What is epistasis?
The term “epistasis” is of Greek and Latin origin, meaning “to stand upon” or “stoppage.” The term was originally used by geneticist William Bateson at the beginning of the twentieth century to define genes that mask the expression of other genes. The gene at the initial location (locus) is termed the epistatic gene. The genes at the other loci are “hypostatic” to the initial gene. In its strictest sense, epistasis describes a nonreciprocal interaction between two or more genes, such that one gene modifies, suppresses, or otherwise influences the expression of another gene affecting the same phenotypic (physical) character or process. By this definition, simple additive effects of genes affecting a single phenotypic character or process would not be considered an epistatic interaction. Similarly, interactions between alternative forms (alleles) of a single gene are governed by dominance effects and are not epistatic. Epistatic effects are interlocus interactions. Therefore, in terms of the total genetic contribution to phenotype, three factors are involved: dominance effects, additive effects, and epistatic effects. The analysis of epistatic effects can suggest ways in which the action of genes can control a phenotype and thus supply a more complete understanding of the influence of genotype on phenotype.
A gene can influence the expression of other genes in many different ways. One result of multiple genes is that more phenotypic classes can result than can be explained by the action of a single pair of alleles. The initial evidence for this phenomenon came out of the work of Bateson and British geneticist Reginald C. Punnett during their investigations on the inheritance of comb shape in domesticated chickens. The leghorn breed has a “single” comb, brahmas have “pea” combs, and wyandottes have “rose” combs. Crosses between brahmas and wyandottes have “walnut” combs. Intercrosses among walnut types show four different types of F2 (second-generation) progeny, in the ratio 9 walnut:3 rose:3 pea:1 single. This ratio of phenotypes is consistent with the classical F2 ratio for dihybrid inheritance. The corresponding ratio of genotypes, therefore, would be 9 A_ B_:3 A_ bb:3 aa B_:1 aa bb, respectively. (The underscore is used to indicate that the second gene can be either dominant or recessive; for example, A_ means that both AA and Aa will result in the same phenotype.) In this example, one can recognize that two independently assorting genes can affect a single trait. If two gene pairs are acting epistatically, however, the expected 9:3:3:1 ratio of phenotypes is altered in some fashion. Thus, although the preceding example involves interactions between two loci, it is not considered a case of epistasis, because the phenotype ratio is a classic Mendelian ratio for a dihybrid cross. Five basic examples of two-gene epistatic interactions can be described: complementary, modifying, inhibiting, masking, and duplicate gene action.
For complementary gene action, a dominant allele of two genes is required to produce a single effect. An example of this form of epistasis again comes from the observations of Bateson and Punnett of flower color in crosses between two white-flowered varieties of sweet peas. In their investigation, crosses between these two varieties produced an unexpected result: all the F1 (first-generation) progeny had purple flowers. When the F1 individuals were allowed to self-fertilize and produce the F2 generation, a phenotypic ratio of nine purple-flowered to seven white-flowered individuals resulted. Their hypothesis for this ratio was that a homozygous recessive genotype for either gene (or both) resulted in the lack of flower pigmentation. A simple model to explain the biochemical basis for this type of flower pigmentation is a two-step process, each step controlled by a separate gene and each gene having a recessive allele that eliminates pigment formation. Given this explanation, each parent must have had complementary genotypes (AA bb and aa BB), and thus both had white flowers. Crosses between these two parents would produce double heterozygotes (Aa Bb) with purple flowers. In the F2 generation, 9/16 would have the genotype A_ B_ and would have purple flowers. The remaining 7/16 would be homozygous recessive for at least one of the two genes and, therefore, would have white flowers. In summary, the phenotypic ratio of the F2 generation would be 9:7.
The term “modifying gene action” is used to describe a situation whereby one gene produces an effect only in the presence of a dominant allele of a second gene at another locus. An example of this type of epistasis is aleurone color in corn. The aleurone is the outer cell layer of the endosperm (food-storage tissue) of the grain. In this system, a dominant gene (P_) produces a purple aleurone layer only in the presence of a gene for a red aleurone (R_) but expresses no effect in the absence of the second gene in its dominant form. Thus, the corresponding F2 phenotypic ratio is 9 purple:3 red:4 colorless. The individuals without aleurone pigmentation would, therefore, be of the genotype P_ rr (3/16) or pp rr (1/16). Again, a two-step biochemical pathway for pigmentation can be used to explain this ratio; however, in this example, the product of the second gene (R) acts first in the biochemical pathway and allows for the production of red pigmentation and any further modifications to that pigmentation. Thus, the phenotypic ratio of the F2 generation would be 9:3:4.
Inhibiting action occurs when one gene acts as an inhibitor of the expression of another gene. In this example, the first gene allows the phenotypic expression of a gene, while the other gene inhibits it. Using a previous example (the gene R for red aleurone color in corn seeds), the dominant form of the first gene R does not produce its effect in the presence of the dominant form of the inhibitor gene I. In other words, the genotype R_ i_ results in a phenotype of red aleurone (3/16), while all other genotypes result in the colorless phenotype (12/16). Thus gene R is inhibited in its expression by the expression of gene I. The F2 phenotypic ratio would be 13:3. This ratio, unlike the previous two examples, includes only two phenotypic classes and highlights a complicating factor in determining whether one or two genes may be influencing a given trait. A 13:3 ratio is close to a 3:1 ratio (the ratio expected for the F2 generation of a monohybrid cross). Thus it emphasizes the need to look at an F2 population of sufficient size to discount the possibility of a single gene phenomenon over an inhibiting epistatic gene interaction.
Masking gene action, a form of modifying gene action, results when one gene is the primary determinant of the phenotype of the offspring. An example of this phenomenon is fruit color in summer squash. In this example, the F2 ratio is 12:3:1, indicating that the first gene in its dominant form results in the first phenotype (white fruit); thus this gene is the primary determinant of the phenotype. If the first gene is in its recessive form and the second gene is in its dominant form, the fruit will be yellow. The fruit will be green at maturity only when both genes are in their recessive form (1/16 of the F2 population).
Duplicate gene interaction occurs when two different genes have the same final result in terms of their observable influence on phenotype. This situation is different from additive gene action in that either gene may substitute for the other in the expression of the final phenotype of the individual. It may be argued that duplicate gene action is not a form of epistasis, since there may be no interaction between genes (if the two genes code for the same protein product), but this situation may be an example of gene interaction when two genes code for similar protein products involved in the same biochemical pathway and their combined interaction determines the final phenotype of the individual. An example of this type of epistasis is illustrated by seed-capsule shape in the herb shepherd’s purse. In this example, either gene in its dominant form will contribute to the final phenotype of the individual (triangular shape). If both genes are in their recessive form, the seed capsule has an ovoid shape. Thus, the phenotypic ratio of the F2 generation is 15:1.
Nonallelic gene interactions have considerable influence on the overall functioning of an individual. In other words, the genome (the entire genetic makeup of an organism) determines the final fitness of an individual, not only as a sum total of individual genes (additive effects) or by the interaction between different forms of a gene (dominance effects) but also by the interaction between different genes (intragenomic or epistatic effects). This situation is something akin to a chorus: great choruses not only have singularly fine voices but also perform magnificently as finely tuned and coordinated units. Knowledge of what contributes to a superior genome would, therefore, lead to a fuller understanding of the inheritance of quantitative characters and more directed approaches to genetic improvement. For example, most economically important characteristics of agricultural species (such as yield, pest and disease resistance, and stress tolerance) are quantitatively inherited, the net result of many genes and their interactions. Thus an understanding of the combining ability of genes and their influence on the final appearance of domesticated breeds and crop varieties should lead to more efficient genetic improvement schemes. In addition, it is thought that many important human diseases are inherited as a complex interplay among many genes. Similarly, an understanding of genomic functioning should lead to improved screening or therapies.
Frankel, Wayne N., and Nicholas J. Schork. “Who’s Afraid of Epistasis?” Nature Genetics 14.4 (1996): 371–73. Print.
Jain, Kavita, and Apoorva Nagar. "Fixation of Mutators in Asexual Populations: The Role of Genetic Drift and Epistasis." Evolution 67.4 (2013): 1143–54. Print.
Mackay, Trudy F. C. "Epistasis and Quantitative Traits: Using Model Organisms to Study Gene-Gene Interactions." Nature Reviews Genetics 15.1 (2014): 22–33. Print.
Pang, Xiaoming, et al. "A Statistical Procedure to Map High-Order Epistasis for Complex Traits." Briefings in Bioinformatics 14.3 (2013): 302–14. Print.
Russell, Peter J. Fundamentals of Genetics. 2nd ed. San Francisco: Benjamin, 2000. Print.
Snustad, D. Peter, and Michael J. Simmons. Principles of Genetics. 6th ed. Hoboken: Wiley, 2012. Print.
Wan, Xiang, et al. "The Complete Compositional Epistasis Detection in Genome-Wide Association Studies." BMC Genetics 14.7 (2013): 1–11. Web. 22 July 2014.
Wolf, Jason B., Edmund D. Brodie III, and Michael J. Wade, eds. Epistasis and the Evolutionary Process. New York: Oxford UP, 2000. Print.
Wood, Stacey J., and Alison A. Motsinger-Reif. “Epistasis: Understanding and Evaluating the Phenomenon in Human Genetics Disease Mapping.” Genetic Predisposition to Disease. Ed. Sara L. Torres and Marta S. Marin. New York: Nova, 2008. 195–213. Print.