Universität Stuttgart
Permanent URI for this communityhttps://elib.uni-stuttgart.de/handle/11682/1
Browse
1 results
Search Results
Item Open Access Generation and characterization of CD40-Flag-FasL : a novel Fas agonist devoid of systemic toxicity(2005) Assohou-Luty, Constance; Wajant, Harald ( Prof. Dr.)Soluble single chain antibody fusion proteins incorporating the extracellular domains of human TNF, FasL or TRAIL acquire functional properties similar to those of their corresponding membrane-bound forms, following interaction with surface antigens expressed by target cells. In this work, this principle is extended to include other selective protein-protein interactions. Soluble fusion proteins comprising, at the N-terminus, the extracellular domains of human CD40, 4-1BB, TNFR1, TNFR2, RANK and, at the C-terminus, the extracellular domain of human FasL were generated. CD40 was also fused to the extracellular domain of human TRAIL. In all cases the two domains were separated by the Flag sequence to facilitate the detection and purification of the fusion proteins, and all the DNA constructs had a leader sequence to target the resulting proteins to the secretory pathway. The proteins had MWs over the expected sizes, a tendency that was particularly striking for RANKed-Flag-FasL and CD40-Flag-FasL, probably due to post-translational modifications such as glycosylation. The proteins were subsequently analysed for their soluble FasL-like activity on HT1080 cells. Despite only moderate expression or barely detectable expression in the cases of TNFR1-Flag-FasL, 4-1BB-Flag-FasL and TNFR2-Flag-FasL, these proteins showed high specific activities. In sharp contrast to the above mentioned fusion proteins, supernatants of CD40-Flag-FasL and RANKed-Flag-FasL, which contained much more protein, showed very low and low specific activity, respectively. Since the activity of soluble FasL can be increased through multimerization (Schneider et al, 1998), the proteins were cross-linked via their internal Flag-tag. Surprisingly, it led to a decrease in activities for TNFR1-Flag-FasL, TNFR2-Flag-FasL and 4-1BB-Flag-FasL. The activity of RANKed-Flag-FasL could be increased by a factor of 50-100 but the cell death-inducing capacity of CD40-Flag-FasL was virtually unchanged by artificial cross-linking. The high specific activity observed for TNFR1-Flag-FasL, 4-1BB-Flag-FasL and TNFR2-Flag-FasL was thus likely the result of the presence of significant amounts of higher MW aggregates of homotrimers in the supernatants. RANKed-Flag-FasL was probably mostly non-aggregated and was therefore activated upon cross-linking. The lack of any M2 cross-linking antibody-mediated effect in the case of CD40-Flag-FasL does not indicate a defect in the FasL part of the protein but might rather reflect, for example, steric hindrance of the M2 antibody. RANKed-Flag-FasL and CD40-Flag-FasL were thus subsequently further analysed with respect to their capacity for target-dependent activation. CD40-Flag-FasL was stably expressed in HEK293 cells and then affinity purified from supernatants of the stable cells. Western blotting analysis showed that CD40-Flag-FasL migrated with a MW of ~50 kDa under reducing conditions however, analysis of the native protein by gel filtration showed that CD40-Flag-FasL eluted with an apparent MW of 471 kDa, corresponding to a 9.4 mer but this is likely to be an overestimation as CD40-Flag-FasL behaved like a homotrimeric molecule. Binding experiments showed that CD40-Flag-FasL could be immobilized specifically by CD40L-expressing cells and, furthermore, that upon binding it induced clustering of Fas on neighboring cells. CD40-Flag-FasL induced apoptosis and NFkappaB activation in a concentration-dependent manner in transfected HT1080 and KB cells expressing CD40L and not in parental cells, despite similar sensitivities of both parental and transfectant cells to cross-linked FasL. When RANKed-Flag-FasL was analysed in a setting allowing target antigen-mediated immobilization, the protein led to gene induction in a RANKL-dependent manner. The EC50 of CD40-Flag-FasL was shown to be ~ 4 logs lower in HT1080-CD40L compared to HT1080 cells. Cell death induced in CD40L-positive cells was accompanied by cleavage of pro-caspase-8 and activation of caspase-3 and could be completely inhibited by zVAD, Fas-Comp or an anti-human CD40L antibody. CD40-Flag-FasL induced apoptosis in a paracrine manner on by-stander cells and proved to be safe when assayed in vivo in mice for acute toxicity, whereas artificial cross-linking of CD40-Flag-FasL induced liver failure in a concentration-dependent fashion. The promising results obtained with CD40-Flag-FasL led to the cloning of a CD40-Flag-TRAIL construct that was shown to be capable of inducing both apoptosis and NFkappaB in a CD40L-dependent manner. CD40-Flag-TRAIL-mediated apoptosis was shown to be TRAIL-specific, to result from the interaction between CD40/CD40L, and to occur via interaction with TRAIL-R2 in both KB-CD40L and HT1080-CD40L cells. These promising results provide the basis for relatively simple development of safer i.e. less toxic FasL or TRAIL derivatives for the treatment of a variety of human pathologies.