University of North Dakota Home
Office Name
'
A to Z Index'Directory'Map
Chemistry department
My UND Blackboard
'
 
Click here to go back to the Faculty page

Dr. Kathryn A. Thomasson

Professor and Director of Graduate Studies

Gold Line that Seperates Text

 

Research in my laboratory incorporates large-scale computer simulations with classical physics, chemistry and biology. Current developments include expanding programs to treat large biological molecules (e.g., DNA and F-actin) and exploring quantum mechanical methods for treating small biological molecules such as short peptides. Specific projects in the group include: (1) glycolytic enzyme interactions with cytoskeletal structures; (2) genome regulatory proteins interacting with DNA, and (3) calculating the circular dichroic spectra of small peptides.

(1) The cytoskeletal structure, F-actin creates a structure in the cell cytoplasm upon which glycolytic enzymes associate and dissociate dynamically, and these dynamic associations are important to organizing the reactions in the glycolytic pathway. Many glycolytic enzymes including fructose-1,6-bisphophsate aldolase (aldolase), glycerladehyde-3-phosphate dehydrogenase (GAPDH), and lactate dehydrogenase (LDH) bind actin reversibly. Other enzymes such as triose phosphate isomerase (TIM) bind indirectly through interactions with the enzymes that bind. We use and enhance existing theoretical methods to better investigate the interactions of F-actin with glycolytic enzymes. The method of Brownian dynamics (BD) predicts the various kinds of complexes between F-actin and the glycolytic enzymes, provides the relative stability of complexes, reproduces experimentally observed factors on binding such as ionic strength, pH, and relative binding affinities, and make predictions about the impact of mutations on the proteins.

 

Figure 1, Research - Dr. Thomasson

 

Complex of aldolase (blue) with a subunit of the cytoskeletal structure F-actin (red). Two subunits of aldolase bind a single subunit of F-actin by forming salt brigdes from lysines on the surface of aldolase to aspartates and glutamates in subdomain 1 of F-actin.

 

(2) Proteins control the copying of DNA in cells. We use BD simulations of protein/DNA interactions to better understand how proteins control DNA transcription. We can predict the how the protein interacts with DNA and how long they associate. We study how these predictions are affected by changes in the composition of the solution, protein or DNA sequence. These are pioneering studies in the dynamics of protein/DNA interactions.

 

Figure 2, Research - Dr. Thomasson

 

Dynamics of an encounter of 434 Cro repressor searching the surface of DNA for its operator. The first 555 steps of a single BD trajectory beginning with the Cro center of mass (+ symbols) 60 Å away from the DNA helix axis. A few representative steps are labeled by their number in the trajectory. Note the predominant motion is towards the left indicating the Cro is sliding along the DNA.

 

 

(3) One of the most serious unsolved problems in computational biology today is the lack of ability to predict the three-dimensional structure of a protein from its primary sequence. The lack of progress in this area is now the scientific bottleneck for understanding proteins and their function, particularly concerning the solution of the human and other genomes and high throughput proteomics. It is critical to provide a computationally fast effective method to predict the structure of a protein as its primary sequence is discovered. One quick way to experimentally determine a solution structure is the circular dichroism (CD) spectrum. We use classical electromagnetic theory (the dipole interaction model) to predict CD for peptides and proteins that can provide an excellent test of proposed protein models when the CD predictions agree with experiment. In collaboration with Dr. Mark Hoffmann, we are developing a classical/quantum mechanical approach to generating the peptide models used to predict CD.

 

Figure 3, Research - Dr. Thomasson

 

CD spectra predicted for cyclo(L-Pro)3. The CD predictions for structures obtained using ab initio 3-21G basis set optimizations with weighting factors created by AM1. The line curves are dipole interaction model predictions using different empirical parameter sets. The ° represent the experimental spectrum. The bandwidths are 4000 cm-1. The units for De are L mol-1 cm-1.

Gold Line that Seperates Text
REPRESENTATIVE PUBLICATIONS

"Dipole Interaction Model Predicted p-p* Circular Dichroism of Cyclo(L-Pro)3 Using Structures Created by Semi-empirical, Ab Initio, and Molecular Mechanics Methods.". S. L. Lowe, K. S. Pierce, J. Czlapinski, G. Kie-Adams, Rajeev Pandey, M. R. Hoffmann, K. A. Thomasson.. J. Peptide Research, 2003, 61, 189-201.


"Brownian Dynamics Simulations of Aldolase Binding GAPDH and the Possibility of Substrate Channeling." I. Ouporov, H. Knull, A. Huber K. Thomasson. Biophys. J., 2001, 18, 2527-2535.


"Brownian Dynamics Simulating the Ionic Strength Dependence of the Nonspecific Association of 434 Cro Repressor Binding B-DNA." F. Yang, I. V. Ouporov, C. Fernandes, D. Motriuk, K. A. Thomasson. J. Phys. Chem. 2001, 105,12601-12608.


"Brownian Dynamics Simulations of the Specific Interactions Between Rabbit Aldolase and G- or F-Actin." I. Ourporov, H. R. Knull, K. A. Thomasson. Biophys. J., 1999, 76, 17-27.

 

 

 

Chemistry Department
Abbott Hall Room 236
151 Cornell Street Stop 9024
Phone: 701-777-2741
Fax: 701-777-2331
Email: info@chem.und.edu