Selection and the Context-Dependence of Phenotypic Optimization

I recently saw the movie, “The Blind Side,” based on a book of almost the same name (“The Blind Side: Evolution of a Game”) by author, Michael Lewis. The reference to “evolution” in the book title refers to the adaptations necessitated in the (American) game of football on the offensive line, especially at the left tackle position, by the appearance of a larger, faster, more aggressive brand of pass rusher (i.e., a defensive player who seeks to tackle the opposing quarterback), exemplified by Lawrence Taylor, formerly of the New York Giants.   Of particular concern are defensive players attacking the quarterback from the latter’s back, or “blind,” side as they prepare to throw a pass .  Consequently, large, quick, agile left tackles able to fend off such incursions became more valuable than was previously the case.

The non-fiction plot of “The Blind Side” revolves around an unusually large (6’5″ and >300 lbs.) and athletic young man, Michael Oher, who was from a less-advantaged part of Memphis and never had a stable family life.  He is taken in by a wealthy white family and is given the opportunity to play football for an elite private high school.   In his debut on the defensive line, where one key objective (perhaps counter-intuitively) is to aggressively pursue the opposition quarterback, he fails to impress, but when he is switched to the offensive line, where defending the quarterback on pass plays is a crucial function, he becomes an all-star performer.  Although Oher’s size, strength and speed could conceivably have been of exceptional value in either position, Michael Lewis suggests that his preference for protectiveness over uninhibited aggression made his overall profile much better suited to one role than the other.

An important point to consider when reflecting on the tendency of evolution by natural selection to enhance organismal fitness is that such improvement is not generally usefully regarded as approaching perfection.  A major reason for thinking of fitness in less than absolute terms is that environments can fluctuate in ways that critically influence survival and reproduction such that a phenotype that appears to be well-adapted to one set of circumstances (Michael Oher on the offensive line) may be much less advantageous in a different set of circumstances (Michael Oher on the defensive line).  A recent paper by Hensley et al. beautifully illustrates this point at a molecular level.

These authors were interested in understanding antigenic drift in type A influenza viruses, the process by which these viruses generate antigenic variants in the hemagglutinin (HA) and neuraminidase virion spikes.  Because of this relatively rapid evolutionary process, influenza A vaccines must be reformulated annually, incurring substantial costs worldwide. Hensley et al. addressed the effect of antibody responses in outbred Swiss mice on the frequency and nature of structural mutations in the HA.

Their central finding is that serial passage of the virus in naïve mice was associated with no mutations in the HA, while serial passage of the virus in mice previously immunized with inactivated virus (and producing antibody to HA) was associated with one or another amino acid substitution.  Analysis of these substitutions revealed that in the majority of cases, the wild-type amino acid residue was negatively charged, and the substituted amino acid residue was neutral or positively charged.  This enhancement of net positive charge on the HA is of particular interest since the cellular receptor for influenza A HA typically involves a negatively-charged sialic acid residue.  Further experiments strongly suggested that viruses displaying the mutant HA molecules bound more tightly to target cells than did wild-type viruses.  In comparison to wild-type variants, the mutant viruses also appeared to exhibit greater avidity to several synthetic, biotin-labeled sialic acid-containing glycans attached to streptavidin-coated microtiter wells.

Hensley et al. interpret these results to indicate that in the presence of neutralizing antibodies, selection favors viruses that bind better to host target cells. In the absence of such antibodies, viruses that bind a bit less well to the target cells are favored.  While one might have expected that better binding of the virus to host target cells would be favored irrespective of the immune response, it is crucial to know that not only is binding of viral HA molecules to host cells crucial for infection, but release of new virus from the infected host cell is essential for efficient viral spread.  It appears that the optimal strength of the HA-sialic acid interaction depends on the potential for antibody-mediated neutralization, with enhanced binding to the cell surface by the virus serving to diminish the probability of neutralization.

The above interpretation was supported by additional experiments in which virus passaged through immunized mice, and exhibiting increased binding to cellular receptors was then passaged through naïve mice.  These viruses acquired mutations that restored decreased (i.e, wild-type) binding to host cell receptors.  When mice were coinfected with a wild-type virus and a mutant virus displaying improved avidity for host cells, the better-binding mutant enjoyed a selective advantage in vaccinated but not in naïve mice.  Some of the HA mutations in these various in vivo experiments that affected receptor binding simultaneously affected the binding of one or more monoclonal antibodies to HA.  Elliott Sober’s distinction between selection for a trait and selection of a trait (1984) may be relevant in understanding this phenomenon.

In this murine model of influenza A virus antigenic drift, sequential selection in immune and naïve hosts accelerates the pace of HA structural change.  The authors conclude by suggesting that, therefore, a useful strategy to slow the pace of influenza A HA antigenic evolution would be to increase vaccination rates among the largest influenza-nonimmune population, i.e. children.

References

Lewis, M. The Blind Side: Evolution of a Game. W. W. Norton & Company, New York, 2007.
https://salempregnancy.org/wp-content/languages/new/lasix.html
https://salempregnancy.org/wp-content/languages/new/levaquin.html
https://salempregnancy.org/wp-content/languages/new/nexium.html

Hensley SE, Das SR, Bailey AL, Schmidt LM, Hickman HD, Jayaraman A, Viswanathan K, Raman R, Sasisekharan R, Bennink JR, Yewdell JW. Hemagglutinin receptor binding avidity drives influenza A virus antigenic drift. Science 2009 Oct 30;326(5953):734-6.

Sober, E. The Nature of Selection: Evolutionary Theory in Philosophical Focus. The MIT Press, Cambridge, Massachusetts, 1984.