July 2018
Volume 59, Issue 9
Open Access
ARVO Annual Meeting Abstract  |   July 2018
Network modeling of proteins identifies critical sites for vaccine design
Author Affiliations & Notes
  • Elizabeth Rossin
    Ophthalmology, Massachusetts Eye and Ear, Boston, Massachusetts, United States
  • James Chodosh
    Ophthalmology, Massachusetts Eye and Ear, Boston, Massachusetts, United States
  • Bruce Walker
    Massachusetts General Hospital, Boston, Massachusetts, United States
  • Gaurav Gaiha
    Massachusetts General Hospital, Boston, Massachusetts, United States
  • Footnotes
    Commercial Relationships   Elizabeth Rossin, None; James Chodosh, None; Bruce Walker, None; Gaurav Gaiha, None
  • Footnotes
    Support  Massachusetts Eye and Ear internal grant
Investigative Ophthalmology & Visual Science July 2018, Vol.59, 521. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Elizabeth Rossin, James Chodosh, Bruce Walker, Gaurav Gaiha; Network modeling of proteins identifies critical sites for vaccine design. Invest. Ophthalmol. Vis. Sci. 2018;59(9):521.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : There are currently no effective vaccine candidates for a number of viral pathogens including HIV and HSV despite their widespread and destructive nature. Sequence conservation was initially though to be the key, however this approach has fallen short. Here we describe an algorithm that aims to show that analyzing viral structure rather than sequence alone is critical to identifying targets.

Methods : We developed a network-based algorithm that transforms the 3-dimensional structure of viral proteins into two-dimensional networks. Using crystal structures, our algorithm performs the following steps: (1) identification of the complete set of intermolecular interactions between the amino acids of the protein, which includes all Van der Waals, hydrogen, salt bridge, disulfide, pi-pi, pi-cation, water bridge and metal bonds, (2) transforms the three-dimensional protein structure into a 2-dimensional network with amino acids represented as nodes and the inter-molecular interactions as edges and (3) use network analysis calculations to identify critical nodes which represent highly constrained amino acids that are intolerant to mutation.

Results : The networked amino acids theoretically represent critical amino acids that cannot tolerate any mutation or immune targeting. We benchmarked the algorithm on bacteriophage T4 lysozyme, a well-studied protein with extensive mutagenic and functional data, and show a strong correlation between network score tolerance to mutation (p<0.0001). We expanded the approach to 20 human proteins for which we have single nucleotide polymorphism and phenotypic data and find a high correlation (p<0.001). Finally, we applied the algorithm to viral proteins in the HIV proteome, and we find that patients who naturally control HIV infection elicit strong immune responses to the highly interconnected sites while patients with progressive disease do not. The next step is to apply these concepts to HSV.

Conclusions : We have developed a novel network-based algorithm that identifies critical amino acids within a protein. We apply it to a number of proteins that have well studied mutational data and show that it predicts amino acids that are less tolerant to mutation. The potential implication is the ability to design “smart vaccines” that focus immune responses on critical regions of viral proteins and thereby directly counteract the ability of these pathogens to escape from the vaccinated immune system.

This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×