HIV Control Is Mediated in Part by CD8+ T-Cell Targeting of Specific Epitopes

  • Florencia Pereyra ,
  • ,
  • ,
  • Carl Kadie ,
  • Damien Z Soghoian ,
  • Daniel Karel ,
  • Ariel Goldenthal ,
  • Oliver B Davis ,
  • Charles E DeZiel ,
  • Tienho Lin ,
  • Jian Peng ,
  • Alicja Piechocka ,
  • Mary Carrington ,
  • Bruce D Walker

Journal of Virology | , Vol 88(22): pp. 12937-12948

Publication | Publication

Natural controllers tend to target structurally constrained epitopes

This is a great paper that provides a great rationale for vaccine design: (1) it’s critical to target specific epitopes; (2) those epitopes need to be those where mutation comes at a cost; and (3) protein structure is a great way to predict which epitopes will be costly. We did this in collaboration with Florencia Pereyra and Bruce Walker at the Ragon Institute (opens in new tab). The idea was to test a bunch of nature controllers and normal non-controllers to see which epitopes they target, whether that explains control, and what characterizes good epitopes.

Citation and access

HIV Control Is Mediated in Part by CD8+ T-Cell Targeting of Specific Epitopes (opens in new tab)

Florencia Pereyra*, David Heckerman*, Jonathan M. Carlson*, Carl Kadie, Damien Z Soghoian, Daniel Karel, Ariel Goldenthal, Oliver B Davis, Charles E DeZiel, Tienho Lin, Jian Peng, Alicja Piechocka, Mary Carrington, Bruce D Walker

Journal of Virology (opens in new tab), 88(22):12937-12948, November 2014.

Abstract

We investigated the hypothesis that the correlation between the class I HLA types of an individual and whether that individual spontaneously controls HIV-1 is mediated by the targeting of specific epitopes by CD8+ T cells. By measuring gamma interferon enzyme-linked immunosorbent spot (ELISPOT) assay responses to a panel of 257 optimally defined epitopes in 341 untreated HIV-infected persons, including persons who spontaneously control viremia, we found that the correlation between HLA types and control is mediated by the targeting of specific epitopes. Moreover, we performed a graphical model-based analysis that suggested that the targeting of specific epitopes is a cause of such control—that is, some epitopes are protective rather than merely associated with control—and identified eight epitopes that are significantly protective. In addition, we use an in silico analysis to identify protein regions where mutations are likely to affect the stability of a protein, and we found that the protective epitopes identified by the ELISPOT analysis correspond almost perfectly to such regions. This in silicoanalysis thus suggests a possible mechanism for control and could be used to identify protective epitopes that are not often targeted in natural infection but that may be potentially useful in a vaccine. Our analyses thus argue for the inclusion (and exclusion) of specific epitopes in an HIV vaccine.

Overview

This is a cool project, and it was a long time in coming, with the first results reported in the 2009 CROI meeting. Here’s the idea: a subset of people (“controllers”) naturally control HIV infection in the absence of therapy. This is obviously something we need to understand. Human genetics plays a big role, with HLA types being the main player (indeed, over 1000 (opens in new tab) papers have been published with “HLA” and “HIV” in the title!). A key function of HLA proteins is to present epitopes (protein fragments) on the cell surface for T-cells to monitor, looking for evidence of viral infection. As you’ll see below, we’ve accumulated LOTS of evidence that this is a very important mechanism, with HIV working very hard to evade this immune response through mutation. One might suspect that controllers then are individuals who target epitopes that can’t easily mutate. Although this is an actively pursued vaccine concept, direct evidence has been lacking. Here’s what we did:

For 342 individuals, roughly 2/3 controller, 1/3 not, Florencia Pereyra and Bruce Walker tested epitope responses against all known epitopes (well, known as of 2008 when this began!). David then used L1-regularized logistic regression to see which HLA alleles and which epitope responses predict control. We found (1) that epitope responses explained away all of the HLA signal (arguing that HLA protectiveness is primarily mediated by WHICH epitopes are targeted, at least in this cohort); (2) that certain “protective” epitopes are consistently associated with control, and certain “nonprotective” epitopes are associated with non-control; (3) that the there is some evidence that the protective epitopes CAUSE control, whereas the nonprotective epitopes are a RESULT of loss of control; and (4) that the protective epitopes are structurally conserved. This last point bears further explanation: current epitope-specific vaccine strategies largely focus on epitopes that are conserved, on the assumption that they’re conserved because they can’t mutate. We found a trend for this, but it wasn’t significant. Why? I don’t know, but an epitope may be conserved because there’s no functional immune response against it in nature; conversely, an epitope may lack conservation despite strong fitness costs associated with mutations within the epitope because targeting the epitope is so darn effective. To get around this, we turned to protein structures. We previously showed that you can use FoldX (opens in new tab) to estimate the change in protein structure energy when you mutate a single base, and that this correlates with the observed change in fitness. To convert this change in structural energy to a single score for a site, we first computed the changes for all 20 amino acids at that site, then took a normalized negative exponetial–ala Boltzmann distribution–to get an implied probability distribution over amino acids at the site. The entropy of this distribution tells you something about much some amino acids are favored over others structurally. Turns out, protective epitopes are ones where the average “structural entropy” over the epitope is very low, meaning mutations at most any site in the epitope are likely to affect the protein structure. Fwiw, if you include both sequence entropy and structural entropy in a logistic model and ask how they predict whether an epitope will be protective, structural entropy is significant while sequence entropy is not.

This provides us a great rationale for vaccine design: (1) it’s critical to target specific epitopes; (2) those epitopes need to be those where mutation comes at a cost; and (3) structure is a great way to predict which epitopes will be costly.

Principal collaborators

Ragon Intitute (opens in new tab)

  • Florencia Pereyra
  • Bruce Walker