Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) and Cystic Fibrosis (CF)

The cystic fibrosis transmembrane conductance regulator (CFTR) is an ATP-binding cassette (ABC) ion channel that regulates salt secretion and reabsorption in epithelial cells. Mutations in the cftr gene cause cystic fibrosis (CF). In 90% of CF patients, the disease-associated mutation is the deletion of phenylalanine 508 (Phe508), which is hypothesized to cause protein misassembly and misfolding. To understand the molecular defect of the CFTR misassembly due to Phe508 deletion, I used computational structural biology tools to develop a theoretical structural model of the CFTR channel. The model predicted that the mutation perturbs a crucial interface between the cytoplasmic and membrane-spanning domains [1,2]. The prediction has been validated extensively in collaboration with Dr. John R. Riordan (UNC-CH Department of Biochemistry and Biophysics). In a subsequent study, we further experimentally validated all of the predicted cytoplasmic-membrane domain interfaces and showed that these interaction sites mediate the regulation of the channel gating [3].  I am currently using rational protein design to introduce rescuing mutations in the interface that is perturbed by Phe508 deletion, and thus prevent the domain misassembly of the CFTR mutant.

In addition to domain misassembly, Phe508 deletion likewise induces aberrant folding of the first nucleotide-binding domain (NBD1). To investigate the molecular origin of this folding defect, I, in collaboration with Dr. Tamas Hegedus of Dr. Riordan’s laboratory, performed molecular dynamics simulations of wild type and mutant NBD1s. We showed that specific loops in the protein can modulate the folding kinetics of the NBD1 domain [4]. We are currently using rational protein design to stabilize these loops and correct the misfolding due to the Phe508 deletion.

CFTR. (A) CFTR is an ion channel in the apical membrane of epithelial cells composed of nucleotide-binding domains (NBD) and membrane-spanning domains (MSD). Deletion of the Phenylalanine 508 (Phe508) is associated to ~90% of cystic fibrosis patients.  (B) Structural model of CFTR identifying the interface perturbed by the Phe508 deletion. Figure adapted from reference [1].

Collaborators: Drs. John Riordan, Tamas Hegedus, Andrei Aleksandrov, Lihua He, and Liying Cui. Jack Riordan and his lab are all from the UNC Biophysics and Biochemsitry Department and the Cystic Fibrosis Center.

References:
[1] AWR Serohijos, T Hegedus, AA Aleksandrov, L He, L Cui, NV Dokholyan, JR Riordan. PNAS 105: 3256-3261 (2008)
[2] T Hegedus, AWR Serohijos, NV Dokholyan, L He, JR Riordan. J Mol Biol 378: 1052-1063 (2008)

[3] L He, AA Aleksandrov, AWR Serohijos, T Hegedus, LA Aleksandrov, L Cui, NV Dokholyan, JR Riordanl. J Biol Chem 283: 26383-26390 (2008)

[4] AWR Serohijos*, T Hegedus*, JR Riordan, NV Dokholyan. PLOS Comp Biol 4: e1000008 (2008) [*Equal contribution]

Structure and Mechanism of the Molecular Motor Dynein

Dynein is a cytoskeletal motor protein that drives the beating of cilia and flagella and the minus-directed transport of molecules and organelles along the microtubule.  In collaboration with Dr. Timothy C. Elston (UNC-CH Department of Pharmacology), I used computational structural biology to construct a complete structural model of dynein’s motor unit [5], an approximately 0.3 MDa domain that hydrolysis ATP and generates force. Employing normal mode analysis, I also studied the conformational dynamics within the motor unit that are potentially associated with dynein’s force generation [5]. More recently, I used simplified protein models to further investigate the conformational dynamics that may accompany the energy transduction of the motor unit [6]. Another major effort of my work is understanding how two identical heads of dynein are coordinated as the dimer walks along the microtubule. Working in close collaboration with Dr. Denis Tsygankov from Dr. Elston’s laboratory, we constructed a mathematical kinetic model of the coordination that is not only consistent with current experimental data, but also determined the necessary requirements for coordination between the two heads [7].

Dynein Motor. (A) Schema of a dynein dimmer. (B) Theoretical model of the motor unit, the site for ATP hydrolysis and force generation. Figure adapted from reference [5].

Collaborators: Drs. Timothy Elston and Denis Tsygankov.(UNC-CH Department of Pharmacology)

References:
[5] AWR Serohijos, Y Chen, F Ding, TC Elston, NV Dokholyan. PNAS 103: 18540-18545 (2006)

[6] AWR Serohijos, D Tsygankov, S Liu, TC Elston, NV Dokholyan. Submitted.
[7] D Tsygankov, AWR Serohijos, NV Dokholyan, TC Elston. J Chem Phys 420: 25011 (2009)

Design of a Cancer Biomarker Probe

Breast cancer is one of the most frequently diagnosed malignancies in American females and is the second leading cause of cancer deaths in women. Early detection of the breast cancer dramatically increases the survival rate of patients. Cysteine-Rich Intestinal Protein 1 (CRIP1) has been identified as a novel marker for early detection of breast cancers. In collaboration with Dr. James P. Basilion, we developed a high-affinity ligand CRIP1 probe, consisting of a ligand peptide fused to a fluorescent molecule [8]. Dr. Basilion’s lab used phage display to identify ligand peptides of modest binding affinity (~36 μM). Then using ab initio modeling of the ligand peptide structure, computational docking, and recently developed free energy estimation protocols, I rationally redesigned the peptides to increase their binding affinity. Synthesis of the redesigned peptide and binding studies demonstrated approximately a 10–28-fold improvement in the binding affinity [8].

Biomarker Probe. CRIP1 is a protein that is overexpressed in breast cancer cells. To detect early cancer onset, we developed a probe to CRIP1 consisting of a cyclized ligand peptide and a fluorescent molecule. Figure adapted from reference [8].

Collaborators: Dr. James Basilion and Jihua Hao (Case Western Reserve University Department of Biomedical Engineering).

Reference:
[8] J Hao*, AWR Serohijos*, G Newton, G Tassone, DC Sgroi, NV Dokholyan, JP Basilion. PLOS Comp Biol 4: e1000138 (2008) [*Equal contribution]