Elucidation of the cellular sensing mechanisms for the contact allergens nickel and cobalt
Allergic contact dermatitis (ACD) caused by the metal ions Ni2+ and Co2+ affect almost one in five people in the western world. Sources for these metals include numerous dayto-day products like metal coins, costume jewelry, body piercing, medical implants, hair dyes and ceramics. Moreover, both metals are widely used in the building, metallurgical ... and electronic industries classifying metal-induced ACD as major work-related disease with considerable impact on professional careers and working ability of affected individuals. Despite the undisputed need for specific and well-tolerated therapies, current treatment options are limited to the use of topical corticosteroids, which are prone to major side effects, particularly upon the long-term administration obligatory with those patients. Here, we dissected the molecular mechanism required for Ni2+-induced hTLR4 activation and extended our analysis to Co2+-induced proinflammatory responses. Using supplementation experiments in naturally hTLR4-deficent primary human keratinocytes and HEK293 cells we show that Co2+ resembles Ni2+ by activating hTLR4/MD2 in a species-dependent manner requiring presence of the non-conserved histidines H456 and H458 within the proposed metal-binding domain of hTLR4 that flanks its dimerisation interface. Using a novel dimerisation-mutant of hTLR4 we provide first direct evidence that metal allergens, similar to LPS, strictly require hTLR4 dimerisation to trigger proinflammatory gene expression. This confirms our structural model, which predicts hTLR4 activation to occur by bridging of two hTLR4 molecules simultaneously bound by two metal ions to the non-conserved histidines H456 and H458. Most importantly, however, we show here that Ni2+ and Co2+, in contrast to LPS, do not require MD2 to induce hTLR4 dimerisation since both metals readily induced complex formation of differently tagged hTLR4 proteins in co-immunoprecipitation studies with MD2-free HEK293 cells. We believe that this observation holds great potential for therapeutic exploitation since administration of a soluble form of hTLR4 (sTLR4) produced in the absence of MD2 was capable to significantly reduce Ni2+ and Co2+-induced cytokine production in metal-sensitive cells while LPS responsiveness remained unchanged. In summary, our data show that metal allergens differ in at least two aspects from LPS: For one, they strictly require presence of the non-conserved histidines H456 and H458 to activate hTLR4 and, secondly, they induce hTLR4 dimerisation independently of MD2.Thus, our results may provide the basis for novel therapies that aim at specifically inhibiting metal-induced ACD in a clinically manageable way.