PhD projects


Structural Biology of Membrane Proteins

PhD project:

Structural and functional characterization of γ-secretase, a core protein complex responsible for Alzheimer’s Disease. Theoretical modeling and NMR approach.

Principal Investigator:

S. Filipek, International Institute of Molecular and Cell Biology, and Warsaw University, Warsaw, Poland; sfilipek|

In collaboration with:

V. Dötsch / F. Bernhard, Centre for Biomolecular Magnetic Resonance, Institute of Biophysical Chemistry, University of Frankfurt; vdoetsch|, fbern|

Alzheimer’s disease (AD), the most common form of dementia in the elderly, is characterized by accumulation of the β-amyloid (Aβ senile plaques in the brain. Aβb is produced by a cascade of proteolytic cleavages of amyloid precursor protein (APP) and the last cut is performed by γ-secretase. The failure of Aβ clearance, mutations in APP or in γ-secretase lead to the pathological aggregation of this peptide1, 2.

γ-Secretase is a complex composed of four different integral membrane proteins: presenilin (PS), nicastrin, APH-1, and PEN-2. The most studied component of the γ-secretase complex is presenilin, which is an integral enzyme in the cleavage of amyloid precursor protein and contributes to the accumulation of Aβ peptide in Alzheimer’s disease. Activation of PS is dependent on its endoproteolysis into an N-terminal fragment (NTF) and C-terminal fragment (CTF). Nicastrin has recently been described as 'the gatekeeper of the γ-secretase complex'. Of the γ-secretase components, only nicastrin is a single-spanning membrane protein containing a massive extramembranous domain; the other components span the bilayer several times and are predominantly buried within the membrane. Thus, the γ-secretase complex residing in the bilayer has relatively small extramembranous domains.

Presenilins (PS-1 and PS-2) provide two aspartic acid residues located within the membrane which form a catalytic core for the intramembranous proteolysis of substrates. Genetic studies show that more than 170 mutations in the PS-1 protein (PSEN1 gene) were associated with familial forms of Alzheimer’s disease (FAD) and new mutations are still being identified3. These pathogenic mutations lead to an overproduction of highly aggregative forms of Aβ within the brain. A vast majority of PS-1 mutations occur within the transmembrane regions indicating that even minute changes in the structure of these regions may radically change properties of the protein including its catalytic characteristics. Residues associated with these pathological mutations appear to form vertical patterns along the helices when mapped on regular α-helices.

The group of S. Filipek performed an analysis of mutation patterns in all ten hydrophobic regions (HRs) of PS-1 and PS-2 using database of AD mutations3. The linear patterns reported previously4 were confirmed and extended to areas spanning as many as three faces of a given HR. The complementary areas of residues free of AD mutations were identified based on the location of non-pathogenic polymorphisms and PS-1 vs. PS-2 amino acid discordances. Taking into account the location of areas of AD mutations and mutation-free areas/regions a preliminary model of PS-1 structure was proposed using a general stick-out-mutation rule. Two molecular models were built differing in the location of CTF helices. The models properly distinguish residues belonging to AD-affected sites and non-pathogenic areas and may be used for classification purposes5.

The occurrence of helical patterns of pathogenic substitutions observed in most HRs of presenilins suggests that these particular regions of putative transmembrane helices are of vital importance for the proper activity or binding properties of γ-secretase. Linear patterns formed by the mutation clusters along HRs seem to be most likely involved in the inter-helical packing. Three levels of this packing are considered: 1) helix-helix interactions within the PS molecule, 2) interhelical interactions with other components of γ-secretase and 3) proper spatial fitting of a substrate to the binding and/or cleaving site. As postulated by the knob-into-holes packing mode of helices, small residues seem to be particularly involved in such a mechanism.

To investigate this mechanism we created a model of PS-1 – APH-1 interface based on (small)xxx(small)xxx(small) motif which involves two transmembrane helices from both proteins. This interface is based on a highly conserved GxxxGxxxG motif in the APH-1 protein. It can form a tight contact with a small residue AxxxAxxxG motif in presenilin. We have built and verified four modes of binding based on similar structures involving GxxxG motifs in glycophorin and aquaporin. The resulting best model employs antiparallel orientations of interacting helices and is in agreement with currently accepted topology of both modeled proteins6.

- Modeling of γ-secretase complex including development of new algorithms for helix-helix packing (Filipek's group, 24 months)

Because all transmembrane parts of the proteins in the γ-secretase complex are composed of α-helices, the modeling of structure of the complex requires elucidation of helix-helix interactions in the membrane environment. Such helix-helix packing will involve interactions within single proteins as well as protein-protein interactions. These both interaction types are important for bundling of helices to produce a delicate balance between all interacting proteins in the γ-secretase complex. Alzheimer’s Disease mutations shift this balance toward overproduction of Aβ 42 form which is highly aggregative. The structure of the membranous part of the complex will be investigated using theoretical methods for wild type and mutated forms. Determined structural parts of the complex will be simulated using Molecular Dynamics in water-membrane systems to elucidate helix-helix contacts and checking stability of investigated system. Simulation of processing of Aβ will be investigated using coarse grain molecular dynamics.

- Structural characterization of the γ-secretase complex by liquid state NMR spectroscopy (V. Dötsch / F. Bernhard, 12 months)

At the Institute of Biophysical Chemistry in Frankfurt, we have established protocols for the production of preparative amounts of the individual components of the γ-secretase complex. The proteins can be produced in mg amounts in less than 2 days and solubilized into suitable detergents or other hydrophobic environments. Moreover, the proteins can efficiently be labeled with stable isotopes and analysed by NMR spectroscopy. Structural and functional details of the proteins can be studied by a variety of newly designed labeling techniques. The structural topology of a core part of the γ-secretase complex, the PS-1 CTF fragment, could already be elucidated and all amino acid residues have been assigned in corresponding NMR spectra. Besides predicted structural features like three transmembrane segments, a new helical component localized in a long loop of the protein has been identified. The experimentally defined topology of CTF will be used in structural modeling and Molecular Dynamics studies. Refined structural models will help to determine specific contacts to other components of the γ-secretase complex and to predict protein interfaces. Proposed complex formations between the individual γ-secretase subunits and the interaction of distinct transmembrane segments will be evaluated in Frankfurt by experimental approaches. Protein interactions will be studied by NMR titration assays on the molecular level. In complementary approaches, a variety of biochemical assays will be implemented in order to analyse the predicted complexes in vitro. Identified residues critical for protein interactions will be modified by directed mutagenesis approaches and structural and functional features of the mutated derivatives will be analysed in Frankfurt. The project will deliver a comprehensive view on the topology and assembly of the γ-secretase complex determined by experimental techniques as well as by theoretical simulations.


  1. A. L. Brunkan, A. M. Goate, (2005).
    Presenilin function and gamma-secretase activity. J. Neurochem. 93, 769-792.
  2. M. P. Mattson, (2004).
    Pathways towards and away from Alzheimer's disease. Nature 430, 631-639.
  3. Alzheimer Disease & Frontotemporal Dementia Mutation Database
  4. Hardy J. and Crook R. (2001). Presenilin mutations line up along transmembrane alpha-helices. Neurosci. Lett. 306, 203-205.
  5. K. Jozwiak, C. Zekanowski, S. Filipek (2006).
    Linear patterns of Alzheimer’s disease mutations along α-helices of presenilins as a tool for PS-1 model construction, J. Neurochem. 98, 1560-1572.
  6. K. Jozwiak, K.A. Krzysko, L. Bojarski, M. Gacia, S. Filipek (2008).
    Molecular models of the interface between APH-1 and presenilin involving GxxxG motifs, ChemMedChem 3, 627-634.
  7. Reckel, S., Sobhanifar, S., Schneider, B., Junge, F., Schwarz, D., Durst, F., Löhr, F., Güntert, P., Bernhard, F. and Dötsch, V. (2008).
    Transmembrane segment enhanced labeling as a tool for the backbone assignment of a-helical membrane proteins. Proc. Natl. Acad. Sci. USA, 105, 8262-7.
  8. Schwarz, D., Junge, F., Durst, F., Frölich, N., Schneider, B., Reckel, S., Sobhanifar, S., Dötsch, V., and Bernhard, F. (2007).
    Preparative scale expression of membrane proteins in E. coli based continuous exchange cell-free systems. Nat. Protocols, 2, 2945-57.
  9. Trbovic, N., Klammt, C., Koglin, A., Löhr, F., Bernhard, F., and Dötsch, V. (2005).
    Efficient strategy for the rapid backbone assignment of membrane proteins. J. Am. Chem. Soc., 127, 13504-5.
  10. Koglin, A., Klammt, C., Trbovic, N., Schwarz, D., Schneider, B., Schäfer, B., Löhr, F., Bernhard, F., Dötsch, V. (2006).
    Combination of cell-free expression and NMR spectroscopy as a new approach for structural investigation of membrane proteins. Magn. Reson., 44, 17-23.