A preliminary evaluation of immune stimulation following exposure to metal particles and ions using the mouse popliteal lymph node assay

Toxicology and Applied Pharmacology 308 (2016) 77–90

Contents lists available at ScienceDirect

Toxicology and Applied Pharmacology journal homepage: www.elsevier.com/locate/ytaap

A preliminary evaluation of immune stimulation following exposure to metal particles and ions using the mouse popliteal lymph node assay Brooke E. Tvermoes a,⁎, Kenneth M. Unice b, Bethany Winans c, Michael Kovochich d, Whitney V. Christian b,1, Ellen Donovan c, Ernest S. Fung d, Brent L. Finley c, Ian Kimber e, Dennis J. Paustenbach c a

Cardno ChemRisk, LLC.; 4940 Pearl East Circle Suite 100; Boulder, CO 80301, United States Cardno ChemRisk, LLC.; 20 Stanwix St. Suite 505; Pittsburgh, PA 15222, United States c Cardno ChemRisk, LLC.; 101 2nd St. Suite 700; San Francisco, CA 94105, United States d Cardno ChemRisk, LLC.; 130 Vantis Suite 170; Aliso Viejo, CA 92656, United States e University of Manchester, Faculty of Life Sciences, Oxford Road, Manchester M13 9PT, UK b

a r t i c l e

i n f o

Article history: Received 1 March 2016 Revised 17 June 2016 Accepted 26 July 2016 Available online 28 July 2016 Keywords: Metals Particles Hypersensitivity Lymph node cell activation Immune stimulation Metal-on-metal implants Popliteal lymph node assay

a b s t r a c t The objective of this preliminary study was to evaluate the threshold for immune stimulation in mice following local exposure to metal particles and ions representative of normal-functioning cobalt-chromium (CoCr) metalon-metal (MoM) hip implants. The popliteal lymph node assay (PLNA) was used in this study to assess immune responses in BALB/c mice following treatment with chromium-oxide (Cr2O3) particles, metal salts (CoCl2, CrCl3 and NiCl2), or Cr2O3 particles together with metal salts using single-dose exposures representing approximately 10 days (0.000114 mg), 19 years (0.0800 mg), and 40 years (0.171 mg) of normal implant wear. The immune response elicited following treatment with Cr2O3 particles together with metal salts was also assessed at four additional doses equivalent to approximately 1.5 months (0.0005 mg), 0.6 years (0.0025 mg), 2.3 years (0.01 mg), and 9.3 years (0.04 mg) of normal implant wear. Mice were injected subcutaneously (50 μL) into the right hind foot with the test article, or with the relevant vehicle control. The proliferative response of the draining lymph node cells (LNC) was measured four days after treatment, and stimulation indices (SI) were derived relative to vehicle controls. The PLNA was negative (SI b 3) for all Cr2O3 particle doses, and was also negative at the lowest dose of the metal salt mixture, and the lowest four doses of the Cr2O3 particles with metal salt mixture. The PLNA was positive (SI N 3) at the highest two doses of the metal salt mixture and the highest three doses of the Cr2O3 particles with the metal salt mixture. The provisional NOAEL and LOAEL values identified in this study for immune activation corresponds to Co and Cr concentrations in the synovial fluid approximately 500 and 2000 times higher than that reported for normal-functioning MoM hip implants, respectively. Overall, these results indicate that normal wear conditions are unlikely to result in immune stimulation in individuals not previously sensitized to metals. © 2016 Elsevier Inc. All rights reserved.

1. Introduction In the late 1990s, second generation metal-on-metal (MoM) hip implants became available in the United States as an alternative to metalon-polyethylene (MoP) hip implants (Hart et al., 2015). This second generation of MoM hip implants offered increased stability, lower volumetric wear, and decreased wear-related osteolysis compared with MoP implants available at that time (Fehring et al., 2015). These devices also provided treatment alternatives for end-stage osteoarthritis in elderly individuals, and were recommended to younger patients hoping to resume an active lifestyle because of their increased resistance to dislocation, as well as, their greater range of hip motion (Bozic et al., 2009; ⁎ Corresponding author. E-mail address: [email protected] (B.E. Tvermoes). 1 WC is currently at Medtronic in Jacksonville, FL.

http://dx.doi.org/10.1016/j.taap.2016.07.020 0041-008X/© 2016 Elsevier Inc. All rights reserved.

Bucholz, 2014; Drummond et al., 2015; Girard et al., 2010; Hart et al., 2015; Matharu et al., 2015). In 2006, MoM was the most common bearing-type implanted for patients b65 years of age in the United States, with 42% of all primary total hip replacements being performed with MoM devices (Bozic et al., 2009; Bucholz, 2014; Hart et al., 2015). It has been estimated that in total, approximately 1 million MoM hip implants have been implanted worldwide (Hart et al., 2015; SCENIHR, 2014). The articulating parts of modern MoM hip implants are comprised primarily of cobalt-chromium (CoCr) alloys, in which Co, Cr, and Ni constitute approximately 64%, 28%, and ≤ 1% of the alloy composition, respectively (ASTM, 2011). It is well understood that all implanted metal alloys, including CoCr alloys, corrode and wear over time (to various degrees), resulting in the release of “wear debris,” which is a mixture of metal particles and solubilized metal ions (i.e., Co2+, Cr3+, and Ni2+) (Bauer et al., 2014; Hallab et al., 2001). For normal-functioning

78

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

CoCr MoM hip implants, the volumetric wear rate is b 1 mm3 per year and the majority of the debris is nano-sized (b100 nm) chromium oxide (Cr2O3) particles that are round, oval-shaped, or needle-shaped (Madl et al., 2015a, 2015b). Blood Co and Cr levels in patients with normal-functioning implants are typically about 2–10 times higher than the “background” concentrations present in the general population. Several recent studies have shown that there are negligible systemic health risks associated with the slightly increased blood Co and Cr concentrations reported in patients with normal-functioning implants (Finley et al., 2012a, 2012b; Finley et al., 2013; Monnot et al., 2014; Tvermoes et al., 2014, 2015). However, it is less clear whether low levels of wear debris might result in adverse effects in the periprosthetic tissue of patients with MoM hip implants. Histological evaluation of periprosthetic tissue retrieved at revision surgery of some failed MoM hip implants has revealed a cell mediated immune response characterized by (T and B) lymphocyte infiltration and lymphoid aggregates (primarily CD3+/CD4+ T cells), as well as, the accumulation of plasma cells associated with macrophages containing metallic wear-debris particles in some cases (Campbell et al., 2010; Davies et al., 2005; Willert et al., 2000). Accordingly, it has been hypothesized that metal sensitization may play a role in influencing prosthesis performance, and may contribute to the etiology of certain types of MoM hip implant failures (Campbell et al., 2010; Davies et al., 2005; Hallab et al., 2001; Pinson et al., 2014; Cook et al., 1991). However, the presence of leukocytes, and many of the histological features described in association with adverse local tissue reactions, have been found to be associated with other types of implant failures (Cook et al., 1991; Mirra et al., 1982). Further, questions as to whether allergic sensitization to metal plays a direct role in the failure of the prosthesis, or if loosening of the prosthesis can induce metal allergy as a result of excessive debris generation, have not been fully resolved (Jacobs et al., 2009; Pinson et al., 2014; Watters et al., 2010). “Sensitization” in this case refers to the induction of a T-cell-mediated immune response to metals such a Co, Cr and Ni (Campbell et al., 2010; Schmidt and Goebeler, 2015; Thomas and Cunningham, 2009; Wang and Dai, 2013). Metal sensitization develops in two phases. The first phase, known as the induction phase, is characterized by immunological priming that results in the acquisition of sensitization. The second phase, or the elicitation phase, occurs when an already sensitized individual is exposed to the same allergen and responds immunologically. The elicitation phase is marked by a more rapid and robust immune response resulting in an allergic reaction. The induction and elicitation phases have different dose-response relationships, and therefore, different thresholds, with the induction of sensitization commonly requiring exposure to higher levels of a particular allergen than elicitation. Historically, several animal models have been used to evaluate the sensitization potential of medical devices, including the guinea pig maximization test (GPMT), the occluded patch test of Buehler in guinea pigs, and the mouse local lymph node assay (LLNA) (ISO, 2010). However, these assays were developed primarily to evaluate skin sensitization potential, and they may not reflect accurately immune reactions within the body that may result from exposures such as those associated with MoM hip implants. The popliteal lymph node assay (PLNA) is an appropriate alternative to the LLNA and guinea pig assays for the assessment of systemic sensitization (rather than skin sensitization) potentially associated with implanted medical devices (e.g., deep tissue exposures) (ISO, 2010; WHO, 1999). The PLNA was developed originally for assessing the sensitizing potential of systemically administered drugs (USFDA, 2002), and for this purpose, the assay has been found to be generally reliable, although it has not been used extensively to evaluate the sensitizing potential of metals and/or metal particles (WHO, 2006). Currently, no information exists regarding the dose of MoM wear debris necessary to induce an internal (i.e., deep tissue) sensitization response. This has not previously been investigated in any detail since,

historically, there has been only a low incidence of potential hypersensitivity reactions reported in MoM hip implant patients (Nasser, 2007). While literature is available on threshold doses of Co, Cr, and Ni ions that induce skin sensitization, or elicit allergic contact dermatitis, these data are not necessarily applicable to the threshold dose of metal required to induce or elicit sensitization responses in deep tissue. The purpose of this study, then, was to investigate the utility of the mouse PLNA as a model for evaluating the induction threshold for immune stimulation associated with exposure to metal particles and ions representative of wear debris generated by normal-functioning MoM hip implants. To this end, the stimulation index (SI) of lymph node cell (LNC) activation was measured and flow cytometric analysis of PLN lymphocyte subsets was performed to evaluate the sensitizing potential of metal particles and/or metal ions injected subcutaneously. 2. Materials and methods 2.1. Animals Female, nulliparous, experimentally naïve BALB/c mice (Charles River Laboratories), aged 6–8 weeks were group housed in metal-free, disposable plastic cages conforming to the Guide for the Care and Use of Laboratory Animals recommendations (Gerber et al., 2011). The animal room was temperature controlled (range: 68.6 ± 71.0 °F), and equipped with a 12-h light/dark cycle. Mice were acclimatized at least five days prior to initiation of the experiment. Rodent chow and distilled water were available ad libitum. All procedures involving laboratory animals and test articles complied with acceptable standards of animal welfare and humane care by the Institutional Animal Care and Use Committee (IACUC) of MB Research (Spinnerstown, PA) or Calvert Laboratories (Olyphant, PA). 2.2. Chemicals and reagents Nickel chloride (NiCl2·6H2O, CAS # 7791-20-0), chromium chloride (CrCl3·6H2O, CAS # 10060-12-5), cobalt chloride (CoCl2·6H2O, CAS # 7791-13-1), chromium oxide particles (Cr2O3, CAS #1308-38-9), 2,4dinitrochlorobenzene (DNCB, CAS #97-00-7), 2,4-dichloronitrobenzene (DCNB, CAS #611-06-3), sodium dodecyl sulfate (SDS, CAS # 151-21-3), bromodeoxyuridine (BrdU) and dimethylsulfoxide (DMSO) were purchased from Sigma Aldrich. TiO2 particles (TiO2, CAS # 1317-70-0) were purchased from US Research Nanomaterials, Inc. Potassium dichromate (K2Cr2O7, CAS # 7778-50-9) was purchased from Fisher Scientific and gold chloride (AuCl3, CAS # 13453-07-1) was purchased from Acros Organics (Fair Lawn, New Jersey). Phosphate buffer saline (PBS) was purchased from Hyclone and syngeneic vehicle BALB/c mouse serum was obtained from Charles River Laboratory (Raleigh, NC). All antibodies used for flow cytometric analyses were obtained from BD Pharmingen (San Jose, CA) or Acris Antibodies (San Diego, CA). 2.3. Characterization of metallic particles The size of the Cr2O3 particles and TiO2 particles were determined by RJ Lee Group (Monroeville, PA) using a Hitachi S5500 Ultra-high Resolution Scanning Electron Microscope at an accelerating voltage of 2.0 kV. Secondary electron contrast was used to maximize the surface definition of the particles to aid in sizing measurements. Diameters were measured using the longest chord joining points on the observable perimeter of the particle. Since these particles were sourced commercially, composition of the particulate was periodically checked using a Bruker energy dispersive spectroscopy (EDS) detector at an accelerating voltage of 20 kV. For sterilization, all particles were autoclaved at 127 °C for 30 min. The test particles were shown to be free of endotoxin using the Gel-Clot assay, which has a sensitivity level of 0.03 EU/mL.

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

2.4. PLNA Immediately prior to dosing, mice were anesthetized with isoflurane and then injected subcutaneously into the right hind footpad with 50 μL of the vehicle or the test article in vehicle. Vehicles were 20% DMSO in PBS for K2Cr2O7, AuCl3, SDS, DNCB, DCNB, and serum:PBS for metal particles, CoCl2, CrCl3, and NiCl2 (Table 1). Control animals received vehicle alone. Each treatment group was comprised of ten animals. Eight to ten animals from each treatment group were used to determine the stimulation index (SI), weight index (WI), cellularity index (CI), and flow cytometric analysis, and two animals from each treatment group were used for histopathological assessment. Since immune responses in the PLNA after footpad injection have been reported to occur between four and ten days post-injection, mice were sacrificed on day four, seven or eleven post-injection (Carey et al., 2006; Fernandez Cabezudo et al., 2007; Stiller-Winkler et al., 1988; Suda et al., 2000). Lymph node weight and footpad thickness were measured for all ten animals in each treatment group. A group of untreated mice were also included in the analysis to facilitate the examination of possible vehicle-mediated responses. 2.5. Chemicals, particles and vehicles DNCB, K2Cr2O7, AuCl3, DCNB, and SDS were prepared in 20% DMSO in PBS at the indicated concentrations (Tables 1 and 2). The DMSO concentration was selected to maximize the solubility of the DNCB positive control, and minimize injury to the animal paw observed at higher concentrations. The positive and negative control doses were selected

Table 1 Summary of treatment concentrations and total doses delivered to the mice. The injection volume was 50 μL for all treatments. Group Vehicle controls Sham 20% DMSO-A 20% DMSO-B Serum:PBS-A Serum:PBS-B Chemical positive controls DNCB

Chemical negative control SDS DCNB

Metal positive controls AuCl3

K2Cr2O7

Particle negative control TiO2 particles

Concentration (mg/ml)

Vehicle control

Dose (mg)

NA NA NA NA NA

NA NA NA NA NA

0 0 0 0 0

2.5 6

20% DMSO-B 20% DMSO-A

0.125 0.3

1.88 2.5 6

20% DMSO-A 20% DMSO-A 20% DMSO-B

0.0938 0.125 0.3

0.313 1.25 2.5 0.125 0.5 1.0

20% DMSO-B 20% DMSO-B 20% DMSO-B 20% DMSO-A 20% DMSO-A 20% DMSO-A

0.0156 0.0625 0.125 0.00625 0.025 0.050

0.419

Serum:PBS-B

0.0210

Serum:PBS-B Serum:PBS-B Serum:PBS-A Serum:PBS-B Serum:PBS-A Serum:PBS-A Serum:PBS-A Serum:PBS-A Serum:PBS-A

0.0000144 0.0101 0.0216 0.0000998 0.0699 0.150 0.000114 0.0800 0.171

79

Table 2 Summary of treatment concentrations and total doses delivered to the mice in the secondary follow up study. The injection volume was 50 μL for all treatments. Group

Concentration (mg/ml)

Vehicle control

Dose (mg)

Vehicle controls 20% DMSO Serum:PBS

NA NA

NA NA

0 0

Metal positive controls AuCl3 K2Cr2O7

2.5 0.5

20% DMSO 20% DMSO

0.125 0.025

0.433 0.01 0.05 0.2 0.8 1.60

Serum:PBS Serum:PBS Serum:PBS Serum:PBS Serum:PBS Serum:PBS

0.0216 0.0005 0.0025 0.01 0.04 0.0800

Cr2O3 particles and/or metal saltsa Cr2O3 particles

Cr2O3 particles + metal saltsa

a Metal salts are formulated in the composition ratio presented in Fig. 1. Dose is expressed as the mass of the CoCl2, CrCl3 and NiCl2 salts.

based on literature review and/or animal tolerance studies conducted prior to the present study (Ikarashi et al., 1992a, 1992b; Kammuller et al., 1989; Pieters and Albers, 1999a, 1999b; Schuhmann et al., 1990). The primary objective was to utilize positive controls that resulted in a response defined by a mean SI N 3 and negative controls that resulted in an absence of a response defined by a mean SI b 3. TiO2 particles, Cr2O3 particles, CoCl2, CrCl3 and NiCl2 suspensions were prepared in a 1:1 mixture of BALB/c mouse serum and PBS at the indicated concentrations (Tables 1 and 2). Surrogate MoM hip implant wear debris was formulated from appropriately sized Cr2O3 commercial particles together with metal salts providing a source of Co2+, Cr3+, and Ni2+ ions. The chemical composition (Cr2O3) and size range (median = 129 nm) of the commercial Cr2O3 particles used in this study are similar to MoM wear particles recovered from hip implant patients and those generated by physiological hip simulators (Madl et al., 2015a; Catelas et al., 2004, 2006). The ratio of total individual metals (particulate + ion) was based on the ASTM F1537 standard specifications for wrought CoCr alloys used for surgical implants where Co, Cr, and Ni content comprise approximately 64%, 28%, and ≤ 1% of the implant alloy, respectively (Fig. 1). With respect to particle and metal ion speciation, the majority of Co and Ni generated from wear debris was assumed to occur as ions, and these metals were injected as soluble salts (Davda et al., 2011). Cr was speciated between particulate and ionic form, with approximately 62% of the Cr injected as Cr2O3 particles and 38% as a Cr3+ containing soluble salt. The remaining 7% of the alloy is typically composed of Mo, which has been reported to be a weak, nonspecific contact irritant (Abdouh et al., 1995) and,

a

Cr2O3 particles and/or metal salts Cr2O3 particles 0.0003 0.202 0.433 Metal saltsa 0.002 1.40 3.0 Cr2O3 particles + metal saltsa 0.0023 1.60 3.43 a

Metal salts are formulated in the composition ratio presented in Fig. 1. Dose is expressed as the mass of the CoCl2, CrCl3 and NiCl2 salts.

Fig. 1. Composition of wear debris. *The remaining 7% of the alloy is typically composed of Mo which has been reported to be a weak, nonspecific contact irritant and therefore was not injected into the mice.

80

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

therefore, was not injected into the mice, and is referred to as “inert metal” in this study. To avoid agglomeration/aggregation of the test substances, stock solutions were sonicated prior to dilution, and dilutions were made on each day of administration. All dilutions were stirred or vortexed until homogeneous, and formulations were vortexed prior to dosing. For metal particle and metal salt treatment, each mouse received a total injected dose of either 0.114, 0.50, 2.5, 10, 40, 80, or 171 μg of Cr2O3 particles together with the metal salt mixture (NiCl2, CoCl2, CrCl3) (Tables 1 and 2; Supplemental Tables 1 and 2). The CoCl2 portion of the total injected dose was 0.080, 0.35, 1.75, 7.01, 28.1, 56.1, and 120 μg, respectively. The CrCl3 portion of the total injected dose was 0.018, 0.08, 0.40, 1.61, 6.45, 12.9, and 27.6 μg, respectively. The NiCl2 portion of the total injected dose was 0.001, 0.005, 0.027, 0.11, 0.44, 0.88, and 1.89 μg, respectively. The Cr2O3 particle portion of the total injected dose was 0.014, 0.06, 0.32, 1.26, 5.05, 10.1, and 21.7 μg, respectively (Supplemental Tables 1 and 2). When compared to possible wear scenarios experienced by MoM hip implant patients, the two highest doses chosen for this study are not physiologically relevant (Fig. 9), but were chosen to help understand if the PLNA could be used to investigate immune stimulation following exposure to metal ions and/or particles. The lowest dose used in this study was chosen to be more physiologically relevant. It is also worth noting that 1% of the final injected dose was comprised of Ni2+, which likely represents a worst case scenario, since the ASTM standard notes that the maximum amount of Ni in a wrought CoCr alloy used for surgical implants can comprise up to 1% of the alloy by weight. Thus, the actual amount of Ni present in wrought CoCr alloys used for surgical implants is likely less than the 1% used in this current study. TiO2 particles were used as a negative control particle as previous studies have shown that TiO2 nanoparticles are not skin sensitizers following single subcutaneous injections of TiO2 particles alone (Hussain et al., 2012; Warheit et al., 2007). DNCB, K2Cr2O7 and AuCl3 were used as positive controls for sensitization, while SDS and DCNB were used as negative controls for sensitization. 2.6. Assessment of footpad swelling and general toxicity Approximately four hours post-injection, the injection site on all animals was evaluated for signs of swelling, distress or other signs of general toxicity (See Tables 1 and 2 in Winans et al. (2016)). Subsequently, all animals were observed once daily for signs of local inflammation, distress, and general toxicity. On days one, two and four post injection, a digital micrometer was used to measure right hind footpad swelling on all animals. Body weights were recorded immediately pre-test (day 0) and at termination on day four, seven or eleven. Water and food consumption were monitored throughout the study by daily visual inspection. 2.7. Determination of the stimulation index 2.7.1. BrdU incorporation. Mice from each group were administered the synthetic nucleoside and thymidine analog bromodeoxyuridine (BrdU) dissolved in PBS at a concentration of 15 mg/mL (3 mg per mouse; intraperitoneal) on day four. Five hours later, the mice were euthanized via carbon dioxide inhalation and the contralateral and ipsilateral PLNs for each individual mouse were excised, placed in PBS, freed from adherent fatty tissue, and weighed. PLNs were then transferred to a glass tube in which cells were extracted by gentle disaggregation with a disposable pestle. The cells were centrifuged, washed in PBS, and re-suspended in RPMI. The isolated LNCs were used for either determination of BrdU incorporation or for flow cytometric analyses. A portion of the isolated LNC were fixed in 75% EtOH and stored up to one week at −20 °C for determining BrdU incorporation. Another portion of the isolated LNCs were stored overnight at 2 to 8 °C for flow cytometric analysis the following

day. If the total number of cells isolated from a single lymph node was b50,000 cells, it was concluded that the isolation of that lymph node was not successful, and that animal was not included in SI, CI, WI or flow cytometric analysis. Animals from the following treatment groups were excluded from analysis: SDS (n = 1); K2Cr2O7 (0.050 mg, n = 2); Cr2O3 particles with metal salts (0.000114 mg, n = 1); Cr2O3 particles (0.0000144 mg, n = 2). Lymph nodes that were used for histological evaluation (n = 2 per treatment group) were fixed in 10% neutral-buffered formalin and embedded in paraffin. To determine BrdU incorporation, 200–250 μL aliquots of isolated LNCs were denatured with HCl Triton X Buffer (1 N HCl, 0.5% Triton X), and samples were neutralized by washing with borate buffer (pH 8.5). Nuclei were washed with a staining buffer and incubated with BrdU-specific antibody FITC conjugate (BD Biosciences, clone B44). The nuclei were washed again with staining buffer and resuspended in PBS containing RNase A (Fisher Scientific) and propidium iodide (PI, Sigma Aldrich). Following a 30-min incubation at room temperature, the total DNA content of the nuclei, as well as, the percentage of nuclei staining positive for BrdU (i.e., percentage of proliferating lymphocytes), were determined with a BD FacScan® flow cytometer. 2.7.2. Tritiated thymidine incorporation. A smaller follow up study was conducted to fill in the dose-response curve as well as to better define the kinetics of the primary PLN reaction observed in the initial dose-response investigation. The follow up study was conducted in a second independent laboratory using 3H-thymidine incorporation to calculate the SI. The dose-response study followed a similar PLNA protocol to that described above except animals were injected with a total metal dose of either 0.5, 2.5, 10, 40 or 80 μg of Cr2O3 particles together with the metal salt mixture (NiCl2, CoCl2, CrCl3) (Table 2 and Supplemental Table 2). One group of mice were also injected with 21.6 μg of Cr2O3 particles alone. In addition, the study also included the following vehicle control groups: serum:PBS and 20% DMSO, as well as, the following positive control groups: AuCl3 (125 μg) and K2Cr2O7 (25 μg). Ten mice from each treatment group were sacrificed four days, seven days and eleven days post injection. On the day of sacrifice each mouse received an intravenous injection, via the lateral tail vein, of 250 μL of PBS containing 20 μCi of 3H-thymidine (Amersham, Buckinghamshire, UK). Five hours later the mice were euthanized via carbon dioxide inhalation and the ipsilateral PLN for each individual mouse was excised and placed in 10 mL of PBS. Single cell suspensions of LNCs were prepared, washed twice with PBS, precipitated with 5% trichloroacetic acid (TCA) and refrigerated at 4 °C for at least 18 h. The samples were then centrifuged, resuspended in 5% TCA and mixed with scintillation cocktail. Incorporation of 3H-thymidine was measured by β-scintillation counting and expressed as disintegrations per minute (dpm) per mouse. A background reading was taken at the beginning, middle and end of the assessment. The SI was calculated using the background subtracted dpm value for each mouse as the numerator and the mean background subtracted dpm value from the vehicle control mice as the denominator. A test material that, at one or more concentrations, caused a threefold or greater increase in proliferation was considered to be positive for immune stimulation. 2.8. Weight (WI), cellularity (CI), and stimulation (SI) indices The PLN WI was calculated by dividing the PLN weight of the test animal by the mean PLN weight of the appropriate vehicle control group listed in Table 1. PLN cell number (cellularity) was measured using a hemacytometer. The CI was calculated by dividing the total cell count of the test animal by the mean cell count of the relevant vehicle control group. Response indices for weight and cellularity were also calculated for the vehicle control groups (DMSO and serum:PBS) by dividing the LN weight or cell count of the vehicle test animal by the mean LN weight or cell count of the untreated group (sham). The vehicle controls showed no change in either weight or cellularity when compared to

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

sham controls (Supplemental Fig. 1). The treatment groups were divided over two separate series of experiments. Each series of experiments had a DMSO and serum:PBS control leading to a DMSO-A and -B and a serum:PBS-A and -B where A denotes the control groups for the first set of experiments and B denotes the control groups for the second set of experiments (Supplemental Fig. 1). The SI was calculated by taking the ratio of the test animal's BrdU labeling index to the mean BrdU labeling index of the relevant vehicle control group. The BrdU labeling index is the total number of BrdU+ lymphocytes per ipsilateral popliteal lymph node of each animal and was calculated by multiplying the total cell number by the percent of BrdU+ cells in the node. The ipsilateral node was used for all indices. 2.9. Flow cytometry Cells were blocked in either rat IgG (for B220, CD3, CD4, and CD8) or hamster/mouse IgG (for I-AD and CD69) for 10 min. Aliquots from each LNC suspension (~5 × 105 cells) were stained with fluorescently-conjugated B cell (B220-FITC, BD Pharmingen, clone RA3-6B2), T cell (CD3PE, BD Pharmingen, clone 17A2; CD4-PE, BD Pharmingen, clone RM45; CD8-FITC, BD Pharmingen, clone 53-6.7) and activation (I-AD-FITC, Acris Antibodies, clone 34-5-3S; CD69-PE, BD Pharmingen, clone H1·2F3) cell surface marker antibodies (30–45 min incubation on ice), fixed with 70% ethanol and analyzed by flow cytometry. LNC analyses were performed with a BD FacScan® flow cytometer using 15 mW of power at 488 nm excitation wavelength. BD CellQuest version 3.3 acquisition software on a Macintosh G4 acquisition system was used to capture and store data on a dedicated secure network drive. Data files were analyzed using CellQuest™ and FlowJo to determine appropriate analysis gate and percent positive LNC populations. 2.10. Corresponding wear rate scenario The injected animal dose can be expressed as a human equivalent wear debris volume, V, using the following equation (Supplemental Table 4):

V¼

   Ma BW h ρ BW a

ð1Þ

where Ma is the wear debris dose expressed as the mass of Co, Cr, Ni and inert metal (mg), BWh is human body weight (70 kg), BWa is the average mouse body weight (0.018 kg), and ρ is the alloy density (8.3 g/cm3). Direct body weight scaling was selected as the most appropriate metric for assessing induction associated with an acute response in the local draining lymph node (USEPA, 2011). The wear debris dose can also be expressed as the number of years of wear, Y, generated by a normal-functioning MoM hip implant using the following equation (Supplemental Table 4): Y ¼ V=A

ð2Þ

where A is annual wear rate of a normal-functioning implant of 1 mm3/ year (Madl et al., 2015a; Madl et al., 2015b; Morlock et al., 2008; Sidaginamale et al., 2013; Sieber et al., 1999). 2.11. Statistical analysis The results are expressed as a mean ± standard error (SE). Twosample equal variance t-tests were used to assess the significance of differences between mean values at p ≤ 0.05.

81

3. Results 3.1. Characterization of metal particles The Cr2O3 and TiO2 particles were characterized by SEM to assess particle morphology and size, and by EDS to confirm the chemical composition of the particles. The diameters of the Cr2O3 particles as measured by SEM (Supplemental Table 3) ranged in size from approximately 30 to 996 nm. While the upper bound of the size range of the particles is larger than what has typically been characterized for particles released from normal-functioning MoM hip implants, the median diameter was 129 nm, which is similar to data indicating that the majority of wear debris particles recovered from human tissue and/or generated using a hip simulator have a diameter b100 nm (Billi et al., 2012; Catelas et al., 2004, 2006; Madl et al., 2015a). EDS analysis showed that the particles contained primarily chromium oxide, which is similar to the chemical composition of the majority of wear debris generated under normal physiological conditions in vitro or in vivo (Fig. 2A, B) (Catelas et al., 2004, 2006; Madl et al., 2015a). TiO2 particles displayed a similar size range as the Cr2O3 particles (Supplemental Table 3), and EDS analysis showed that the particles contained primarily titanium oxide (See Fig. 1 in Winans et al. (2016)). 3.2. Stimulation, weight, and cellularity indices The mean SI, WI and CI were calculated to assess the potential sensitizing capacity of each test article (Figs. 3 and 4). 3.2.1. Stimulation index. In the LLNA, the threshold for a biologically significant response is an SI ≥ 3; as such, a similar threshold was used in this current study to indicate a positive response in the PLNA. The average SIs for mice dosed with the negative controls DCNB, TiO2 or the irritant SDS were less than three and were not significantly different from the vehicle control. The average SIs observed in mice dosed with known sensitizing agents DNCB (0.125 mg or 0.3 mg), AuCl3 (0.0625 mg and 0.125 mg), and K2Cr2O7 (0.050 mg) were significantly different from the vehicle control and had average SIs greater than three (Fig. 3a). Further, treatment with the two highest doses of the metal salt mixture (0.0699 mg and 0.150 mg) and the two highest doses of Cr2O3 particles with metal salts (0.0800 mg and 0.171 mg) resulted in average SIs that were above three and significantly different from vehicle controls (Fig. 4). Treatment with the lowest dose of the metal salt mixture alone (0.0998 × 10−3 mg) or the lowest dose of Cr2O3 particles with the metal salt mixture (0.1143 × 10− 3 mg) resulted in average SIs that were not significantly different from the vehicle control and were less than three. In addition, treatment at all doses of the Cr2O3 particles alone resulted in average SIs b 3 that were not significantly different from the vehicle control (Fig. 4a). 3.2.2. Weight index. In the PLNA, the threshold for a positive response has been defined as WI ≥ 2. The average WIs for mice dosed with the negative controls DCNB, TiO2, and the irritant SDS were less than the threshold value of two, and were not significantly different from the vehicle control (Fig. 3b). In contrast, treatment with the two highest doses of AuCl3 (0.0625 mg and 0.125 mg) resulted in average WIs N 2 that were significantly greater than the vehicle control. Treatment with the lowest dose of AuCl3 (0.0156 mg) and the highest dose of DNCB (0.3 mg) resulted in average WIs that were significantly greater than the vehicle control, but less than the threshold value of two (Fig. 3b). All three doses of K2Cr2O7 resulted in average WIs less than the threshold value of two that were not significantly different from the vehicle control. Similarly, all three doses of Cr2O3 particles alone resulted in average WIs less than two that were not significantly different from the vehicle control. Treatment with the lowest dose of the metal salt mixture alone (0.0998 × 10−3 mg) or the lowest dose of Cr2O3 particles with the metal

82

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

Fig. 2. SEM image (A) and chemical composition (B) of Cr2O3 particles.

salt mixture (0.1143 × 10−3 mg) also resulted in average WIs less than two that were not significantly different from the vehicle control. Treatment with the two highest doses of the metal salt mixture (0.0699 mg and 0.150 mg) and the two highest doses of Cr2O3 particles with the metal salt mixture (0.0800 mg and 0.171 mg) resulted in average WIs that were significantly different from vehicle controls, but were less than the threshold value of two (Fig. 4b). 3.2.3. Cellularity index. The threshold for a positive response in the PLNA has been defined as a CI ≥ 5. The average CIs for mice dosed with the negative controls DCNB, TiO2, and the irritant SDS were less than the threshold value of five, and were not significantly different from the vehicle control (Fig. 3c). Treatment with the highest dose of DNCB (0.3 mg) or the highest dose of AuCl3 (0.125 mg) resulted in average CIs greater than five that were significantly different from the vehicle control. Treatment with the lowest two doses of AuCl3 (0.0156 mg and 0.0625 mg), the lowest dose of K2Cr2O7 (0.006 mg) and the lowest dose of DNCB (0.125 mg) resulted in average CIs that were significantly greater than the vehicle control, but less than the threshold value of five. The two highest doses of K2Cr2O7 (0.025 mg and 0.050 mg) resulted in average CIs less than the threshold value of five that were not significantly different from the vehicle control. All three doses of Cr2O3 particles alone resulted in an average CI less than the threshold value of five that was not significantly different than the vehicle control (Fig. 4c). Treatment with the lowest dose of the

metal salt mixture alone (0.0998 × 10− 3 mg) or the lowest dose of Cr2O3 particles with the metal salt mixture (0.1143 × 10−3 mg) resulted in average CIs less than five that were not significantly different from the vehicle control. Treatment with the two highest doses of the metal salt mixture (0.0699 mg and 0.150 mg) and the two highest doses of Cr2O3 particles with the metal salt mixture (0.0800 mg and 0.171 mg) resulted in average CIs that were significantly greater than the vehicle control, but were less than the threshold value of five. 3.2.4. Weight of evidence evaluation. Using a weight of evidence approach, the three known sensitizers, DNCB (0.125 mg and 0.3 mg), AuCl3 (0.0625 mg and 0.125 mg) and K2Cr2O7 (0.050 mg) showed signs of immune stimulation by the PLNA four days after treatment (Table 3), providing reassurance that, as performed here, the PLNA provides a legitimate method for assessing sensitizing potential. Immune stimulation was not observed four days after treatment with any dose of the Cr2O3 particles alone, or following treatment with the lowest dose of the metal salt mixture (CoCl2, CrCl3, and NiCl2), or the lowest dose of Cr2O3 particles with the metal salt mixture. Exposure to the two highest doses of the metal salt mixture (0.0699 mg and 0.150 mg) and Cr2 O3 particles with the metal salt mixture (0.080 mg and 0.171 mg) resulted in clear immune stimulation four days after treatment. In these groups, and at these doses, the average SIs were greater than three, and the average WIs and CIs were significantly increased relative to the vehicle control and were

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

83

a) Stimulation Index - Controls 12

*

11 10 9

*

8

SI

7

*

6 5 *

4

* *

3

*

2 1 0 DNCB 0.125 mg

DNCB 0.3 mg

Chemical Positive Controls

SDS DCNB 0.0938 mg 0.125 mg

DCNB 0.3 mg

AuCl3 AuCl3 AuCl3 0.0156 mg 0.0625 mg 0.125 mg

Chemical Negative Controls

K2Cr2O7 K2Cr2O7 K2Cr2O7 TiO2 0.006 mg 0.025 mg 0.050 mg 0.0210 mg

Metal Positive Controls

Particle Neg. Cont.

b) Weight Index - Controls 3 *

2

*

*

WI

*

1

0 DNCB 0.125 mg

DNCB 0.3 mg

Chemical Positive Controls

SDS DCNB 0.0938 mg 0.125 mg

DCNB 0.3 mg

AuCl3 AuCl3 AuCl3 0.0156 mg 0.0625 mg 0.125 mg

Chemical Negative Controls

K2Cr2O7 K2Cr2O7 K2Cr2O7 TiO2 0.006 mg 0.025 mg 0.050 mg 0.0210 mg

Metal Positive Controls

Particle Neg. Cont.

c) Cellularity Index - Controls 8 *

7 6

*

CI

5 4

*

*

*

3 *

2 1 0 DNCB 0.125 mg

DNCB 0.3 mg

Chemical Positive Controls

SDS DCNB 0.0938 mg 0.125 mg

DCNB 0.3 mg

Chemical Negative Controls

AuCl3 AuCl3 AuCl3 0.0156 mg 0.0625 mg 0.125 mg

K2Cr2O7 K2Cr2O7 K2Cr2O7 TiO2 0.006 mg 0.025 mg 0.050 mg 0.0210 mg

Metal Positive Controls

Particle Neg. Cont.

Fig. 3. PLNA control indices at day 4 including (a) stimulation index (SI), (b) weight index (WI), and (c) cellularity index (CI) following footpad injection. The dashed line indicates an SI value of 3, WI of 2, or a CI of 5 which are the threshold values for a positive response in the LLNA and/or PLNA. Data are presented as the mean LLNA and/or + SE. Asterisk indicates p b 0.05 for two sample mean comparison of test article group and respective vehicle.

84

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

a) Stimulation Index - Test Articles 10 9 8 *

7

SI

6

* *

5 *

4 3 2 1 0 Cr2O3 0.0144E-3 mg

Cr2O3 0.0101 mg

Cr2O3 0.0216 mg

Metal salts 0.0998 E-3 mg

Metal salts 0.0699 mg

Metal salts 0.150 mg

Cr2O3+ metal salts 0.0800 mg

Cr2O3+ metal salts 0.171 mg

Cr2 O3 Particles+Metal Salts

Metal Salts (CoCl2 , NiCl2 , and CrCl3 )

Cr2 O 3 Particles

Cr2O3+ metal salts 0.114E-3 mg

b) Weight Index - Test Articles 3

*

2

*

*

WI

*

1

0 Cr2O3 0.0144E-3 mg

Cr2O3 0.0101 mg

Cr2O3 0.0216 mg

Metal salts 0.0998 E-3 mg

Metal salts 0.0699 mg

Metal salts 0.150 mg

Cr2O3+ metal salts 0.0800 mg

Cr2O3+ metal salts 0.171 mg

Cr2 O3 Particles+Metal Salts

Metal Salts (CoCl2 , NiCl2 , and CrCl3 )

Cr2 O 3 Particles

Cr2O3+ metal salts 0.114E-3 mg

c) Cellularity Index - Test Articles 8 7 6

* *

CI

5

*

*

4 3 2 1 0 Cr2O3 0.0144E-3 mg

Cr2O3 0.0101 mg

Cr2O3 0.0216 mg

Cr2 O3 Particles

Metal salts 0.0998 E-3 mg

Metal salts 0.0699 mg

Metal salts 0.150 mg

Metal Salts (CoCl2 , NiCl2 , and CrCl3 )

Cr2O3+ metal salts 0.114E-3 mg

Cr2O3+ metal salts 0.0800 mg

Cr2O3+ metal salts 0.171 mg

Cr2 O 3 Particles+Metal Salts

Fig. 4. PLNA test article indices at day 4 including (a) stimulation index (SI), (b) weight index (WI), and (c) cellularity index (CI) following footpad injection. The dashed line indicates an SI value of 3, WI of 2, or a CI of 5 which are the threshold values for a positive response in the LLNA and/or PLNA. Data are presented as the mean + SE. Asterisk indicates p b 0.05 for two sample mean comparison of test article group and respective vehicle.

approaching the threshold values of two and five, respectively. In addition, all four treatment groups had B220+ ratios N1.25. The histology also showed some evidence of immune stimulation in one of the

mice treated with the highest dose of Cr2O3 particles with the metal salt mixture (histology data presented in Tables 3–6 of Winans et al. (2016)).

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90 Table 3 Classification using a weight of evidence approach. The fraction of animals meeting the criteria outlined below are presented. The bold font indicates the treatment groups or doses considered to be positive in the PLNA using a weight of evidence approach. Treatment group

Dose (mg) Primary evidence of immune stimulation

Secondary Histologya,b evidence of immune stimulation

SI ≥ B220+/control WI ≥

CI ≥

3

≥ 1.25

2

5

Chemical positive controls DNCB 0.125 0.3

6/8 6/8

6/8 5/8

1/10 2/10

1/8 4/8

1/2 0/2

Chemical negative control SDS 0.0938 DCNB 0.125 0.3

1/7 0/8 0/8

2/7 2/8 0/8

0/9 2/10 0/10

0/7 0/8 0/8

0/2 0/2 0/2

Metal positive controls AuCl3 0.0156 0.0625 0.125 K2Cr2O7 0.00625 0.025 0.050

0/8 4/8 8/8 1/8 3/8 4/6

1/8 6/8 8/8 0/8 2/8 3/6

1/10 5/10 8/10 0/10 0/10 0/8

0/8 1/8 3/8 0/8 0/8 0/6

0/2 0/2 0/2 0/2 NP/NP 2/2

1/8

2/8

0/10

0/8

0/2

0/6 0/8 0/8 0/8 5/8 7/8 0/7 6/8 6/8

1/6 6/8 2/8 4/8 7/7 8/8 2/7 7/8 6/8

1/8 0/10 1/10 0/10 1/10 2/10 0/9 2/10 5/10

0/6 0/8 0/8 0/8 1/8 2/8 0/7 4/8 1/8

1/2 0/2 0/2 0/2 0/2 0/2 0/2 0/2 1/2

Particle negative control TiO2 particles 0.0210 Cr2O3 particles and/or metal salts Cr2O3 particles 0.0000144 0.0101 0.0216 Metal saltsc 0.0000998 0.0699 0.150 Cr2O3 particles + 0.000114 c 0.0800 metal salts 0.171

c

a Two animals from each group were used for histological evaluation. For more information please refer to Winans et al., 2016. Animals considered positive for histology were reported to have at least 3–4 secondary follicles/germinal centrals or a score of +++ or ++++ for lymphocyte hyperplasia in the paracortex. b “NP” indicates that lymph node tissue was not present for analysis on the prepared slide. c Metal salts are formulated in the composition ratio presented in Fig. 1. The dose is expressed as the mass of the CrCl3, CrCl2 and NiCl2 salts.

85

3.3. Flow cytometry immunophenotyping The ratio of B220+ sample/B220+ control in combination with SI data has been used in the LLNA to help distinguish between sensitizing agents (N 1.25) and irritants (b1.25) (Supplemental Table 5) (Gerberick et al., 2002). In this study, mice dosed with DNCB (0.125 mg and 0.3 mg), the two highest doses of AuCl3 (0.0625 mg and 0.125 mg), and the highest dose of K2Cr2O7 (0.050 mg) had B220+ ratios N 1.25 in combination with an average SI N 3 (Supplemental Table 6). B220+ ratios N 1.25 in combination with an SI N 3 were also seen following treatment with the two highest doses of the metal salt mixture (0.0699 mg and 0.150 mg) and the two highest doses of the Cr2O3 particles with the metal salt mixture (0.0800 mg and 0.171 mg). These treatment groups were also associated with a significant increase in the percentage of B220+ cells relative to controls (Figs. 5, 7). Treatment with Cr2O3 particles alone at the low (0.0144 × 10 –3 mg) and high (0.0216 mg) dose yielded a B220+ratio below the 1.25 threshold in combination with an average SI b 3. Mice dosed with SDS (a known irritant), the medium dose of Cr2O3 particles alone (0.010 mg), and the lowest dose of the metal salt mixture (0.0998 × 10 –3 mg) all had a B220+ test:vehicle ratio above the 1.25 threshold but average SIs b 3 (Supplemental Table 6). The expansion of B cells, T cells, and activated antigen presenting cells (APCs) in the ipsilateral lymph nodes were evaluated by assessing the number of B220+ B cells, CD4+ T cells, CD8+ T cells and I-AD+ CD69+ cells (Figs. 5–7). Figs. 5–7 presents the mean values for the lymphocyte subsets determined in the PLN 4 days after substance treatment. Relative to the vehicle controls, there was no significant change in the number of B220+ B cells, CD8+ T cells, or I-AD+ CD69+ cells following treatment with the negative controls SDS, DCNB, or TiO2 (Fig. 6). In contrast, treatment with both doses of DNCB (0.125 mg and 0.3 mg), the lowest dose of K2Cr2O7 (0.006 mg) and the two highest doses of AuCl3 (0.0625 mg and 0.125 mg) resulted in a significant increase in the number of CD4+ T cells, CD8+ T cells, B220+ cells and activated APC relative to vehicle control. There was no significant increase in the number of CD4+ T cells, CD8+ T cells, B220+ cells, or activated APCs relative to vehicle controls following treatment with Cr2O3 particles alone at any dose. In contrast, treatment with the two highest doses of the metal salt mixture (0.0699 mg and 0.150 mg) and the two highest doses of Cr2O3 particles with the metal salt mixture (0.0800 mg and 0.171 mg) resulted in a

Fig. 5. The number of B220+, CD4+, CD8+ and I-AD + CD69+ cells in the PLN following footpad injections for (a) vehicles, (b) positive and negative controls, and (c) test articles. Mice were sham injected, or injected with the controls or test articles as indicated. Four days later, the number of B220+, CD4+, CD8+, and I-AD+ CD69+ cells in the PLN were determined by flow cytometry. Data are presented as the mean + SE.

86

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

significant increase in the number of CD4+ T cells, CD8+ T cells, B220+ cells, and activated APC relative to vehicle controls (Fig. 7).

stimulation was evaluated in two independent laboratories using two slightly different techniques (BrdU incorporation and 3H-thymidine incorporation); yet, the mean stimulation index reported in both labs following exposure to 0.08 mg of metal was very similar (6.39 vs 5.56) providing confidence in the data and methodology. Overall, the findings from this study suggest that the immune response elicited after treatment with the metal salt mixture, or with the Cr2O3 particles with the metal salt mixture, is dose- and time-dependent, and is largely driven by the metal salt mixture, since all doses of the Cr2O3 particles alone failed to elicit an immune response up to eleven days after treatment. However, further work, such as a repeated dosing study, would be helpful to confirm that the Cr2O3 particles alone are not capable of inducing an immune response following prolonged or repeated exposures at physiologically relevant doses. In this current study, a provisional NOAEL for immune stimulation of 0.01 mg of metal was identified, and a provisional LOAEL of 0.04 mg of metal was also identified.

3.4. Dose-response curve and kinetics of the PLN response to metal particles and metal ions On the basis of the dose-response curves shown in Fig. 4, an initial LOAEL and NOAEL of 0.08 mg and 0.114 × 10−3 mg of metal, respectively, was identified. Additional testing was undertaken to further refine the dose-response curve and a provisional LOAEL and NOAEL was identified of 0.04 mg of metal and 0.01 mg of metal, respectively, based on 3 H-thymidine incorporation. The two highest doses tested (0.08 mg of metal and 0.04 mg of metal) resulted in average SIs that were equal to or above three and significantly different from vehicle controls. Maximal immune response for these two doses occurred on day 4 and returned to background levels by day 7. The results of this study also exclude the possibility that the lack of responsiveness (SI b 3) observed at the lower dose groups (0.0005 mg, 0.0025 mg, and 0.01 mg) were caused by a temporal shift in PLN reactivity as even by day 11 none of the SIs were above 3 (Fig. 8). In addition, all five doses of Cr2O3 particles plus metal salts resulted in average CIs less than the threshold value of five at all three time points (Supplemental Fig. 3).

4.1. Previous work using mice to investigate the sensitization potential of metal ions and particles The results of this current study are in line with previous work using the primary PLNA or reporter antigen PLNA to assess the sensitization potential of various metals in mice. In the BALB/c strain mice, immune stimulation, as indicated by an increase in PLN cell number or increased PLN weight, has been reported following treatment with 0.05 or 0.06 mg HgCl2, 0.05 mg PbCl2, and 0.02 or 0.04 mg of various platinum containing compounds (Albers et al., 1996; Carey et al., 2006; Schuppe et al., 1992; Stiller-Winkler et al., 1988). Therefore, the provisional LOAEL (0.04 mg of metal) identified in this study is consistent with previous studies reporting immune stimulation in BALB/c mice following metal exposure. In mice, NOAELs have also been identified using the PLNA after treatment with 0.04 mg CdCl2 and 0.03 mg HgCl2 (Carey et al., 2006; Stiller-Winkler et al., 1988). Thus, the finding that the lowest treatment of 0.114 × 10 -3 mg Cr2O3 particles with the metal salt mixture did not lead to immune stimulation is not surprising, as it is at least an order of magnitude less than previously identified metal NOAELs in the PLNA. The NOAEL of 0.01 mg of metal identified in this study is in line with previous reports.

4. Discussion In this present investigation, a novel approach has been employed using the mouse PLNA to evaluate the induction threshold for immune stimulation associated with exposure to metal particles and ions representative of implant wear generated by normal-functioning MoM devices. One of the advantages of this current study, as compared with previous studies such as that conducted by Brown et al. (2013), was that a mixture of metal particles and ions was used to assess possible immune stimulation. The chemical composition (Cr2O3) and size range (median = 129 nm) of the commercial Cr2O3 particles used in this study are similar to MoM wear particles recovered from hip implants patients and those generated by physiological hip simulators (Madl et al., 2015a; Catelas et al., 2004, 2006). Therefore, the combination of metal ions with metal particles is more representative of the constituents released from MoM implants, and thus provides a physiologically relevant exposure scenario. In addition, immune

* *

* * * *

* * *

*

*

* *

*

*

* * *

Fig. 6. The number of B220+, CD4+, CD8+ and I-AD+ CD69+ cells in the PLN following footpad injections for the positive and negative controls. Mice were injected with the controls as indicated. Four days later, the number of B220+, CD4+, CD8+, and I-AD+ CD69+ cells in the PLN were determined by flow cytometry. Data are presented as the mean ± SE. * indicates statistical significance compared to the vehicle control.

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

4.2. Relevance to human exposure An important objective of this study was to correlate the animal data to exposure scenarios relevant to MoM hip implant patients so as to start to build an understanding of the amount of metal particles and/ or ions necessary to induce deep tissue immunological responses. To this end, injection doses were converted to volumetric wear rates and test article concentrations were directly compared with metal concentrations reported in the synovial fluid of MoM hip implant patients. Dose-response data from the mouse LLNA are commonly used to predict the relative potency of skin sensitizing chemicals in humans. For this purpose an EC3 value is used, as this is the concentration of test chemical required to cause a threefold increase in the SI relative to controls (Basketter et al., 1999). Previous studies have shown that mouse LLNA EC3 values correlate closely with human NOAEL values (Api et al., 2015; Basketter et al., 2005; Griem et al., 2003). This provides a precedent for the possibility that direct comparisons can be drawn between dose-response data in mice and an immune stimulation threshold in man, and suggests that the PLNA provides an appropriate model to address the aims of the investigation reported here. 4.2.1. Comparison with synovial fluid concentrations. The animal doses administered in this study were compared to human doses on both a concentration and body weight basis to address uncertainties in the differences between the animal and human exposure scenarios. Lass et al. (2014) measured metal ion concentrations in hip aspirate from normalfunctioning MoM hip implants and reported a median concentration of 113.4 μg/L for Co (range: 3.9–176 μg/L) and a median concentration of 54 μg/L for Cr (range: 1.5–334 μg/L). Assuming a 70 kg adult and an average hip fluid volume of 2.7 mL for an asymptomatic hip, this equates to an exposure of approximately 0.004 μg Co/kg and 0.002 μg Cr/kg (Moss et al., 1998). In this study, administered Co and Cr concentrations as high as 63,700 μg/L and 27,900 μg/L, respectively, were not associated with immune stimulation. In mice these exposure concentrations equate to approximately 177 μg Co/kg and 77 μg Cr/kg using the average weight of the mice (0.018 kg) and an injection volume of 50 μL. At sufficiently high doses, however, immune stimulation was observed following treatment with approximately 254,700 μg/L of Co and 111,400 μg/L of

87

Cr. In mice this equates to an exposure of approximately 707 μg Co/kg and 310 μg Cr/kg. On μg/kg basis, this is thousands of times higher than the amount of Co and Cr reported in the synovial fluid of normalfunctioning MoM hip implant patients (Fig. 9). In MoM patients with failed devices, synovial fluid concentrations as high as 64,550 μg/L for Co and 263,300 μg/L for Cr have been reported (Beraudi et al., 2013; Sampson and Hart, 2012). Assuming a 70 kg adult and an average hip fluid volume of 6.1 mL for a symptomatic hip, this equates to an exposure of approximately 5.6 μg Co/kg and 23 μg Cr/kg (Moss et al., 1998). This is appreciably less than the provisional NOAEL identified in this study for Co (177 μg Co/kg) and Cr (77 μg Cr/kg) and provisional LOAEL identified for Co (707 μg Co/kg) and Cr (310 μg Cr/kg). 4.2.2. Human equivalent dose and corresponding wear scenario. The doses used in mice in this current study can also be converted to a volumetric wear rate and then compared with the volumetric wear rate associated with a normal-functioning MoM hip implant, which is typically considered to be b1 mm3 per year (~0.003 mm3/day) (Morlock et al., 2008; Sidaginamale et al., 2013; Sieber et al., 1999). Using Eqs. 1 and 2, the human equivalent wear debris volume associated with the seven injected doses was approximately 0.027, 0.12, 0.6, 2.3, 9.3, 19 and 40 mm3, which is approximately equivalent to 10 days, 1.5 months, 0.6 years, 2.3 years, 9.3 years, 19 years and 40 years of cumulative wear from a normal-functioning hip implant (at an average rate of approximately 0.003 mm3/day) injected in a signal bolus dose (Eqs. 1 and 2; Supplemental Table 4). Thus, the provisional NOAEL for immune stimulation identified in this current study was equivalent to ~ 2.3 years (2.3 mm3) of wear generated by a normal-functioning MoM hip implant administered in a single dose and the provisional LOAEL for immune stimulation corresponded to approximately 9.3 years of cumulative wear (9.3 mm3) generated by a normal-functioning MoM hip implant injected in a single dose. Langton et al. (2011) reported one of the highest wear rates for an MoM hip implant of 95.5 mm3/year, which represents an extreme wear scenario. With a density of 8.3 g/cm3 for CoCrMo alloy, a low wear scenario (1 mm3/year) and a high wear scenario (95.5 mm3/ year) results in the release of approximately 8.3 mg of total metal and 793 mg of total metal in one year, respectively. Assuming a 70 kg

Fig. 7. The number of B220+, CD4+, CD8+ and I-AD+ CD69+ cells in the PLN following footpad injections for the test articles. Mice were injected with the test articles as indicated. Four days later, the number of B220+, CD4+, CD8+, and I-AD+ CD69+ cells in the PLN were determined by flow cytometry. Data are presented as the mean ± SE. * indicates statistical significance compared to the vehicle control.

88

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90 9

D4

D7

D11 *

SI

6

*

3 *

*

*

0.0025 mg

0.01 mg

0 Cr2O3 Particles 0.0216 mg

0.0005 mg

0.04 mg

0.08 mg

human, the amount of metal released in one year equates to a daily wear rate of approximately 0.32 μg metal/kg-day (normal wear) and 31 μg metal/kg-day (extreme wear). In comparison, the mice in this study were injected with a single bolus dose of wear, equivalent to approximately 3.2 μg/kg, 13.8 μg/kg, 69.1 μg/kg, 276 μg/kg, 1105 μg/kg, 2211 μg/kg, and 4744 μg metal/kg. The NOAEL for immune stimulation was 276 μg/kg, and is approximately 800 times higher than that associated with normal wear. The provisional LOAEL for immune stimulation was 1105 μg/kg, which is approximately 3500 times higher than the amount of daily wear in a normal-functioning implant, and 36 times higher than the amount of daily wear experienced by an MoM patient under extreme wear conditions.

Cr2O3 + metal salts Fig. 8. Dose-response curve and kinetics of the PLN stimulation index calculated from 3Hthymidine uptake by the PLN following treatment with metal particles or metal particles plus metal ions 4 days, 7 days and 11 days post treatment. The dashed line indicates an SI value of 3 which is the threshold value for a positive response in the PLNA. Data are presented as the mean ± SE. Asterisk indicates p b 0.05 for two sample mean comparison of test article group and respective vehicle.

4.3. Areas for future study In this study, mice received a single bolus dose of commercial wear particles, whereas MoM hip implant patients are exposed to wear debris on a daily basis for years. However, the development of sensitization is a

a) Cobalt

b) Chromium

Fig. 9. Administered cobalt and chromium concentrations in the PLNA assay of current study relative to cobalt and chromium concentrations reported in MoM hip implant patients requiring revision (Beraudi et al., 2013; Davda et al., 2011; De Smet et al., 2008; Kwon et al., 2011; Langton et al., 2010; Mao et al., 2011; Reito et al., 2015; Sampson and Hart, 2012; Tower, 2010).

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

threshold based mechanism, meaning there is a level of exposure (threshold) below which sensitization is not likely to occur because there is not sufficient amounts of antigen/allergen present to stimulate the immune system and trigger a response. The exposure scenarios used in this study, then, were designed to investigate the threshold for immune stimulation following treatment with metal particles and ions representative of wear debris generated by normal-functioning MoM hip implants. It is unclear whether the daily dose or the cumulative dose is the best dose metric for predicting adverse effects. The exposure scenarios and the PLNA-based model system employed in this study were not designed to investigate or characterize disease progression after the acquisition of sensitization. Under the exaggerated exposure scenarios used in this study, immune stimulation was only observed at local tissue concentrations orders of magnitude higher than what would be experienced daily by a MoM hip implant patient with a normal-functioning device. It is plausible that an unknown or unexpected mechanism with repeated dosing might affect the induction threshold dose for local immune effects. 5. Conclusions Overall, the data from this study indicate that the PLNA is a useful model for assessing the immune stimulating potential of metal particles and metal ions, including Co2+ and Cr3+. Immune stimulation was not observed four days after injection of Cr2O3 particles alone at all doses tested in the PLNA. The provisional LOAEL identified in this study for immune stimulation corresponds to experimental Co (254,700 μg/L) and Cr (111,400 μg/L) concentrations, roughly 2000 times higher than Co (113.4 μg/L) and Cr (54 μg/L) concentrations found in the synovial fluid of normal-functioning MoM hip implant patients. Taken together, under these study conditions, the provisional NOAEL and LOAEL identified in this study indicate that normal wear conditions are unlikely to result in immune stimulation in individuals not previously sensitized to metals based on the estimated corresponding Co and Cr concentrations in the synovial fluid. Conflict of interest Nine of the authors (BT, KU, BW, MK, EF, WC, ED, BF and DP) are or were employed by Cardno ChemRisk, a consulting firm that provides scientific advice to the government, corporations, law firms, and various scientific/professional organizations. Cardno ChemRisk has been engaged by DePuy Orthopaedics, Inc., a manufacturer of prosthetic devices, some of which contain cobalt and chromium, to provide general consulting and expert advice on scientific matters, as well as litigation support. This paper was prepared and written exclusively by the authors, without review, or comment by DePuy employees or counsel. It is likely that this work will be relied upon in medical research, nutrition research and litigation. One of the authors (DJP) has previously testified on behalf of DePuy in hip implant litigation. It is possible that any or all of the authors may be called upon to serve as expert witnesses on behalf of DePuy. Funding for this study and the preparation of this paper was provided by DePuy. The study and the preparation of the paper, including the synthesis of the findings, the conclusions drawn and recommendations made are the exclusive professional work product of the authors and may not necessarily be those of their employer or the financial sponsor of the study. Transparency document The Transparency document associated with this article can be found, in the online version.

89

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.taap.2016.07.020.

References Abdouh, M., Krzystyniak, K., Flipo, D., Therien, H.M., Fournier, M., 1995. Cytometric profile of molybdenum-induced contact sensitization versus a strong allergen reaction to oxazolone in murine auricular lymph node (ALN) test. Int. J. Immunopharmacol. 17, 545–554. Albers, R., van der Pijl, A., Seinen, W., Pieters, R., Bloksma, N., 1996. The autoimmunogenic chemicals HgCl2 and diphenylhydantoin stimulate IgG production to TNP-Ficoll and TNP-OVA, supporting and extending the graft-versus-host hypothesis for chemical induction of autoimmunity. Immunology 89, 468–473. Api, A.M., Basketter, D., Lalko, J., 2015. Correlation between experimental human and murine skin sensitization induction thresholds. Cutan. Ocul. Toxicol. 34, 298–302. ASTM (American Society for Testing Materials), 2011. ASTM F1537-11. Standard Specification for Wrought Cobalt-28 Chromium-6 Molybdenum Alloys for Surgical Implant (UNS R31537, UNS R331538, and UNS R31539). American Society for Testing Materials, West Conshohocken, PA. Basketter, D.A., Clapp, C., Jefferies, D., Safford, B., Ryan, C.A., Gerberick, F., Dearman, R.J., Kimber, I., 2005. Predictive identification of human skin sensitization thresholds. Contact Dermatitis 53, 260–267. Basketter, D.A., Lea, L.J., Cooper, K., Stocks, J., Dickens, A., Pate, I., Dearman, R.J., Kimber, I., 1999. Threshold for classification as a skin sensitizer in the local lymph node assay: a statistical evaluation. Food Chem. Toxicol. 37, 1167–1174. Bauer, T.W., Campbell, P.A., Hallerberg, G., 2014. How have new bearing surfaces altered the local biological reactions to byproducts of wear and modularity? Clin. Orthop. 472, 3687–3698. Beraudi, A., Catalani, S., Montesi, M., Stea, S., Sudanese, A., Apostoli, P., Toni, A., 2013. Detection of cobalt in synovial fluid from metal-on-metal hip prosthesis: correlation with the ion haematic level. Biomarkers 18, 699–705. Billi, F., Benya, P., Kavanaugh, A., Adams, J., McKellop, H., Ebramzadeh, E., 2012. The John Charnley award: an accurate and extremely sensitive method to separate, display, and characterize wear debris: part 2: metal and ceramic particles. Clin. Orthop. 470, 339–350. Bozic, K.J., Kurtz, S., Lau, E., Ong, K., Chiu, V., Vail, T.P., Rubash, H.E., Berry, D.J., 2009. The epidemiology of bearing surface usage in total hip arthroplasty in the United States. J. Bone Joint Surg. Am. 91, 1614–1620. Brown, C., Lacharme-Lora, L., Mukonoweshuro, B., Sood, A., Newson, R.B., Fisher, J., Case, C.P., Ingham, E., 2013. Consequences of exposure to peri-articular injections of micro- and nano-particulate cobalt-chromium alloy. Biomaterials 34, 8564–8580. Bucholz, R.W., 2014. Indications, techniques and results of total hip replacement in the United States. Revista Médica Clínica Las Condes 25, 756–759. Campbell, P., Ebramzadeh, E., Nelson, S., Takamura, K., De Smet, K., Amstutz, H.C., 2010. Histological features of pseudotumor-like tissues from metal-on-metal hips. Clin. Orthop. 468, 2321–2327. Carey, J.B., Allshire, A., van Pelt, F.N., 2006. Immune modulation by cadmium and lead in the acute reporter antigen-popliteal lymph node assay. Toxicol. Sci. 91, 113–122. Catelas, I., Campbell, P.A., Bobyn, J.D., Medley, J.B., Huk, O.L., 2006. Wear particles from metal-on-metal total hip replacements: effects of implant design and implantation time. Proc. Inst. Mech. Eng. H J. Eng. Med. 220, 195–208. Catelas, I., Medley, J.B., Campbell, P.A., Huk, O.L., Bobyn, J.D., 2004. Comparison of in vitro with in vivo characteristics of wear particles from metal-metal hip implants. J. Biomed. Mater. Res. B Appl. Biomater. 70, 167–178. Cook, S.D., McCluskey, L.C., Martin, P.C., Haddad Jr., R.J., 1991. Inflammatory response in retrieved noncemented porous-coated implants. Clin. Orthop. 264, 209–222. Davda, K., Lali, F.V., Sampson, B., Skinner, J.A., Hart, A.J., 2011. An analysis of metal ion levels in the joint fluid of symptomatic patients with metal-on-metal hip replacements. J. Bone Joint Surg. (Br.) 93, 738–745. Davies, A.P., Willert, H.G., Campbell, P.A., Learmonth, I.D., Case, C.P., 2005. An unusual lymphocytic perivascular infiltration in tissues around contemporary metal-on-metal joint replacements. J. Bone Joint Surg. Am. 87, 18–27. De Smet, K., Calistri, A., De Haan, R., Campbell, P.A., Ebramzadeh, E., Pattyn, C., Gill, H.S., 2008. Metal ion measurement as a diagnostic tool to identify problems with metalon-metal hip resurfacing. J. Bone Joint Surg. Am. 90, 202–208. Drummond, J., Tran, P., Fary, C., 2015. Metal-on-metal hip arthroplasty: a review of adverse reactions and patient management. J. Funct. Biomater. 6, 486–499. Fehring, T.K., Carter, J.L., Fehring, K.A., Odum, S.M., Griffin, W.L., 2015. Cobalt to chromium ratio is not a key marker for ALTR in metal on metal hips. J. Arthroplast. 30, 107–109. Fernandez Cabezudo, M.J., Petroianu, G., Al-Ramadi, B., Langer, R.D., 2007. Iosimenol, a new non-ionic dimeric contrast medium, does not induce immunoreactivity in the popliteal lymph node assay. Br. J. Radiol. 80, 713–718. Finley, B.L., Monnot, A.D., Gaffney, S.H., Paustenbach, D.J., 2012a. Dose-response relationships for blood cobalt concentrations and health effects: a review of the literature and application of a biokinetic model. J. Toxicol. Environ. Health, B 15, 493–523. Finley, B.L., Monnot, A.D., Paustenbach, D.J., Gaffney, S.H., 2012b. Derivation of a chronic oral reference dose for cobalt. Regul. Toxicol. Pharmacol. 64, 491–503. Finley, B.L., Unice, K.M., Kerger, B.D., Otani, J.M., Paustenbach, D.J., Galbraith, D.A., Tvermoes, B.E., 2013. 31-day study of cobalt (II) chloride ingestion in humans: pharmacokinetics and clinical effects. J. Toxicol. Environ. Health A 76, 1210–1224.

90

B.E. Tvermoes et al. / Toxicology and Applied Pharmacology 308 (2016) 77–90

Gerber, J.C., et al., 2011. Guide for the Care and Use of Laboratory Animals. eighth ed. Institute of Laboratory Animal Resources, National Research Council, Washington, DC. Gerberick, G.F., Cruse, L.W., Ryan, C.A., Hulette, B.C., Chaney, J.G., Skinner, R.A., Kimber, I., 2002. Use of a B cell marker (B220) to discriminate between allergens and irritants in the local lymph node assay. Toxicol. Sci. 68 (2), 420–428. Girard, J., Bocquet, D., Autissier, G., Fouilleron, N., Fron, D., Migaud, H., 2010. Metal-onmetal hip arthroplasty in patients thirty years of age or younger. J. Bone Joint Surg. Am. 92, 2419–2426. Griem, P., Goebel, C., Scheffler, H., 2003. Proposal for a risk assessment methodology for skin sensitization based on sensitization potency data. Regul. Toxicol. Pharmacol. 38, 269–290. Hallab, N., Merritt, K., Jacobs, J.J., 2001. Metal sensitivity in patients with orthopaedic implants. J. Bone Joint Surg. Am. 83, 428–436. Hart, A.J., Sabah, S.A., Henckel, J., Lloyd, G., Skinner, J.A., 2015. Lessons learnt from metalon-metal hip arthroplasties will lead to safer innovation for all medical devices. Hip Intl. http://dx.doi.org/10.5301/hipint.5000275. Hussain, S., Smulders, S., De Vooght, V., Ectors, B., Boland, S., Marano, F., Van Landuyt, K.L., Nemery, B., Hoet, P.H., Vanoirbeek, J.A., 2012. Nano-titanium dioxide modulates the dermal sensitization potency of DNCB. Part. Fibre Toxicol. 9, 15. Ikarashi, Y., Ohno, K., Tsuchiya, T., Nakamura, A., 1992a. Differences of draining lymph node cell proliferation among mice, rats and guinea pigs following exposure to metal allergens. Toxicology 76 (3), 283–292. Ikarashi, Y., Tsuchiya, T., Nakamura, A., 1992b. Detection of contact sensitivity of metal salts using the murine local lymph node assay. Toxicol. Lett. 62, 53–61. ISO (International Organization for Standardization), 2010. ISO 10993-10:2010. Biological Evaluation of Medical Devices. Part 10, Tests for Irritation and Skin Sensitization. third ed. ISO, Geneva August 1, 2010. Jacobs, J.J., Urban, R.M., Hallab, N.J., Skipor, A.K., Fischer, A., Wimmer, M.A., 2009. Metalon-metal bearing surfaces. J. Am. Acad. Orthop. Surg. 17, 69–76. Kammuller, M.E., Thomas, C., De Bakker, J.M., Bloksma, N., Seinen, W., 1989. The popliteal lymph node assay in mice to screen for the immune disregulating potential of chemicals–a preliminary study. Int. J. Immunopharmacol. 11, 293–300. Kwon, Y.-M., Ostlere, S.J., McLardy-Smith, P., Athanasou, N.A., Gill, H.S., Murray, D.W., 2011. “Asymptomatic” pseudotumors after metal-on-metal hip resurfacing arthroplasty: prevalence and metal ion study. J. Arthroplast. 26, 511–518. Langton, D.J., Jameson, S.S., Joyce, T.J., Hallab, N.J., Natu, S., Nargol, A.V., 2010. Early failure of metal-on-metal bearings in hip resurfacing and large-diameter total hip replacement: a consequence of excess wear. J. Bone Joint Surg. (Br.) 92, 38–46. Langton, D.J., Joyce, T.J., Jameson, S.S., Lord, J., Van Orsouw, M., Holland, J.P., Nargol, A.V., De Smet, K.A., 2011. Adverse reaction to metal debris following hip resurfacing: the influence of component type, orientation and volumetric wear. J. Bone Joint Surg. (Br.) 93, 164–171. Lass, R., Grubl, A., Kolb, A., Stelzeneder, D., Pilger, A., Kubista, B., Giurea, A., Windhager, R., 2014. Comparison of synovial fluid, urine, and serum ion levels in metal-on-metal total hip arthroplasty at a minimum follow-up of 18 years. J. Orthop. Res. 32, 1234–1240. Madl, A.K., Kovochich, M., Liong, M., Finley, B.L., Paustenbach, D.J., Oberdorster, G., 2015a. Toxicology of wear particles of cobalt-chromium alloy metal-on-metal hip implants part II: importance of physicochemical properties and dose in animal and in vitro studies as a basis for risk assessment. Nanomedicine 11, 1285–1298. Madl, A.K., Liong, M., Kovochich, M., Finley, B.L., Paustenbach, D.J., Oberdorster, G., 2015b. Toxicology of wear particles of cobalt-chromium alloy metal-on-metal hip implants part I: physicochemical properties in patient and simulator studies. Nanomedicine 11, 1201–1215. Mao, X., Wong, A.A., Crawford, R.W., 2011. Cobalt toxicity—an emerging clinical problem in patients with metal-on-metal hip prostheses? Med. J. Aust. 194, 649–651. Matharu, G.S., Pandit, H.G., Murray, D.W., Treacy, R.B., 2015. The future role of metal-onmetal hip resurfacing. Int. Orthop. 39, 2031–2036. Mirra, J.M., Marder, R.A., Amstutz, H.C., 1982. The pathology of failed total joint arthroplasty. Clin. Orthop. 170, 175–183. Monnot, A.D., Christian, W.V., Paustenbach, D.J., Finley, B.L., 2014. Correlation of blood Cr (III) and adverse health effects: application of PBPK modeling to determine non-toxic blood concentrations. Crit. Rev. Toxicol. 44 (7), 618–637. Morlock, M.M., Bishop, N., Zustin, J., Hahn, M., Ruther, W., Amling, M., 2008. Modes of implant failure after hip resurfacing: morphological and wear analysis of 267 retrieval specimens. J. Bone Joint Surg. Am. 90 (Suppl. 3), 89–95. Moss, S.G., Schweitzer, M.E., Jacobson, J.A., Brossmann, J., Lombardi, J.V., Dellose, S.M., Coralnick, J.R., Standiford, K.N., Resnick, D., 1998. Hip joint fluid: detection and distribution at MR imaging and US with cadaveric correlation. Radiology 208 (1), 43–48. Nasser, S., 2007. Orthopedic metal immune hypersensitivity. Orthopedics 30, 89–91. Pieters, R., Albers, R., 1999a. Assessment of autoimmunogenic potential of xenobiotics using the popliteal lymph node assay. Methods 19 (1), 71–77.

Pieters, R., Albers, R., 1999b. Screening tests for autoimmune-related immunotoxicity. Environ. Health Perspect. 107 (Suppl. 5), 673. Pinson, M.L., Coop, C.A., Webb, C.N., 2014. Metal hypersensitivity in total joint arthroplasty. Ann. Allergy Asthma Immunol. 113, 131–136. Reito, A., Parkkinen, J., Puolakka, T., Pajamaki, J., Eskelinen, A., 2015. Diagnostic utility of joint fluid metal ion measurement for histopathological findings in metal-on-metal hip replacements. BMC Musculoskelet. Disord. 16, 393. Sampson, B., Hart, A., 2012. Clinical usefulness of blood metal measurements to assess the failure of metal-on-metal hip implants. Ann. Clin. Biochem. 49, 118–131. SCENIHR (Scientific Committee on Emerging and Newly Identified Health Risks), 2014f. Scientific Committee on Emerging and Newly Identified Health Risks (SCENIHR) Opinion on the Safety of Metal-on-Metal Joint Replacements with a Particular Focus on Hip Implants. European Commission, Luxembourg. Schmidt, M., Goebeler, M., 2015. Immunology of metal allergies. J. Dtsch. Dermatol. Ges. 13, 653–659. Schuhmann, D., Kubicka-Muranyi, M., Mirtschewa, J., Günther, J., Kind, P., Gleichmann, E., 1990. Adverse immune reactions to gold. I. Chronic treatment with an Au(I) drug sensitizes mouse T cells not to Au(I), but to Au(III) and induces autoantibody formation. J. Immunol. 145, 2132–2139. Schuppe, H.C., Haas-Raida, D., Kulig, J., Bomer, U., Gleichmann, E., Kind, P., 1992. T-cell-dependent popliteal lymph node reactions to platinum compounds in mice. Int. Arch. Allergy Immunol. 97, 308–314. Sidaginamale, R.P., Joyce, T.J., Lord, J.K., Jefferson, R., Blain, P.G., Nargol, A.V., Langton, D.J., 2013. Blood metal ion testing is an effective screening tool to identify poorly performing metal-on-metal bearing surfaces. Bone Joint Res. 2, 84–95. Sieber, H.P., Rieker, C.B., Kottig, P., 1999. Analysis of 118 second-generation metal-onmetal retrieved hip implants. J. Bone Joint Surg. (Br.) 81, 46–50. Stiller-Winkler, R., Radaszkiewicz, T., Gleichmann, E., 1988. Immunopathological signs in mice treated with mercury compounds–I. Identification by the popliteal lymph node assay of responder and nonresponder strains. Int. J. Immunopharmacol. 10, 475–484. Suda, A., Iwaki, Y., Kimura, M., 2000. Differentiation of responses to allergenic and irritant compounds in mouse popliteal lymph node assay. J. Toxicol. Sci. 25, 131–136. Thomas, W.R., Cunningham, P.T., 2009. Hypersensitivity: Immunological. Encyclopedia of Life Sciences (ELS). John Wiley & Sons Ltd, Chichester, pp. 1–11. Tower, S., 2010. Arthroprosthetic cobaltism: identification of the at-risk patient. Alaska Med. 52, 28–32. Tvermoes, B.E., Paustenbach, D.J., Kerger, B.D., Finley, B.L., Unice, K.M., 2015. Review of cobalt toxicokinetics following oral dosing: implications for health risk assessments and metal-on-metal hip implant patients. Crit. Rev. Toxicol. 45, 367–387. Tvermoes, B.E., Unice, K.M., Paustenbach, D.J., Finley, B.L., Otani, J.M., Galbraith, D.A., 2014. Effects and blood concentrations of cobalt after ingestion of 1 mg/d by human volunteers for 90 d. Am. J. Clin. Nutr. 99, 623–646. USEPA (United States Environmental Protection Agency), 2011. Recommended Use of Body Weight 3/4 as the Default Method in Derivation of the Oral Reference Dose. EPA/100/R11/00010 Final. Environmental Protection Agency, Washington, D.C. USFDA (United States Food and Drug Administration), 2002. Guidance for Industry. Immunotoxicology Evaluation of Investigational New Drugs. October 2002. U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research, Washington, D.C. Wang, Y., Dai, S., 2013. Structural basis of metal hypersensitivity. Immunol. Res. 55, 83–90. Warheit, D.B., Hoke, R.A., Finlay, C., Donner, E.M., Reed, K.L., Sayes, C.M., 2007. Development of a base set of toxicity tests using ultrafine TiO 2 particles as a component of nanoparticle risk management. Toxicol. Lett. 171 (3), 99–110. Watters, T.S., Cardona, D.M., Menon, K.S., Vinson, E.N., Bolognesi, M.P., Dodd, L.G., 2010. Aseptic lymphocyte-dominated vasculitis-associated lesion: a clinicopathologic review of an underrecognized cause of prosthetic failure. Am. J. Clin. Pathol. 134, 886–893. WHO (World Health Organization), 1999. Principles and methods for assessing allergic hypersensitization associated with exposure to chemicals. Environmental Health Criteria Volume 212. World Health Organization, Geneva. WHO (World Health Organization), 2006. Principles and methods for assessing autoimmunity associated with exposure to chemicals. Environmental Health Criteria 236. WHO Press, Geneva. Willert, H.G., Buchhorn, G.H., Fayyazi, A., Lohmann, C.H., 2000. Histopathologische veranderungen bei metall/metall-gelenken geben hinweise auf eine zellvermittelte uberempfindlichkeit. Osteologie 9, 165–179. Winans, B., Tvermoes, B.E., Unice, K.M., Kovochich, M., Christian, W.V., Donovan, E., Fung, E.S., Finley, B.L., Kimber, I., Paustenbach, D.J., 2016. Data on Immune Stimulation Following Exposure to Metal Particles and Ions Using the Mouse Popliteal Lymph Node Assay (Data in Brief, accepted for publication).