Structur-Function Analysis of ADAM17
By studying the ADAM17 ectodomain, we were able to identify and characterise an important functional region consisting of a membrane-proximal domain (MPD) and a directly C-terminal adjacent, highly conserved amphipathic helix called CANDIS (Conserved Adam seventeeN Dynamic Interaction Sequence) [1], [2].
We were able to solve the structure of the MPD by NMR and found that this domain can exist in two disulfide isomers [1]. One of these disulfide isomers has a very flexible, open conformation, while the other has a closed, very rigid conformation. This open conformation keeps ADAM17 in an active state. We were able to show that this conformation enables ADAM17 to interact with phosphatidylserine on the outer layer of the cell membrane, which leads to a conformational change of the entire ADAM17 ectodomain and triggers the actual proteolytic shedding process [1], [3], [4]. Phosphatidylserine is only presented at the cell surface when the maintenance of cell membrane asymmetry is disturbed. This can be triggered by drastic processes such as apoptosis. But also the activation of scramblases by various signalling pathways leads to a transient phosphatidylserine presentation. In this context, we were also able to show that the amphipathic helix CANDIS binds more effectively to the cell membrane when phosphatidylserine is presented at the surface [3]. Furthermore, we have shown that CANDIS is directly required for substrate recognition and binding [2]. When the MPD is converted from its open, flexible conformation to the closed conformation, e.g. by protein disulphide isomerases, ADAM17 loses the ability to interact with phosphatidylserine and bind its substrates via CANDIS [2], [3], [4]. ADAM17 is thus fixed in an inactive conformation.
Phosphatidylserine-dependent membrane binding of active ADAM17 is presumably accompanied by a conformational change of the entire ADAM17 ectodomain. This presumably brings the catalytic domain into close proximity to the membrane. This model is corroborated by the fact that the cleavage sites in ADAM17 substrates are positioned in the stalk regions close to the membrane. We have further corroborated this model by showing that shedding by ADAM17 depends on the position of the interface in the stalk region of the substrates. The further the cleavage site is positioned from the membrane, the less ADAM17-dependent shedding occurs [5].
Understanding these molecular relationships opens up new possibilities for inhibiting the activity of ADAM17. For example, the use of protein disulphide isomerases or redox reagents could favour the inactive conformational state of ADAM17. Another possibility is to develop antibodies against CANDIS to interfere with binding to ADAM17 substrates. An important question that is still open is to what extent iRhoms might play a role in this conformational change.
Reference
[1] Düsterhöft, S. et al. (2013). "Membrane-proximal domain of a disintegrin and metalloprotease-17 represents the putative molecular switch of its shedding activity operated by protein-disulfide isomerase." J Am Chem Soc 135(15): 5776-5781. DOI: 10.1021/ja400340u
[2] Düsterhöft, S. et al. (2014). "A disintegrin and metalloprotease 17 dynamic interaction sequence, the sweet tooth for the human interleukin 6 receptor." J Biol Chem 289(23): 16336-16348. DOI: 10.1074/jbc.M114.557322
[3] Düsterhöft, S. et al. (2015). "Extracellular Juxtamembrane Segment of ADAM17 Interacts with Membranes and Is Essential for Its Shedding Activity." Biochemistry 54(38): 5791-5801. DOI: 10.1021/acs.biochem.5b00497
[4] Sommer, A. et al. (2016). "Phosphatidylserine exposure is required for ADAM17 sheddase function." Nat Commun 7: 11523. DOI: 10.1038/ncomms11523
[5] Düsterhöft, S. et al. (2020). "Distance dependent shedding of IL-6R." Biochem Biophys Res Commun 526(2): 355-360. DOI: 10.1016/j.bbrc.2020.03.093
