Adaptive and Auto-Immunity| Volume 139, ISSUE 5, SUPPLEMENT , S7, May 2019

041 Anti-thyroid peroxidase antibodies target a cytoplasmic protein in keratinocytes and may contribute to blister formation in pemphigus vulgaris

      In addition to having reactivity to the classical targets, desmoglein (Dsg)3 and -1, there is accumulating evidence that Pemphigus vulgaris (PV) patients harbor non-Dsg autoantibodies. The disease relevance of non-Dsg targets in PV is under investigation. We have shown that PV patients carry anti-thyroid peroxidase (TPO) antibodies at significantly higher levels than healthy controls and that anti-TPO reactivity is driven by HLA status and the absence of Dsg-reactivity. We have also shown functional effects of anti-TPO on cell adhesion and cellular processes associated with blister formation in human keratinocytes. TPO is known as an enzyme primarily involved in iodination of thyroglobulin in the thyroid gland with no known function in the skin. The exact target of anti-TPO antibodies in skin remains unclear. To clarify, we assessed TPO mRNA and protein expression by RT-PCR, Western Blot (WB) and immunofluorescence (IF) in the human keratinocyte cell line HaCaT. We show that TPO is indeed expressed at the mRNA level in HaCaT cells by PCR. At the protein level, we show definitive cytoplasmic staining with anti-TPO Abs, with no overlap with anti-Dsg3 binding or cytoskeletal components by IF. Interestingly, by Western Blot, we detect binding of anti-TPO antibody to a protein of ∼75kDA size, which is not the predicted size for the canonical TPO variant (103 kDa). Variants of TPO of sizes close to 75 kDa have been described. Additionally, a number of TPO orthologs such as lacto-, myelo- and eosinophil peroxidase exhibit a high degree of sequence similarity to TPO ectodomains. Our data conclusively indicate that anti-TPO antibodies do target a keratinocyte-associated molecule, which may be a thus far unknown isoform of TPO or a structurally similar target antigen. Precise molecular characterization of this molecule will be a necessary step to understand the mechanisms by which anti-TPO autoantibodies modulate keratinocyte function in the context of PV.