Search

This review is dedicated to John Newsom-Davis, who was an exceptional colleague and friend, always exchanging ideas with respect and consideration. We shall not forget his involvement and passion in search for the truth on the role of thymectomy in the management of Myasthenia Gravis (MG). In this short review, we shall summarize what we learnt from DNA microarrays applied to MG thymus. We shall focus on three main comparisons of the thymic transcriptomes: 1) highly hyperplastic MG patients versus non-MG adults; 2) corticosteroid-treated versus untreated seropositive MG patients; and 3) seronegative versus seropositive MG patients.

The thymus is believed to play an important role in the pathogenesis of myasthenia gravis (MG). The 80% of MG patients with anti-acetylcholine receptor autoantibodies fall into three clinical subgroups: 1) thymoma; 2) early-onset MG (

Here we show that in myasthenic thymus several cell types, including thymic epithelial cells (TEC) and immune cells, were the source and the target of the neurotrophic factor brain-derived growth factor (BDNF). Interestingly, many actively proliferating medullary thymocytes expressed the receptor TrkB in vivo in involuted thymus, while this population was lost in hyperplastic or neoplastic thymuses. Furthermore, in hyperplastic thymuses the robust coordinated expression of BDNF in the germinal centers together with the receptor p75NTR on all proliferating B cells strongly suggests that this factor regulates germinal center reaction. Finally, all TEC dying of apoptosis expressed BDNF receptors, indicating that this neurotrophin is involved in TEC turnover. In thymomas both BDNF production and receptor expression in TEC were strongly hindered. This may represent an attempt of tumour escape from cell death.

The alternative TrkAIII splice variant is expressed by murine and human thymus. Alternative TrkAIII splicing predominates in postembryonic day E13 (E17 and E18), postnatal murine (3 week and 3 month) and human thymuses, with TrkAIII mRNA expressed by selected thymocyte subsets and thymic epithelial cells (TECs) and a 100 kDa immunoprecipitable TrkAIII-like protein detected in purified thymocyte and whole thymus extracts. FACS and immunohistochemical analysis indicate a non-cell surface localisation for the TrkAIII-like protein in cortical CD4+/CD8+ double positive and, to a lesser extent, single positive thymocyte subsets at the cortex/medulla boundary and in Hassle's corpuscles, reticular epithelial and dendritic cells of the thymic medulla. TrkA(I/II) expression, on the other hand, predominates in sub-capsular regions of the thymus. TrkAIII-like immunoreactivity at the cortex/medulla boundary associates with regions of thymocyte proliferation and not apoptosis. A potential role for thymic hypoxia in thymocyte alternative TrkAIII splicing is supported by reversal to TrkAI splicing by normoxic but not hypoxic culture and induction of Jurkat T cell alternative TrkAIII splicing by the hypoxia mimic CoCl2. In contrast, TEC expression of TrkAIII predominates in both normoxic and hypoxic culture conditions. The data support a potential role for TrkAIII in thymic development and function, of particular relevance to intermediate stage CD4+/CD8+ thymocyte subsets and TECs, which potentially reflects a reversible thymocyte and more permanent TEC adaptation to thymic environment. Since intracellular TrkAIII neither binds nor responds to NGF and can impede regular NGF/TrkA signalling (Tacconelli et al., Cancer Cell, 2004), its expression would be expected to provide an alternative and/or impediment to regular NGF/TrkA signalling within the developing and developed thymus of potential functional importance.