Current
Projects
 |
|
Redesign of a fasciculin toxin from the venom of Green Mamba
snake
Fasciculin
(FAS) is a snake toxin that is a very powerful inhibitor of
acetylcholinesterase (AChE), an enzyme that terminates impulse transmission
at cholinergic synapses. FAS binds with very high pM affinity to
AChE from Humans (hAChE) and similar species.
Hence the Fas-AChE complexes are excellent systems for computational
studies of protein-protein interactions. We have designed a number
of mutants of Fas with improved affinity towards AChE from topedo
Californica (tAChE). To
verify our computational results, we developed an efficient
procedure for expression of Fas in E. Coli. Such a procedure
has previously failed in the hands of several labs. Computationally designed
Fas mutants were tested experimentally and were found to enhance
binding affinity of Fas to tAChE. In addition, the Fas
mutants optimized for binding to tAChE considerably decreased
affinity to hAChE.
Interestingly,
Fas is not potent against AChE from Drosophila (dAChE). The hAChE and dAChE are very similar both
at the sequence and the structure level. However, the FAS binding
pocket of dAChE has
evolved to loose its geometric complimentarity to FAS. Our goal is
to redesign FAS to reverse its binding specificity, inducing its
binding to dAChE and
abolishing its binding to hAChE. For this purpose, we
first created a structural model of the dAChE-FAS complex and then
redesigned FAS for best interactions with dAChE. Our computational
results are currently being tested in the laboratory by expressing
the FAS mutants and assaying them for binding to both hAChE and dAChE. This project is in
collaboration with the Structural Proteomics center in Weizmann and
Prof. Joel Sussman.
Top
|
 |
|
Synthetic mimicry of protein binding sites through
structure-based design
Molecules capable of
mimicking protein binding and/or functional sites present useful tools for a range of biomedical
applications, including the inhibition of protein-ligand
interactions. Mimetics of large and sequentially
discontinuous protein binding sites can presently be generated
through structure-based design and chemical synthesis. Computational protein design could further
enhance the affinity and specificity of such molecules for their
targets. The interaction of the synaptic enzyme acetylcholinesterase
(AChE) with its inhibitor fasciculin-2 (FAS) serve as a model for
this study, since the structure and function of this complex is well
understood. Assembled peptides mimicking the
discontinuous binding site of AChE for FAS will be synthesized,
functionally evaluated, and computationally optimized regarding
their affinities to FAS. The results are expected to further enhance the scope and
quality of structure-based and computational strategies for the
design of inhibitors of molecular interactions involving large
protein-protein interfaces. This project is in collaboration with
Prof. Jutta Eichler from University of Erlangen.
Top
|
 |
|
Development of computational methods for protein
interface design and multi-state protein
design
Computational design of protein-protein interfaces
could become an easy and inexpensive method of creating
protein-based inhibitors for a particular disease-associated
pathway. Such inhibitors should possess high affinity to a selected
target as well as high binding specificity for the same target.
However, presently available computational methods often fail to
supply the designed proteins with these important characteristics.
Hence, we are working on improving the computational protocols for
design of protein-protein interfaces. 1) To develop a better energy
function for design of protein-protein interfaces, we are optimizing
the relative weights of each energy term using a database of
protein-protein complexes with known structures. 2) To guarantee the high
binding specificity of our designed proteins, we are developing
algorithms that incorporate many protein “states” or structures into
our design procedure, allowing to select sequences that are optimal
for one particular state and are suboptimal for alternative
states.
Top
|
 |
|
Redesign of Prion Proteins
A number of neurodegenerative
diseases have been recently identified to share a common mechanism,
which involves a conversion of some soluble protein into its
pathological insoluble conformer. Among the best-studied protein
misfolding diseases are the transmissible spongiform
encephalopathies (TSEs), in which a conversion of a prion protein
plays a major role. It is known that prions can exist in monomeric,
dimeric, and higher-order aggregate conformations. The dimeric prion
conformation has been proposed to be an intermediate on the pathway
to aggregation. We would like to test this hypothesis by designing
mutations into the prion sequence that induce formation of prion
heterodimers. Our mutant prions should recruit the wild type prions
into forming dimers and hence, inhibit the aggregation process. This
project is in collaboration with Albert Taraboulos from the Hadassah
Medical School in Hebrew University.
Top
|
 |
|
Evolution of promiscuous proteins
In nature, some proteins are
more social than others, interacting with a large number of
partners. These “promiscuous” proteins play key roles in cellular
signaling pathways whose disruption may lead to diseases such as
cancer. The amino acid sequences of such proteins must have evolved
to be optimal for combined interactions with all natural partners.
However, the evolutionary process leading to this promiscuity is not
fully understood. We address this subject by predicting amino acid
sequences that would be most compatible for interaction with each
partner on its own and those most compatible for binding multiple
proteins. We find that these two types of sequences are
substantially different, the latter more closely resembling the
natural sequences of promiscuous proteins. We also find that
promiscuous proteins contain certain regions that are necessary for
interfacing with all of their partners, while other regions convey
specific interactions with each particular target protein. We
analyze the tradeoffs required for such proteins to bind multiple
partners and find that only some degree of compromise is typically
needed in order to permit interactions that are seemingly
antagonistic. We conclude that the simulations reported here mimic
well the natural evolution of proteins that associate with multiple
partners.
Top
|
 |
|
Reversing the Ras signaling: design of effectors
that preferentially bind to the inactive Ras
state
Proteins in the Ras family are very important
molecular switches for a wide variety of signaling pathways that
control cytoskeletal integrity, proliferation, cell adhesion,
apoptosis, and others. These proteins are small GTPases that cycle
between the GTP- and the GDP-bound states. While the GTP- and the
GDP-bound Ras states are structurally very similar, only in the
GTP-bound state, Ras is able to bind tightly to its effector
proteins and activate downstream signaling pathways. We set our goal
to understand how relatively small conformational changes in Ras
could control its affinity to effector proteins. By designing a
mutant of the Ras effector Raf kinase with a 100-fold improved
affinity for the Ras-GDP state, we were able to solve the first
crystal structure of Ras effector bound to the natively inactive Ras
state. Interestingly, the structure demonstrates that Ras is found
in conformation similar to that of Ras-GTP and not Ras-GDP. This
project is in collaboration with Prof. Christian Herrmann from Ruhr
University Bochum.
Top
|
 |
|
Designing novel binders for Ras
Protein engineering approaches have been used to
convert various scaffold proteins into antibody-like molecules that
bind a particular target protein. Such novel binding domains could
find useful applications in various fields including drug and
biomarker design and in
vivo imaging. Computational methods could greatly facilitate the
discovery of such novel binding domains, reducing the time and the
cost of the process and at the same time, expanding the number of
available scaffold proteins. We developed a general approach for
computational design of novel binding partners for
structurally-characterized targets. This approach is being tested by
designing novel binding partners for Ras – a small GTP-binding protein that
is an essential molecular switch for a wide variety of
signaling pathways. Computationally designed small libraries of
novel Ras binders are now constructed experimentally and screened
for binding to Ras using ELISA. In the preliminary experiments, we
have identified several positive hits of Ras novel binders that are
presently being characterized using biophysical and structural
methods. This project is in collaboration with the Israel Structural
Proteomics center and with Prof. Christian Herrmann from Ruhr
University Bochum.
Top
| |
