Correction: Under "Major concepts", the ratio for recessive epistasis is "9:4:3", and not 12:4 as previously shown. The ratio in the main text was correct as originally posted.
Textbook assignment: Chapter 13, Pages 393 - 404. These notes include some material not in the textbook and should be regarded as supplemental reading for this lecture.
Major concepts:
Modification of dihybrid ratios: A large portion of this lecture focuses on various ways in which the simple Mendelian F2 dihybrid ratio of 9:3:3:1 can be modified because of altered phenotypic distributions at one or both of the loci being examined, or because of interactions between alleles at two independently assorting loci. None of the examples in this lecture involve linkage or crossing over.
Effects of altered phenotypic distribution at one of the loci: Independent assortment of alleles at two different loci that have different patterns of dominance, partial dominance or codominance can lead to unusual phenotypic ratios. The example cited in class during the previous lecture involves the mating of two individuals who are both blood type AB and also heterozygous for recessive albinism. The AB blood types will segregate in a 1:2:1 ratio and the albinism will segregate in a 3:1 ratio. Since these are independently assorting events, the product rule will yield six different phenotypes in the following ratios.
Epistasis: The term epistasis is derived from a Greek word that means stoppage. Some purists still attempt to limit its usage to situations where alleles at one genetic locus directly block or otherwise significantly alter phenotypic expression of alleles at another locus. However, it is becoming increasingly common to use the term in a more generic sense to describe any situation in which alleles at two different genetic loci interact to influence a single well defined phenotypic property such as coat color or the shape of a chicken's comb. In cases where direct masking of expression occurs, the locus whose expression is masked is described as hypostatic, and the locus whose alleles cause the masking is described as epistatic.
Dominance-based genetic notation: To keep the descriptions of genotype as simple as possible, We will follow the standard convention of using a capital letter followed by a hyphen (A-) to describe any genotype that contains at least one copy of the dominant allele (AA, Aa). Thus, the designation A-B- will describe any genetic combination that includes at least one dominant allele at each locus(AABB, AABb, AaBB, or AaBb).
Epistasis in Bombay phenotype: We have already observed an example of epistasis in failure of expression of type A and type B blood groups by individuals with the Bombay phenotype. The progeny of matings of individuals who are blood type AB and heterozygous FUT1*H/*O will have the following phenotypic ratios:
Shape of comb in chickens: Our textbook introduces epistasis with a brief summary of the interactions of two separate loci in determining the shape of the comb on a chicken's head. As illustrated in figure 13.14, four different shapes are possible depending on which alleles are expressed at the two loci. In this example, there is no alteration of the expected 9:3:3:1 dihybrid ratio, and the only evidence of epistasis is the four quite different phenotypes generated by the interaction of the alleles.
Altered phenotypic ratios: Figure 13.15 illustrates 5 different modified F2 ratios that can be obtained in dihybrid crosses with various types of interaction that involve only fully dominant and fully recessive behavior of alleles at two different genetic loci The situation can become even more complex if intermediate phenotypes due to codominance or partial dominance are allowed, as illustrated by the interaction between the ABO locus and the FUT1 locus (Bombay phenotype) discussed above.
Complementary gene action (F2 ratio = 9:7): Blocking any step in a sequential enzymatic process can prevent synthesis of the final product of the pathway. Anthocyanin pigment synthesis in sweet peas is an example where blocking either of two steps will prevent pigment formation. Only the double dominant phenotype produces both of the needed enzymes and is able to synthesize the pigment (figure 13.17). The F2 progeny ratio is 9 pigmented (A-B-) to 7 unpigmented (A-bb, aaB-, or aabb) as shown in figure 13.16. Similar results can also be obtained obtained if the products of two independently coded enzymes must interact to yield the final product (figure 13.18). As described in boxed example 13.3 (and in figure 13.16), crossing two strains of sweet peas that are white because of different mutations in the anthocyanin biosynthetic pathway can result in F1 progeny with purple flowers. Please note that some purists do not consider complementary gene action to be an example of epistasis because it involves two steps in the same pathway rather than modification of the effects of one gene by another. However, all four of the textbooks that I have used for this course over a period of six years have included it under epistasis.
Similarity to Complementation: Complementary gene action is the same in principle as complementation, which can be defined as the ability of two defective genomes to work together to support a function that neither alone can support. We have already examined a few examples of complementation. For example, in chapter 7 (pages 201-202 and figure 7.25) we saw that an F' plasmid containing a lactose operon with a defective permease gene (LacY- ) was able to restore the ability to utilize lactose to an E. coli cell whose chromosomal lactose operon had a defective beta-galactosidase gene (LacZ- ). We also saw complementation between Drosophila mutations blocked at different steps in the synthesis of brown pigment (figure 6.11). In example 3 in that figure, a vermillion imaginal disc implanted into a cinnabar larva was capable of synthesizing brown pigment. A third example that we studied is the ability of auxotrophic strains of Neurospora blocked at two different steps in the biosynthesis of an essential nutrient to complement each other (Lecture 8 notes). We will see more examples of complementation, including its use in genetic fine structure analysis in a future lecture (textbook pages 498 - 504).
Both loci must affect the same phenotype: Complementary gene action is in essence complementation between haploid genomes with defects in two different genetic loci, such that the diploid genome that they produce has one functional allele at each of the genetic loci, and is thus able to support full function. Thus, in the example described above, crossing two strains with white flowers can yield progeny with purple flowers. It is also necessary to specify that both defects (or recessive traits) affect the same phenotype. Without this restriction, any two true-breeding recessive strains could be considered to complement each other. For example a cross between round green peas and wrinkled yellow peas will yield an F1 hybrid that exhibits both dominant traits and thus produces round yellow peas. Although similar in principle, this is not considered to be complementary gene action or complementation because two different phenotypic properties are involved.
Duplicate gene action (F2 ratio = 15:1): If either of two genes can achieve the same result, both halves of the redundant process must be blocked to prevent phenotypic expression. Only the double recessive genotype will exhibit the recessive phenotype (15:1 phenotypic ratio in F2). The example presented in our textbook is Winter wheat. A dominant allele at either of two loci causes the wheat to have a Spring growth pattern that includes lack of ability to survive over the Winter. Only the double recessive exhibits the Winter wheat growth characteristics, which permit it to be planted and germinated in the Fall and then to survive in a dormant state through the Winter and mature during the following Spring and early Summer.
Dominant suppression (F2 ratio = 13:3): The textbook describes two closely related situations in which the dominant allele at one of the genetic loci suppresses expression at the other locus. The slightly different phenotypic ratios that are obtained reflect the nature of the dominant and recessive forms at the second locus. If the recessive phenotype is the same as the phenotype caused by the dominant inhibitory allele, a 13:3 ratio is obtained. If not, a 12:3:1 ratio is obtained, as described below under dominant epistasis. The example of dominant suppression cited in our textbook involves feather color in chickens. A coat color locus C determines whether the feathers are pigmented (C-) or white (cc). An inhibitory locus I blocks all expression of feather pigment when the dominant allele (I-) is expressed and has no effect when recessive (ii). Thus, the F2 phenotypes are 3/4 I-, all white, 3/16 iiC-, colored, and 1/16 iicc, also white. This yields an F2 ratio of 13 white : 3 colored.
Dominant epistasis (F2 ratio = 12:3:1): In a very similar situation, a dominant inhibitory allele can mask expression of a second locus whose phenotypes are both different from the inhbitied state. The example in the textbook involves a locus whose alleles produce red or yellow coloration in onions plus a dominant inhibitory allele at a second locus that blocks all pigment production and results in a white color (figure 13.22). In this case 3/4 of the F2 progeny are white, with a 3:1 ratio of red to yellow in the remaining 1/4, yielding a 12:3:1 ratio of white:red:yellow. The textbook emphasizes the "leaky" nature of the yellow allele, which allows some color in the yellow onions. However, an entirely similar situation can occur anytime that the locus that is suppressed by the inhibitor yields two phenotypes that are both different from the suppressed phenotype. A widely cited example is fruit color in squash, where dominant white is epistatic over a second locus that makes the fruit yellow in the dominant form and green in the recessive form. This results in an F2 ratio of 12/16 white, with the remaining colored fruit in a typical 3:1 ratio of 3/16 yellow and 1/16 green.
Recessive epistasis (F2 ratio = 9:4:3): In recessive epistasis, expression of the homozygous recessive state at one locus totally blocks expression of traits controlled by a second locus. Thus, for example, an albino animal with no pigmentation is incapable of expressing the effects of alleles at other loci that influence coat color. Similarly, recessive mutations in Drosophila that cause total absence of wings or eyes block phenotypic expression of mutations at other loci that modify wing or eye phenotypes. The example cited in the textbook involves coat color in Labrador retrievers. The recessive yellow mutation (ee) restricts pigment deposition in the hair such that the effects of alleles at a second locus (B) that would otherwise cause the coat color to be black (B-) or chocolate (bb) cannot be seen. Thus, among the 3/4 of the F2 progeny that are E-, 3/4 will be black and 1/4 will be chocolate. The 1/4 of the progeny that are ee will all be yellow. This yields an F2 ratio of 9:4:3 black:yellow:chocolate (figure 13.23). .
Fruit shape in summer squash (F2 ratio = 9:6:1): An interesting example that is not mentioned in our textbook is the the genetic control of the shape of the fruit in summer squash. Elongated fruit are homozygous recessive (aabb) at two separate loci that influence shape. At the other extreme, fruit that have at least one dominant allele at each of the two loci (A-B-) are flattened to a disc shape. Fruit that are recessive at one of the loci but have at least one dominant allele at the other exhibit an intermediate spherical shape, with no obvious phenotypic difference based on which locus carries the dominant allele (A-bb or aaB-). This results in an F2 phenotypic ratio of 9/16 disc-shaped, 6/16 spherical, and 1/16 elongated.
Coat color in mice: Coat color in mice provides some fascinating examples of epistatic interactions. These are not described in the textbook, except for some indirect (and rather confusing) references in the end-of-chapter exercises. Please regard the material that follows as a supplement to the text material on epistasis. We will be examining interactions among three genetic loci that affect coat color. To keep the discussion as simple as possible, we will continue to use the older designations of A/a, B/b, and C/c for these genetic loci. However, please be aware that the preferred designations for B and C are now tryp1 and tyr, respectively, based on current knowledge of the biochemical functions of their gene products..
The agouti (A/a) locus causes cyclic switching between black and yellow hair pigment in the wild-type (A-) phenotype (figure 13.8). This is what gives wild mice their mousy gray color. The recessive aa phenotype is solid black because the yellow pigmentation is not switched on at all. Dominant yellow (Ay) occurs when agouti is continuously overexpressed, as described in the previous lecture (figure 13.16).
The brown (B/b) locus is named for the recessive phenotype, which causes the hair pigment to be brown instead of black. A functional wild type allele (B- genotype) is required to synthesize black pigment. Combined with aa, the bb genotype yields a solid brown (rather than solid black) mouse. Combined with A-, it yields a cinnamon (rather than agouti) mouse. The official designation for this locus is now Tyrp1 for tyrosinase-like protein 1. Just as is the case for human genetic loci, mouse loci are also being renamed as the exact molecular identities of their coding sequences become known.
The coat color (C/c) locus must be in the dominant form (C-) to obtain any coat color at all. The recessive phenotype (cc) is albino, with no coat pigmentation of any sort. The official name for this locus is now Tyr for tyrosinase. The absence of a functional tyrosinase enzyme blocks all pigment production prior to the branch points in the biosynthetic pathways that lead to various colors of pgiment, such as yellow and black (or brown).
Epistatic effect of cc genotype: The cc recessive (albino) phenotype exhibits complete epistasis over any combination of A and B genes. Thus mice with a cc genotype are always white irrespective of the other coat color genes they carry. As an example, the F1 progeny from a cross of a true-breeding black mouse (aaCC) and a true breeding white strain that is wild-type at the agouti locus (AAcc) will all be agouti (A-C-) and the F2 progeny will be:
Trihybrid cross of agouti, brown and coat color genes: The overall interplay of these three genes causes a complex pattern of inheritance of coat color that can be analyzed readily through use of the product rule, and diagramed through use of the branching line approach. Remember that cc is totally epistatic over any other effects, and that in animals that have a C- genotype, the A and B genes interact to yield agouti (A-B- phenotype), black (aaB- phenotype), cinnamon (A-bb phenotype), and brown (aabb phenotype). The expected phenotypic combinations from AaBbCc x AaBbCc would be as follows:
If you would like to learn more about these loci, or other genetic loci in the laboratory mouse, you may wish to visit the web site entitled Mouse Genome Informatics that is maintained by the Jackson Laboratories with support from the National institutes of Health. You can find a brief introduction describing how to use that web site on our course web page Other Genetics Links.
Additional loci: Last year, there was report that could be pulled up on the Mouse Genome Informatics web site that listed a total of 92 different loci that in some way influence coat color or the distribution of pigmented and white areas in the laboratory mouse. The web site has been reorganized in a manner that no longer lumps all of these loci into a single category. However, to see related but slightly shorter lists, you can go to the web site's query form, scroll down to "Phenotype", and type "color" into the search box, then scroll back up to "Retrieve" and click on it. This will bring up a list of 63 loci that influence color in some way. Similarly, a search for "pigment" brings up 77 loci. In either case, you can then click on the abbreviation for a specific locus, and when the page for the locus cames up, you can click on phenotype to obtain more information about the specific phenotypic effects of that locus. For more information on the A, B, and C loci discussed above, examine a, Tyrp1 and Tyr, respectively. For each locus. you will find that there are many diverse alleles, going far beyond the fully dominant and fully recessive alternatives that have been discussed in these notes.
Brown, scarlet, and white eyes: The deep red color of the eyes of wild type Drosophila is caused by the presence of two different types of pigment. The mutant strain brown (bw/bw) fails to make a scarlet pigment, and has brown eyes because only brown pigment is made. It is confusing, but very important to remember that brown-eyed flies fail to make scarlet pigment. Similarly, the mutant strain scarlet (st/st) has scarlet eyes because they fail to make brown pigment (this mutation blocks another step in the pathway (figure 6.13) that is also blocked by the vermillion and cinnabar mutations). Thus, the locus called brown codes for an enzyme involved in the synthesis of scarlet pigment, and the locus called scarlet codes for an enzyme involved in the synthesis of brown pigment. Note that naming genetic loci for the phenotypic effects of recessive (loss of function) mutations can become very confusing, as illustrated by this case.
Loss of both pigments results in white eyes: In the double mutant strain (bw/bw, st/st), neither pigment is made and the eyes are white. Thus, the F2 of a cross of a true-breeding brown-eyed fly and a true-breeding scarlet-eyed fly will yield 9/16 wild type, 3/16 brown-eyed, 3/16 scarlet-eyed, and 1/16 white-eyed. As will be examined in more detail in the lecture on sex linkage, a loss of function mutation in the sex-linked genetic locus, designated w for white eyes, can have the same phenotypic effect as the simultaneous presence of homozygous b/b and st/st mutations, both of which are autosomal.
Penetrance: For some genotypes, the expected phenotype is not always expressed. This phenomenon is referred to as penetrance, which is defined as the fraction of individuals with a particular genotype that express at least some degree of the expected phenotype. As an example, a genetic predisposition to cancer is not expressed in everyone with the genotype. Thus, for genotypes known to have less than 100% penetrance, it is not safe to assume that absence of the phenotype means absence of the genotype. The textbook points out that at least some cases of lack of penetrance may reflect epistasis. However, there are other cases in which the cause for non-penetrance is not known, such as the skipping of a generation that can sometimes occur in dominant polydactyly (Figure 13.26) and in dominant type D brachydactyly (figure 13.27).
Expressivity: Among those individuals that express a phenotype to some extent, the intensity of the expression is referred to as expressivity. The textbook cites the variable expressivity of purple spotting on the seed coats of peas as an example (figure 13.24). Another cited example is type D brachydactyly, which affects only one thumb in some individuals and both thumbs in others, as well as sometimes being non-penetrant (figure 13.27). Yet another good example is the expression of genes that cause white spotting in mice and various other animals, such as cats, in which the area of pigmentation loss can vary greatly from one individual to another. Expressivity is also very much influenced by environmental factors, such as temperature of various body parts, as will be discussed below.
Temperature effects: Many mutations are known that confer temperature sensitivity on phenotypic expression. In the most common cases, a gene product is functional at a lower temperature and loses function at a higher temperature. This generally reflects loss of heat stability of a functional protein (usually an enzyme) due to an amino acid substitution. Such mutations are referred to as conditional or temperature sensitive. The temperature at which normal function is retained is referred to as permissive, and that at which function is lost as restrictive (or non-permissive). The ability to produce temperature sensitive mutations makes it possible to assemble collections of mutations that are lethal at restrictive temperatures, but capable of being maintained in a viable and normally reproductive state at permissive temperatures. This allows genetic analysis of many processes that are essential for survival and could not otherwise be studied because of the lethality of loss-of-function mutations.
Siamese cats and Himalayan rabbits: Two common examples of temperature sensitivity that do not involve loss of viability are seen in Siamese cats and Himalayan rabbits. In both cases, pigment production is temperature sensitive and occurs only in extremities such as ears, nose and limbs that have a lower surface temperature than the main part of the body (see figure 13.25 in the textbook for the Himalayan rabbit).