Faculty Research

Research Summary

The eight stranded a/ b barrel protein fold is the largest family of protein structures (representing at least 10% of all known protein structures) and has the widest range of enzyme functions. My research focuses on elucidating the structural basis for folding specificity and thermodynamic stability in this class of enzymes and the future application of those principles in protein design. We are currently investigating "Fold Specific" protein structure adaptation strategies which suggest the pathways that a protein takes to increased stability are specific to the specific structural architecture of its polypeptide chain.

Directed Evolution Approach

We use random PCR mutagenesis to create a library of genetic variants for a variety of enzymes including beta-galactosidase. These variants are placed in an expression host that is a beta-galactosidase negative yet positive for the lactose transport genes. A temperature selection is applied. Using chromogenic substrates for screening, or nutrient restriction for selection, variants that display the selected phenotype (activity or stability at higher or lower temperature) can be selected. These variants are subjected to future rounds of mutagenesis and selection until variants with dramatically different enzymatic properties are produced. Using multiple successes from these “Directed Evolution” experiments, coupled with careful analysis of structure, patterns of mutations begin to emerge. These patterns help us posit generalized, fold-specific, molecular mechanisms of protein structural adaptation to temperature. These principles are then tested using site directed mutagenesis and rational design.

Bioinformatics Approaches

The Panasik lab also looks at the same problem from a different angle through the lens of Bioinformatics. Here, all extant psychrophilic, mesophilic, and thermophilic alpha/beta barrel enzymes that have high resolution crystal structures are annotated and compared. We analyze the distribution of the structural features most commonly purported as the molecular basis of thermostability; ion pairs, hydrogen bonds, solvent accessible hydrophobic surface area buried upon folding, and flexibility.
We find certain ion pairs interactions are enriched in thermophilic alpha/beta barrel proteins. We will compare these patterns observed in the Protein Data Bank (PDB) of all known 3D protein structures to those patterns of adaptation that we see in our directed evolution experiments.

An example of an ion pair interaction that is more commonly found in thermophilic alpha/beta barrels.
Other Folds The above methods are applied to enzymes of other protein fold families to determine structural adaptation strategies to temperature that are specific to those folds.

Entropic Stabilization in Thermostable Enzymes

Computational Biophysical Approaches.

Using ab initio Monte Carlo simulations programmed in PYTHON, the Panasik Lab estimates conformational space available to the unfolded state of a peptide sequence and the conformational space that is disallowed due to steric exclusion of side chain and main chain atoms. We propose novel classes of amino acid substitutions that increase thermostability and have posited a new model of protein thermostability through entropic stabilization

where more non-native-like conformations of the unfolded state are preferentially disallowed due to atom-atom repulsions and steric clash in a thermostable sequence (shown in red) as compared to a homologous mesophilic sequence (shown in blue). More of the conformations ruled out due to steric clash for the thermophilic sequence have higher RMSDs as compared to the native like conformation for that sequence than mesophilic homologous sequences do.

Representations of Conformational Space:

The size of conformational space available to the folded and unfolded states of a protein is not the only informative feature in protein folding and stability. The type of conformational space, i.e. which conformations are available and not available are perhaps far more informative. In order to differentiate between different local conformations of the protein backbone it becomes nescessary to fnd a way to represent the entire range of conformations in a manner that is easy to visualize and to quantitate.


Above is a new 3D representation for multiple conformations of tetrapeptide sequences called 'Dihedral Space'. Each point represents a different tetrapeptide conformation. This 3D representation of tetrapeptide conformations is similar to the 2D Ramachandran plot for dipeptides. This figure is a sample of 10% of all tetrapeptide conformations in the PDB.

Tetra peptide conformations found in a 60 x 60 x 60 degree cube of dihedral space called a "Hexostate" indicates that similar regions in dihedral space correspond to similar conformations. Graphic courtesy of Prof. N. Panasik, Jr., Claflin University. Conformations that result when phi, psi angles from dissallowed regions of the Ramachandran Plot cluster and inhabit regions of dihedral space that are sparcely populated in the PDB. Graphic courtesy of Prof. N. Panasik, Jr., Claflin University.

Dihedral Space can be divided into 60x60x60 cubes representing different conformations and the distribution of conformations present in all proteins in the PDB can be analyzed. Graphic courtesy of Prof. N. Panasik, Jr., Claflin University.

Recent grant awards (http://www.scepscoridea.org/Cyberinfrastructure/index.html ) have increased Claflin’s bandwidth and allow for computational time on the Palmetto High Performance Computing Cluster (http://citi.clemson.edu/condoprogram ) and the Kraken (http://www.nics.tennessee.edu/computing-resources/kraken ) at Oak Ridge National Labs.
Also, a new NSF Early CAREER Award
 (http://nsf.gov/awardsearch/showAward.do?AwardNumber=1054787&WT.z_pims_id=12771 ) has expanded research of alpha/beta barrel enzymes into bio-fuel related enzymes. Here, bio-fuel related enzymes such as cellulases and glucanases are adapted through a range of temperatures while further elucidating the patterns of temperature adaptation that alpha/beta barrel enzymes exhibit. Computational approaches are taken to identify possible mutations as well as provide a biophysical mechanism of action for some mutations identified through Directed Evolution approaches.