Stress and the Expression of Phenotypic Variation

Although selection acts at the phenotypic level, adaptive responses to stressors ultimately will be dictated by the presence of genetic variation in traits in the direction which selection acts. Typically, narrow sense heritability for a single trait (h2 = VA/VP), which is the slope of the relationship between the additive genetic variance (VA) and total phenotypic variance (VP), is used to predict a population's response to selection (R = h S; see Units of Selection). An alternative predictor is evolvability, an expression of the ratio between the additive genetic variance and mean (x) of a trait, such as VA/x . The additive genetic variance (VA) is the proportion of the genetic component that contributes to the resemblance between parents and offspring, while the phenotypic variance includes additive genetic variance, genetic interactions between alleles within or between loci and environmental variance. Changes in any one of these components may alter a population's response to selection.

The debate as to whether narrow-sense heritability (VA/VP) or evolvability will decrease or increase as a consequence of stressful environments is still to be resolved. Recent empirical studies have emphasized the absence of simple answers to this question by showing both increases and decreases in heritability depend on traits and species measured (Table 1). Heritability is predicted to decrease under stressful conditions due to an increase in environmental variation, increasing phenotypic variation without changing VA. In addition, stressful conditions can directly decrease VA, as seems to happen in some animal populations when food is scarce and the proportion of VP due to genetic factors decreases relative to environmental factors. Heritability and selection responses can increase under stressful conditions due to the release of cryptic genetic variation as a by-product of canalization, or through an increase in mutation or recombination rate in stressed organisms. The results of numerous laboratory experiments and those undertaken in natural populations have emphasized that the heritability or evolvability of a trait are not stable measures but depend on environmental conditions.

Canalization refers to the process that maintains pheno-typic constancy despite fluctuations in both the genotype and environment. A classic example of a 'canalized' trait is scuttelar bristle number in Drosophila (which rarely varies from four). A highly canalized phenotype will vary little within a range of conditions and this is often referred to as the zone ofcanalization; outside this range the trait may show an increase in variation. Canalization is thought to be maintained through genetic mechanisms and result in an accumulation of cryptic genetic variation not expressed within a trait's zone of canalization. However, stressful conditions may disrupt the buffering mechanism and lead to an increase in variation. A heat shock protein (Hsp90) has been implicated as a candidate mechanism for canalization; this protein is present in plants, fungi, and animals, and plays a role in folding proteins, particularly those involved in signal transduction, cell cycle, and cell death. Experiments on Drosophila, Arabidopsis, and fungi, have demonstrated increases in variation following suppression/disruption of Hsp90 with chemical inhibitors or temperature stress. Most recently, experiments manipulating Hsp90 expression in fungi demonstrated its involvement in the evolution of drug resistance; the suppression ofHsp90 allowed new mutations to have immediate effects on the phenotype. Furthermore, the drug-resistant phenotype became stable without the suppression of Hsp90 following intense selection. This article highlights the important role cryptic genetic variation and thus canalization may have in rapid

Table 1 Trends in heritability estimates in stressful environments





D. buzzati (insect) D. melanogaster

Stator limbatus Pholcus phalangioides

Rana temporaria (amphibian)

Bufo calamita Parus caeruleus (bird) Pica Pica

Crassostrea gigas (bivalve molluscs)

Salmo salar (fish)

Ovis Canadensis (mammal)

Ovis aries

Tamiasciurus hudsonicus

Wing Length Wing Length

Bristle number

Development time

Egg length Lifetime fecundity Body size Development time Body mass Development time Mass

Body length Tail length

Survival to 25 stage Gosner

Nestling tarsus length

Tarsus length

Body mass

Immune response



Reproductive effort Juvenile length Adult body mass Male parasite resistance Female parasite resistance Growth rate

High temperature High temperature Low temperature High temperature Low temperature High temperature Low temperature Environmental quality

Food abundance


Osmotic stress Parasite load

Food abundance and quality

Food abundance

Habitat quality



Annual quality

'+' indicates an increase in heritability while '-' represents a decrease and '=' means no change. Multiple estimates represent more than one study, while brackets indicate statistical significance was not tested or results were only suggestive. Data from:

aHoffmann AA and Merila J (1999) Heritable variation and evolution under favorable and unfavorable conditions. Trends in Ecology and Evolution 14: 96-101.

bBubily OA and Loeschcke V (2001) High stressful temperature and genetic variation of five quantitative traits in Drosophila melanogaster. Genetics 110: 79-85.

cBubily OA and Loeschcke V (2002) Effect of low stressful temperature on genetic variation in five quantitative traits in Drosophila melanogaster. Heredity 89: 70-75.

dCharmantier A and Garant D (2005) Environmental quality and evolutionary potential: Lessons from wild populations. Proceedings of the Royal Society of London Series B 272: 1415-1425.

evolution following an environmental stress. The idea of cryptic genetic variation in itself is interesting from an adaptive perspective; is cryptic genetic variation a by-product of canalization or a pre-adapted mechanism to cope with stressors?

Mutation is the ultimate source of genetic variation, and under stressful conditions mutation rates may increase, introducing new variation available for selection when variation is potentially needed most. For example, some studies of the bacterium E. coli suggest that the stressful starvation conditions encountered during stationary phase incubation result in a temporary increase in mutation rate due to decreased fidelity of DNA replication and a reduction in DNA repair. There is also evidence that recombination frequencies may be associated with stress levels and produce an increase in combinations of different alleles in offspring. Several experiments on Drosophila have shown increases in recombination frequencies under extreme culturing temperatures and starvation, while overcrowding in mice has also been found to increase recombination. Abiotic stress factors like heat and increased salinity can stimulate somatic recombination in Arabidopsis.

Although an increase in mutation and recombination rates may promote novel variation during environmental stress, there are also likely to be deleterious effects in stressful environments. Most mutations are deleterious and selection acts against them; an increase in mutation rate may increase the mutational load in a population exposed to stressful conditions. Many new combinations of alleles generated by recombination may also be deleterious, generating costs. It is often not clear if an increase in recombination/mutation rate is adaptive or simply a consequence of exposure to stressful conditions.

In addition to influencing the heritability and evolvability of traits directly, stressful conditions can also exert indirect effects through changes in population size. A decrease in population size is expected to lead to lower levels of additive genetic variance and decrease the adaptive potential of populations as alleles are lost in small populations due to genetic drift; however, the decrease in size often needs to be quite substantial to have much impact on trait heritability. The adaptive potential of small populations can also be reduced through a loss of fitness due to inbreeding.

Theory suggests that the effects of population size on evolvability can be unpredictable when there are nonadditive genetic interactions among alleles within (dominance) or between (epistasis) loci. In these cases additive genetic variation may increase when population size declines rapidly. This additive genetic variance, previously hidden from selection by epistatic or dominance interactions, may now contribute to a selection response. There is some empirical evidence that additive genetic variance can increase after a population bottleneck for morphometric and behavioral traits in the house fly, and viability in the flour beetle and D. melanogaster; however, the evolutionary relevance of such increases is not clear. In the case of viability in D. melanogaster, although the response to selection was higher in the bottlenecked lines, it was not high enough to overcome the initial level of inbreeding depression caused by bottlenecking. As such, although a population bottleneck increased additive genetic variance, in this case it is not likely to be evolutionary significant.

Project Earth Conservation

Project Earth Conservation

Get All The Support And Guidance You Need To Be A Success At Helping Save The Earth. This Book Is One Of The Most Valuable Resources In The World When It Comes To How To Recycle to Create a Better Future for Our Children.

Get My Free Ebook

Post a comment