Th1 type lymphocyte reactivity to metals in patients with total hip arthroplasty

Implant related metal hypersensitivity remains a relatively unpredictable and poorly understood phenomenon. [1–3] Dermal hypersensitivity to metal is common, affecting about 10–15% of the population. [1, 2, 4, 5] Dermal contact and ingestion of metals have been reported to cause immune reactions which most typically manifest as skin hives, eczema, redness and itching. [1, 5–7] Although little is known about the short and long term pharmacodynamics and bioavailability of circulating metal degradation products in vivo,[4, 8, 9, 9, 10] there are many case and group studies documenting metal reactivity responses temporally associated with implantation of metal components.

All metals in contact with biological systems corrode[11, 12] and the released metal ions, while not sensitizers on their own, can act as haptens, activating the immune system by forming complexes with native proteins. [4, 13, 14] Metal-protein complexes are considered to be candidate antigens for eliciting hypersensitivity responses. Metals known as sensitizers (haptenic moieties in antigens) are beryllium,[15] nickel,[5–7, 15] cobalt[15] and chromium[15] while occasional responses have been reported to tantalum,[16] titanium [17, 18] and vanadium. [16] Nickel is the most common metal sensitizer in humans followed by cobalt and chromium. [1, 4–7]

The specific T-cell subpopulations associated with metal hypersensitivity, the cellular mechanism of recognition/activation, and the antigenic metal-protein determinants created by implant metals remain incompletely characterized. Th1 (T-helper type 1) and Th2 (T-helper type 2) correspond to CD4+ αβ TCR T cell subsets that provide help to cells of both the innate and adaptive immune systems. We hypothesized that adaptive Th1 mediated responses will be the more prevalent type of hypersensitivity response to metal in patients with total hip arthroplasty (THA). We investigated this hypothesis by evaluating the reactivity of peripheral blood lymphocytes to metal challenge agents (Cobalt, Chromium, Nickel and Titanium) in subjects with and without hip arthroplasties using proliferation and cytokine assays. We studied lymphocyte responses in three different subject groups: i) young healthy individuals without implants and without a history of metal allergy were used to assess general lymphocyte reactivity to various concentrations of metal challenge agents; ii) individuals with hip arthroplasty were determined to be either metal-reactive or metal-nonreactive by cell proliferation studies, and iii) age matched controls without implants were used to assess lymphocyte subtype dominance by cytokine release profiles.

Contrasting the metal reactivity responses of healthy subjects without implants and subjects with implants was used to assess the immunogenic effects of metals found retrospectively in orthopaedic implant cohorts (n = 31 total subjects). Informed consent was obtained from all subjects after Institutional Review Board review and approval. Group 1 consisted of young healthy subjects without any metallic implants (n = 8, 4 male and 4 female subjects with an average age of 29, range 23 to 36) and was primarily used for determination of relative metal toxicity and upper limits of challenge dose. In Group 1 subjects lymphocyte proliferation in response to a variety of metals (Aluminum-Al⁺³, Cobalt-Co⁺², Chromium-Cr⁺³, Copper-Cu⁺², Iron-Fe⁺³, Molybdenum-Mo⁺⁵, Manganeese-Mn⁺², Nickel-Ni⁺², Vanadium-V⁺³ and Sodium-Na⁺² chloride solutions, Sigma, St Louis, MO) at a variety of concentrations (0.0, 0.05, 0.1, 0.5, 1.0 and 10.0 mM) was studied to establish toxicity thresholds. Group 2 consisted of 15 hip arthroplasty single-implant type subjects with well functioning (average Harris Hip Score = 91) total hip arthroplasties of the same design (Harris-Galante, Zimmer, Warsaw, IN) with 7 males and 8 females (average age 69 yrs, range 55–80 yrs), and an average hip arthroplasty time in-situ of 158 months (range 117–174 months). This implant system is comprised of a Titanium-6%Aluminum-4%Vanadium alloy stem with a Cobalt-Chromium-Molybdenum femoral head, an Ultrahigh Molecular Weight Polyethylene (UHMWPE) acetabular cup in a Titanium-6%Aluminum-4%Vanadium alloy lining. A sample size of 14 in each group will have an 80% power to detect a probability of 0.81 that an observation in one group is less than an observation in the other group using a two-sided Mann-Whitney test with a 0.05 significance level. Two of 15 reported a history of metal allergy and 5 of 15 had moderate osteolysis (proximal focal lesions in excess of 0.5 cm² in total area on an anteroposterior radiograph and/or distal [diaphyseal] focal lesions greater than 1 cm² in total area on an anteroposterior radiograph were correlated with the magnitude of articular surface wear). [19] Group 3 subjects were age-matched controls (to Group 2) with no implants (Group 3, n = 8, 4 male, 4 female, average age 70, range 49 to 80). Group 2 subjects with hip arthroplasty were evaluated for lymphocyte reactivity to metals using lymphocyte transformation testing (proliferation assays), and IFN-gamma, IL-2, and IL-4 cytokine analysis. The metal challenge agents for Group 2 subjects were a subset of the more prevalent implant-alloy metals used for appropriate dose determination in Group 1 and consisted of 0.1 mM CrCl₃, 0.1 mM NiCl₂, 0.1 mM CoCl₂ (Sigma, St. Louis, MO) and approx. 0.001 mM titanium (using Titanium saturated culture medium produced through incubation with Titanium beads). Despite its prominence as an implant alloy, titanium could only be tested at such low concentrations because of the insoluble nature of titanium at physiologic pH and its subsequent inability to form ions in solution.

Proliferation assays

(Lymphocyte Transformation Tests): Serum and Peripheral Blood Mononuclear Cells (PBMCs) were obtained from mononuclear cell fractionation of blood obtained by peripheral venipuncture after obtaining informed consent and Institutional Review Board approval. Peripheral blood mononuclear cells were isolated from 30 mL of peripheral blood using density gradient separation (Ficoll-isopaque, Pharmacia, Piscataway, NJ). Ficoll separated mononuclear cells are generally comprised of 85–95% lymphocytes with 5–13% monocytes and <0.1% dendritic cells with limited contamination (i.e. <5% erythrocytes and <3 granulocytes). Among mixed peripheral blood mononuclear cells populations, lymphocytes are the only cells capable of competitively significant in vitro proliferation upon challenge with an allergen. Thus we used standard Lymphocyte Transformation Testing (LLT) protocol of mononuclear cells to measure lymphocyte proliferation and cytokine production (15–30 × 10⁶ cells per subject). [20, 21] Lymphocytes in all assays were incubated with Dulbeccos Modified Eagles Medium and 10% autologous serum with either no metal (plain media) as a negative control, 0.01 mg/ml phytohemagglutinin (PHA) as a positive control, or metal. Each metal challenge treatment was conducted in quadruplicate (4 wells/treatment). [³H]-thymidine was added (1 mCi/culture well) during the last 12 hours of incubation after 5 1/2 days of treatment. At day six, [³H]-thymidine uptake was measured using liquid scintillation. Incorporated radioactivity was measured using liquid scintillation Beta plate analysis (Wollac Gatesburg, MD). The amount of [3H]-thymidine incorporation for each metal treatment was normalized to that of the nontreated control producing a ratio, referred to here as the proliferation index or stimulation index, SI.

The SI was calculated using measured radiation counts per minute (cpm):

(1)

Six days of incubation were used to reproduce, in vitro, the time lag associated with in vivo lymphocyte proliferation in a delayed type hypersensitivity (DTH) response. [20, 21]

The detection of in increase in proliferation and IFN-gamma in the absence of detectable IL-4 in the PBMCs of metal-reactive hip arthroplasty patients (SI>2) supports our hypothesis that Th1 type reactivity may dominate lymphocyte reactivity responses to metals in patients with TJRs. This finding helps to identify the cellular pathways in metal-induced reactivity responses to orthopaedic implants. Typically, the principle function of Th1 cells is to aid other leukocytes, i.e. macrophages and natural killer (NK) cells, in the protective response against intracellular microbes.

Lymphocyte metal-reactivity reported here using controls (primary human lymphocytes from 8 healthy individuals with no implants and no history of metal allergy revealed that more reactive metals induce greater lymphocyte reactivity at higher concentrations tested, i.e. Nickel and Chromium (0.05–0.5 mM). Other metals (i.e. Manganese, Copper, and Vanadium) induced cell toxicity at concentrations where other metal ions (Nickel) induced a proliferative response (as low as 0.05 mM). However, metal challenge at 0.1 mM demonstrated no toxicity or statistically elevated reactivity in Group 1 normal controls (Fig 1) and was thus used as an appropriately high dose for Group 2 subjects with implants. These results of Group 1 controls are consistent with past patch-test investigations where Nickel has shown the highest prevalence of metal reactivity among the general population at an approximately 14% incidence. [1] Roughly 2 of 8 (25%) of the Group 1 young healthy controls without implants demonstrated metal reactivity to Cobalt, Chromium, or Nickel (at 0.1 mM) compared to 7 of 15 (46%) of the Group 2 subjects with well functioning hip arthroplasties (SI>2, p < 0.05 t-test) and 1 of 8 (12%) of the Group 3 age matched controls. This incidence of metal reactivity in subjects with implants (Group 2) is higher than that of the general population with total joint arthroplasty, yet lower than a previously reported 60% incidence associated with failing TJAs (prior to revision). [25–27] These results support previous reports indicating greater sensitivity associated with lymphocyte transformation testing than dermal patch testing. [28–33]

The subjects with well performing hip arthroplasties demonstrated a higher incidence and greater degree of metal reactivity, as determined by lymphocyte transformation testing to chromium, than healthy controls. This may indicate a sensitizing effect of hip arthroplasty on peripheral lymphocytes. Although the relationship between lymphocyte transformation testing and clinical TJA outcome remains undetermined, these results are consistent with the theory of individual dependent generalized idiopathic gradual sensitization to metal debris from implant degradation. The clinical significance of these findings is unknown, in part, because it is not currently feasible to compare implant performance in prospective groups with and without metal reactivity. Continued efforts to follow-up and correlate implant performance with lymphocyte reactivity will ultimately help determine the clinical utility and diagnostic capability of such assays.

These reactivity results are novel in that direct cytokine evidence from metal-stimulated PBMCs of individuals with hip arthroplasties supports the hypothesis that a Th1 activation type paradigm is identifiable in as-tested metal-reactive subjects with TJA. This Th1 reactivity may be important to the pathogenesis of aseptic osteolysis. [34, 35] While activated osteoclasts are considered to be the effector cell type in bone resorption, all cell types within the periprosthetic space can contribute to the osteolytic cascade. Particles phagocytosed by macrophages and metal-protein complexes interacting with lymphocytes can lead to the production of factors, which act in both an autocrine and paracrine fashion to contribute to either increased bone resorption or reduced bone formation. IFN-gamma is produced by Th1, cytotoxic T-cells and Natural Killer T-cells and plays a pivotal role in the regulation of immune responses. IFN-gamma has been previously reported to be a potent inhibitor of osteoblast functions directly [36–38] and has been established as the predominant macrophage activating factor by priming or stimulating the expression of both class I and class II MHC (major histocompatibility complex) molecules, additional co-stimulatory macrophage cytokines (e.g. IL-1, IL-6 and TNF-α) and surface molecules (e.g. ICAM-1, B7, and CD40). [39–42]

When combined, the release of cytokines IFN-gamma and IL-2 by lymphocytes can induce macrophage production of TNF-α which can establish an autocrine stimulatory loop required for continued macrophage activation and recruitment. [43–47] Therefore the present study demonstrates likely mechanisms by which lymphocyte mediated metal reactivity may contribute to the etiology of macrophage- and particle-induced osteolysis (Figure 6). IFN-gamma released by metal-activated Th1 lymphocytes is also important to macrophage activation because it is necessary for TNF-alpha released from macrophages to synergize with IFN-gamma in an autocrine fashion to induce a variety of macrophage activation genes including nitric oxide synthase, also implicated in aseptic osteolysis. [37, 48] Therefore, it may be less important that a Th1 or Th2 paradigm has been determined by cytokine profiling and more important that certain cytokines themselves have been identified especially in the context of recent reports where the identification of discrete T-helper cell populations seem to be constantly updated and revised. [23]

IL-4, in the presence of activated T-cells, is able to trigger an IL-4-dependent activation of antibody secreting B-cells from resting B-cell populations. The implications of the lack of IL-4 secretion from lymphocytes in Group 2a and 2b subjects is that B-cell reactivity is not dominantly stimulated in these in vitro elicited metal responses (Figure 7). [42, 49] Th2 cytokines such as IL-4 and IL-10 have been shown to have very powerful inhibitory effects on nearly every facet of macrophage function. Both IL-4 and IL-10 inhibit the production of such inflammatory cytokines as IL-6 and TNF-alpha [43, 47] and the generation of reactive oxygen intermediates. [42, 50, 51] Given that the osteolysis-inducing effects of IFN-gamma released from Th1 lymphocytes are in direct contrast to the bone formation effects of released from Th2 T-cells (Figs 6 and 7), if metal reactivity responses of hip arthroplasty patients resulted in IL-4 release from Th2 cells, the ramifications to peri-implant bone homeostasis would be vastly different than the Th1-derived IFN-gamma release detected in this investigation. [24, 43, 47, 52]

Peri-implant lymphocyte response of an individual to metal challenge remains a complex process, which can only be approximated in vitro. Our findings of IFN-gamma release in the absence of IL-4 secretion in those subjects with hip arthroplasty that demonstrated metal-reactivity (SI>2, p < 0.05) support a Th1-type reactivity paradigm. The involvement of a specific lymphocyte subtype (Th1) in the metal reactivity response implicates an adaptive immunity response, which represents a departure from the innate-only immunity response (nonspecific) typically associated with implant degradation products. It remains uncertain to what degree lymphocyte metal-reactivity contributes to the pathogenesis of poor implant performance. Continued clinical correlation between metal reactivity and implant performance (with LTT-like assays) as well as further investigation into basic mechanisms of metal antigenicity are required to build a more certain etiological connection.