Molecular Characterisation of MIC2/E2 (CD99) Ligands

There are two specific aims to this proposal, firstly to identify the molecular ligand(s) for the ectodomain of CD99, and secondly to use biochemical and genetic approaches to identify downstream effectors which interact with CD99 intracellular domains. The CD99 ectodomain ligand(s) will be screened with recombinant CD99::IgG-fusion proteins and will be enriched by classical expression cloning. As a technical inovation we plan within this project to develop a panel of Fc-receptor independent fusion scaffolds, which will facilitate the screening of libraries from Fe-receptor expressing cells. CD99 has originally been described as a human thymus leukemia antigen, an Ewing`s sarcoma specific membrane marker molecule and as a putative adhesion molecule (termed E2) involved in spontaneous rosette formation of T cells with erythrocytes. The type I integral membrane protein CD99 is the product of the mic-2 gene which is located in the pseudoautosomal (pairing) region of human X and Y chromosomes. Particularly high CD99 expression levels with a strong membranous immunoreactivity are characteristic for Ewing`s sarcoma cells and peripheral primitive neuroectodermal tumors (pPNET). Ewing`s sarcoma, the second most common malignant bone tumor of children and young adults, is an aggressive osteolytic tumor with a marked propensity for dissemination. Immunohistochemical detection of strong membrane localized CD99 expression, especially with low affinity monoclonal antibodies, contributes to the differential diagnosis of "small round tumor cells", although it lacks absolute specificity. CD99 is also expressed on cells of hematopoietic origin, with particularly high expression on early CD34+ hematopoietic precursors. Significantly, experimental CD99 downregulation in B cells leads to the generation of a Sternberg-Reed giant cell like phenotype. These findings together with the notion that dominant active Rac as well as over-expression of CD99 are able to overcome phenotypic changes of cells towards Hodgkin cells, point to a critical role of CD99 in the regulation of cell morphology, differentiation and dignity. That CD99 is a signal tansducing molecule has been well established in several cellular systems. We and others have shown that ligation of CD99 induces a calcium flux in T cells and that CD99 co-stimuli lead to the activation of T cells, i.e. IL-2 promoter activation, proliferation, and induction of Th1-type cytokine production. In the absence of activating signals via the antigen receptor, especially in thymocytes, CD99 transduced signals target mainly pro- apoptotic pathways and result in phosphatidyl serine exposure at the cell membrane followed by programmed cell death. Physiological ligands for the CD99-ecotodomain and the intracellular CD99 domain have not been defined, so far. The characterization of such CD99 ligands will help to understand in more detail the cellular activation processes observed upon CD99 ligation in B and T lymphocytes and will also shed important light on the pathophysiology of malignant cell types which show either over-expression of CD99, such as Ewing sarcoma family tumors, or down- regulation of CD99, such as the Hodgkin cells of the Reed-Sternberg phenotype. The availability of recombinant CD99-ligands might ultimately become a useful tool to influence diseases in which the expression of CD99 is deregulated and will help to understand T cell co-stimulation more precisely.

CD99 has originally been described as a human thymus leukemia antigen, an Ewing`s sarcoma specific membrane marker molecule, as a cellular adhesion molecule and a costimulatory molecule for T Lymphocytes. Within this project we aimed at defining putative CD99 ligands since no physiological ligands for the CD99 molecule have been defined when this project was framed. With the help of CD99-Ig fusion proteins we have defined the first ligand for CD99 on hematopoietic cells, i.e. B- lymphocytes. CD99L is functionally relevant since CD99-Ig leads to the co-stimulation of B lymphocytes which have been preactivated with CD40 plus IL-4. The functionally relevant interaction between CD99 and CD99L is dependent on a serum protein with a molecular mass larger than 100 kDa. In detailed epitope mapping studies with CD99 ectodomain deletion variants we assigned more than 20 mAbs to eight CD99 epitopes. Interestingly, we found that so called CD99R mAbs bind to more than one protein domain. Apart from the ongoing functional characterization of the various epitopes, we were able to define an epitope, which is selectively expressed on myeloid precursor cells but not on later differentiation stages. The intracellular signal transducing elements of the CD99 molecule have been characterized with the help of Ig::CD7::CD99 chimeric constructs. With this technology we were able to define two intracellular signaling domains. Ongoing studies shall clarify the relevance of these domains in primary T lymphocytes. In summary, during this project we have gained considerable new knowledge about the biology and structure of an important cell differentiation molecule. It will help us to better understand the biology of certain malignancies (Ewing`s sarcomas, leukemias) as well as lymphocyte communication.