The project started with the question if and to what extent the nuclear envelope is a target of or a host for cellular signalling. Signalling molecules on their way into the nucleus have to pass the nuclear envelope at the nuclear pores, where the outer nuclear membrane (ONM), part of the rough endoplasmic reticulum, folds into the inner nuclear membrane (INM). Large protein complexes, the nuclear pore complexes (NPC), control the transport of macromolecules between nucleus and cytoplasm. The inner nuclear membrane is tightly bound to the intermediate filaments of the lamina (LAM) via their integral INM-proteins binding to the lamins. Here, signalling processes controlling nuclear structure and regulation of gene expression should be established.

At that time, not very much was known about the most interesting substructure related to the question of signalling at the nuclear envelope, the inner nuclear membrane, since neither its protein composition nor protein interactions at the INM were sufficiently characterized. Also, no specific residents of the outer nuclear membrane were known as this membrane was considered to be rough endoplasmic reticulum. We had therefore to first address the protein composition of the nuclear envelope, and in particular of the INM.

For practical reasons, we focussed on nuclear membrane proteins, which, in a formal way, can be sorted in three categories by their localization: specific ONM-proteins could for example anchor and position the cell nucleus inside a cell. A second category would involve the pore membrane proteins that anchor the nuclear pore complexes at the nuclear pores. The third category comprises proteins of the inner nuclear membrane, which in most cases bind to the lamins of the lamina. Besides these proteins that are confined to their compartments by specific interactions, their interactors should also be found in a proteomics screen. For pore membrane proteins, this are the NPC proteins and components of nucleocytoplasmic transport; the INM proteins should bind components of the chromatin (proteins and nucleic acids) and regulators of gene expression.

To this end, we developed and refined purification and subfractionation techniques for the nuclear envelope and necessary protein chemical and mass spectrometric procedures using the integral INM protein LAP-2beta as a model protein. We started a first thorough analysis of the nuclear-envelope protein composition, identifying about 150 proteins from defined nuclear-envelope subfractions. 19 of these proteins were either not very well characterized or completely unknown as nuclear-envelope constituents. To validate new nuclear-envelope proteins that putatively are integral membrane proteins, we cloned cDNAs of such candidates into expression vectors, adding sequences coding for antibody epitopes. We transiently expressed these proteins in fibroblasts and analysed their subcellular localization. As a result, we verified three new integral INM proteins, LUMA, SUN-1 and KIAA 0668 / SUN-2, and detected two LAP-2 isoforms for the first time at the protein level (Dreger et al. (2001)).

While we proceeded to characterize LUMA, the 70-kDa protein SUN-2 was further characterized in collaboration with Dr. Didier Hodzic (Washington University, St. Louis), who coordinated this study (Hodzic et al. (2004).

----------------------------------------------------------

Henning Otto / updated: 12.04.2011