National Science Foundation     |     Directorate for Engineering  (ENG)
Division of Chemical, Bioengineering, Environmental, & Transport Systems  (CBET)
CBET Award Achievements  (Formerly "CBET Nuggets")
Notable Accomplishments from CBET Awards
Development of Computational Tools and Experimental Verifications for Protein Design
Costas Maranas  -  The Pennsylvania State UniversityUniversity Park, PA
Patrick Cirino

The Maranas and Cirino engineering groups, at Pennsylvania State University, are developing new computational and experimental tools to design new proteins that could be used to speed up biochemical reactions or recognize foreign toxins that are associated with disease.  They have redesigned a number of proteins using this methodology including an enzyme used to make a sugar substitute so that the reaction can be carried out using cheaper starting materials.  This work could lead to the design of more stable, and more effective proteins that carry out biochemical reactions, or used as protein therapeutics to diagnose and treat disease.

Background:  Proteins are versatile molecules tuned to perform a diverse set of functions, including catalysis, signaling and regulation, and medicinal applications.  Their potential to provide solutions to challenges in biomass treatment, biofuels production, biosensing, and wastewater and environmental pollution has long been recognized.  Regrettably, proteins with potential commercial utility often are unstable, and do not function with alternative, more cost-effective substrates.
Protein design is effective at identifying proteins with improved performance characteristics. Purely experimental approaches are challenged by the enormous protein design space.  An alternative approach pursued here is to develop accurate computer-based tools that can predict potential protein designs before stepping into the lab.
The research teams are developing computational tools to design or redesign proteins to bind more cost-effective substrates as well as antibodies to tightly bind antigens, and then conducting experiments to test the effectiveness of the developed computational frameworks and reveal opportunities for improvement.

Results:  The Iterative Protein Redesign and Optimization (IPRO) algorithm previously created by the Maranas group has been improved to include biologically important water solvent effects as well as to run in parallel, which significantly decreases the time necessary to computationally design a target protein.  The protein CbXR naturally converts D-xylose to xylitol (high value-added sugar-substitute with many commercial applications) utilizing the cofactor NADPH.  The improved IPRO algorithm was able to redesign CbXR to utilize NADH (which is 10X as abundant and more stable) as its cofactor, leading to a potentially cost-effective catalyst for industrial xylitol production.  This NSF-funded work was featured as the cover article of Protein Science (Protein Sci  18(10), 2125-2138., 2009).  A redesigned nicotinamide binding pocket is shown in Figure 1A.
Antibodies are proteins in vertebrate immune systems that bind foreign molecules and mark them for elimination from the body.  Antibody-based therapeutics have ushered in new treatment options unavailable only a few years ago.  The OptCDR computational procedure has been developed as a method to design novel libraries of antibodies to bind any pre-specified antigen.  It works by first identifying combinations of protein structures that are likely to bind the antigen well and then filling in their amino acid sequences.  To date, it has been applied to generate antibodies to bind a peptide from hepatitis C and vascular endothelial growth factor, a protein implicated in the progression of a number of cancers.  Figure 1B highlights a novel OptCDR-designed antigen binding site.  This method will be published soon.
The OptGraft algorithm was recently introduced for grafting a binding site from one protein into a new protein scaffold.  This method identifies the best locations to graft the new binding pocket via geometric optimization, and then utilizes an energy function to determine what mutations may be needed in neighboring residues to conserve the desired geometry.  There are many reasons to develop such an algorithm; one can introduce a second functionality to an existing protein, as well as grafting new functionality into a stable protein.  This method was used to guide the transfer of a calcium binding pocket from thermitase protein into the first domain of CD2 protein.  Experimental verification was performed validating the efficacy of this algorithm (Protein Sci  18(1), 180-195., 2009).  Figure 1C pictorially illustrates the grafting from a donor to a target protein scaffold.
Regulatory proteins with customized effector specificity (proteins that target cells that perform a specific function in response to a stimulus) can find applications in metabolic engineering, molecular reporting and biosensing.  The well-studied AraC regulatory protein naturally switches from a transcriptional repressor to an activator upon binding L-arabinose.  In efforts to alter effector specificity and better understand the determinants of molecular recognition in this protein, the IPRO computational framework was extended to include specificity in the design of protein variants (Biophysical Journal  92, 2120-2130, 2007).  Mutations that minimize the binding energy with the desired ligand are identifed, while at the same time explicit constraints are introduced that maintain the binding energy for all decoy ligands above a threshold necessary for binding.  This amended framework was demonstrated by computationally altering the effector binding specificity of AraC for unnatural ligands.  The results demonstrated the importance of systematically suppressing the binding energy for competing ligands.  Pinpointing a small set of mutations within the binding pocket greatly improved the difference in binding energies between targeted and decoy ligands, even when they are very similar.

Costas Maranas Image 1
Figure 1.  Highlights from the Maranas and Cirino Team's work on Developing Computational Tools and Experimental Verifications for Protein Design:
(AThe nicotinamide binding site of a redesigned CbXR is shown, with the alternative substrate, NADH bound. Residues in yellow are designs selected by IPRO that computationally increased the binding affinity for NADH, and were shown to experimentally have new activity.
(BOptCDR-designed complementarity determining regions in complex with their target antigen, a peptide from the capsid of hepatitis C.
(CPictorial description of OptGraft transferring the naturally occurring binding pocket of a donor protein into a new scaffold without disrupting its function.
Image Credit:  George Khoury, Robert Pantazes, Patrick Cirino, and Costas Maranas
                    The Pennsylvania State University

Scientific Uniqueness:  The team has created a set of novel computational methods (IPRO, OptCDR, OptGraft) for the identification of promising protein designs followed by experimental verification.  The team successfully used energy functions coupled with optimization to design proteins that were experimentally validated to have new, more useful properties for industrial and medicinal applications.  In addition, the Maranas group developed the first method to design libraries of antibodies to bind any specified antigen.  These generalized approaches can be used by the scientific community to design other proteins of scientific, industrial, and medicinal interest.

This project addresses the NSF Strategic Outcome Goals, as described in the NSF Strategic Plan 2006-2011, as follows:
Primary Strategic Outcome Goal:      (1) Discovery:  NSF emphasizes investigations that are multi-disciplinary and require systematic approaches to address complex problems.  This research has created algorithms to discover novel protein variants that are capable of having new or improved properties and functionalities relative to their naturally-found forms.  These computational methods are geared to the design of antibodies with targeted affinity and biocatalysts for the bioproduction of chemicals.
                                                                   (1) Discovery Categories:
                                                                           -  Biology
                                                                           -  Computer & Information Science and Engineering
                                                                           -  Engineering
                                                                           -  Mathematical and Physical Sciences

Secondary Strategic Outcome Goal:  (2) Learning:  Three graduate students involved in this research learned methods of designing computer algorithms, biophysics, and experimental mutagenesis techniques.  They obtained a broad knowledge in the field of protein engineering while gaining a deep understanding of the molecular interactions that lead to successful protein designs.  Top undergraduates have been recruited and introduced to research and protein design, and they have been given primary authorship of their work.

                                                                   (2) Learning Categories:
                                                                           -  Undergraduate Education and Undergraduate Student Research
                                                                           -  Graduate Education and Graduate Student Research
                                                                           -  International Research Experiences for Undergraduate and
                                                                              Graduate Students

This Award Achievement represents Transformative Research.  This research is transformative in the way that computational methods are interfaced with experimental efforts to arrive at more efficient paradigms for the design of proteins and antibodies.

The Intellectual Merit of this research:  The intellectual merit of this work lies in the development of sophisticated computational methods drawing from many disciplines ranging from combinatorial optimization to biophysics of molecular interactions to meet the challenge of the a priori design of proteins (i.e. enzymes, antibodies) with pre-specified function.

The Broader Impacts of this research include:
Broadening the participation of underrepresented groups::  A key aspect of this research was providing scientific research opportunities to the largest undergraduate chemical engineering population in the country at Penn State.  Top undergraduates are recruited and retained, and are paired with graduate students to mentor them, but are trained to have ownership of their work.  These undergraduates have orally presented their work at the American Institute of Chemical Engineers and American Chemical Society national meetings, and have published technical papers in leading journals with primary authorship.  The exposure gained by these undergraduates to the process of scientific research and discovery is invaluable, and in recent years has motivated them to pursue further studies at leading graduate programs.  In addition, this research has been integrated to establish successful applications of the developed algorithms with Penn State's International Genetically Engineered Machines program (iGEM), which is aimed at providing undergraduates and high-school students hands-on exposure to synthetic biology.
Dissemination of results broadly to enhance scientific and technological understanding:  The protein design algorithm IPRO has been improved and made accessible to the scientific community on the Maranas group website ( so that others can systematically design proteins of their own interest.

Area of Emphasis (Themes) for FY 2010 Highlights included in this research project:
(1Interdisciplinary, high-risk, and potentially transformative
(2Speeds translation of promising fundamental research into innovations that can be commercialized
(3Enhances health and quality of life
(4Infuses computational thinking into all areas of engineering, bringing computational capabilities into the traditional experimentation-observation-analysis-theory research paradigm (CDI)
(5Nurtures a world-class engineering workforce and a technically literate population

Program Director:
Theresa A. Good
CBET Program Director - Biotechnology, Biochemical, and Biomass Engineering
NSF Award Number:   0639962
Award Title:   Development of Computational Tools and Experimental Verifications for Protein Design
PI Name: 
  Costas Maranas and
Patrick Cirino
Institution Name:   The Pennsylvania State University;  University Park, PA
Program Element Code:   1491
CBET Award Achievement:

  FY 2010

Top of Page

This Award Achievement was Updated on 19 August 2010.