Tetraspanins are a superfamily of membrane proteins which span the membrane four times; they are found predominantly at the cell surface but are also located on intracellular vesicles. Tetraspanins (with a few exceptions) do not have conventional receptor ligand functionality and instead form lateral associations with other molecules within the membrane. Binding partners include, but are not limited to, MHC proteins, integrins, signalling proteins and other members of the tetraspanin superfamily. This large network of interactions has led to the idea of tetraspanin enriched microdomains (TEMs) or the tetraspanin web, in which tetraspanins function by bringing together proteins to form functional clusters which allow processes (such as signal transduction or adhesion) to take place more efficiently. Due to the numerous diverse binding partners, tetraspanins have been implicated in a number of cellular and pathological processes.
Despite tetraspanins being involved in fundamental physiological processes, relatively little is known about the function of individual members. Difficulties arise as only a few monoclonal antibodies are available to the native proteins and mouse knock outs often show only a mild phenotypic change. Our group and others have used recombinant human EC2 domains in the form of GST fusion proteins to assess tetraspanin involvement in several processes. This region is thought to attribute specificity to individual tetraspanin members and when recombinant versions are added to cells exogenously, they have been shown to modulate different cellular events including adhesion, migration and fusion.
Due to a number of inherent drawbacks associated with bacterially expressed recombinant EC2 domains (such as LPS contamination and inferior folding), the initial aim of this work was to express the recombinant proteins in a mammalian host. Despite multiple attempts to express the proteins in mammalian or insect cells using different vector systems, this was not successful, although it was demonstrated that DNA was integrated into the host genome and that EC2 encoding mRNA was expressed. Following this, it was decided to focus on bacterial expression and use the EC2 domains generated to further our understanding of tetraspanin involvement in IgE mediated degranulation this is a critical first step in Type I Hypersensitivity and although several tetraspanin family members have been implicated in this pathway, their exact involvement is not yet clear. Here, recombinant EC2 domains of tetraspanins CD9, CD63, CD81 and CD151 were used for the first time in conjunction with RBL-2H3 cells (a cell line commonly used as a model for mast cell degranulation) to examine tetraspanin involvement in IgE mediated degranulation. These particular tetraspanins were selected because past studies
iii
have implicated these members in mast cell activation, but dispite an anti-CD63 antibody being able to down regulate degranulation, in this instance the EC2 domains did not exhibit any modulating effect on this form of degranulation in RBL-2H3 cells. The activity of these particular recombinant proteins was demonstrated in two other functional assays; bacterial adhesion to endothelial cells and bacterial induced giant cell formation.
Later work sought to characterise the EC2 proteins in terms of their secondary structure, LPS content and their ability to bind to cells, with the hope of elucidating their mechanism of action. The binding of each EC2 domains to two cell lines was examined: RBL-2H3 where EC2 domains show no effect on degranulation, and HEC-1B cells, where EC2 domains were previously shown exhibit biological activity by reducing bacterial adhesion. At highest concentrations utilised, the EC2 domains were shown to bind to both cell lines significantly more than the GST control protein, but attempts to examine the specificity of the EC2 interaction with the cells by competitive inhibition gave inconclusive results. Whilst this may have been due to technical issues it is tempting to speculate that this indicates cellular interactions that do not follow conventional binding mechanisms. The LPS content of the EC2 domains was shown not to correlate with their ability to modulate bacterial-induced cell fusion. Finally, to facilitate structural and future functional studies, attempts were made to optimise the removal of the GST tag from the fusion proteins. CD spectroscopy was then performed and attributed the EC2 domains of CD9 and CD81 with 50% and 52% α-helical structure, respectively, as expected for these proteins.
Although initial aims of producing mammalian EC2 domains were not fulfilled, they were successfully produced in bacteria and used in RBL degranulation assays. later work indicated that LPS contamination was not the causative component of EC2 preparations and confirmed some level of secondary structure. Furthermore, the apparent lack of effect of recombinant EC2 domains in RBL-2H3 activation may shed light on tetraspanin interaction with the high affinity IgE receptor.
