Let's take another few steps away from the beauty and simplicity of Mendelian genetics. Kind of the same thing as when you were learning to count, once you had 1 to 100 mastered, they told you there were also thousands, millions and billions! Then it turns out there are numbers in between numbers, like 1.2234. Then negative numbers… We started with single, independent characters, determined solely by the genotype at a single locus. That was the 1 to 100 stage. Then we talked about linked traits and modifications of dominance relationships between alleles at a single locus. Here we cover how different genes at different loci interact with each other and the environment to define phenotypes.
EpistasisOften, one gene can alter the effect of a second gene. Thus, two or more loci interact to determine a single phenotype. Let's look at coat color in mice. Colored bands in individual hairs produce the typical, wild type, brownish-gray coat color in mice. At least two loci affect this trait. At the first locus, the recessive allele a is responsible for albino mice, while its dominant counterpart A allows for wild type color. At the second locus the allele B is dominant and makes the banded agouti pattern (named after a Central and South American rodent related to the guinea pig). Allele b, which is recessive, makes solid colored hairs with no bands, so that the mouse looks black. If the two loci are independent and thus don't interact with one another, then a cross between two heterozygote individuals at both loci (that is, AaBb and thus phenotypically agouti) should produce offspring in the typical 9:3:3:1 ratio of a dihybrid cross. However, the resulting distribution is 9 agouti, 3 black and 4 albino mice. As it turns out, the genotype of the first locus suppresses the expression of alleles at the second locus when homozygous recessive. In other words, an individual that is homozygous a (that is, aa) at the first locus is albino regardless of its genotype at the second locus. The explanation for this is in gene function: the first locus codes for a product involved in the early stages of pigment formation. The genotype aa blocks all pigment formation practically at the beginning of the process, so it doesn't really matter what the genotype is for latter stages.
PleiotropyIn this phenomenon, a gene affects more than a single phenotype. The product of a gene is likely to have more than one function and play a role in the development of many apparently independent characters. Genes are, in a way, like people—everyone has different roles in different aspects of life. It is Tuesday: you feed the cat in the morning, then go to school and ace the physics lab with Mary, win a soccer game against a rival school, and then help your Dad fix the TV. So you had many roles throughout the day, involving different activities and people, and some things like acing the physics lab or fixing the TV got done because of your involvement.
Let's continue with our comparison: if you were a gene, the activities you participated in would be different biological processes or pathways leading to a specific phenotype, and the other people you interacted with, likewise, are other genes associated with the same network. Now, what would have happened if you were sick on Tuesday? Your sister had to feed the cat and thus missed the bus to work; Mary struggled through the physics lab on her own, your soccer team lost the game as their best defender was missing, and then Dad skipped fixing the TV because no one was there to help him move it. All of these different things were affected because you stayed home sick. The same is true for a mutation in a pleiotropic gene: the different networks it belongs to do not function the way they usually do, so that the various phenotypes the gene plays a role in might be modified.
Let's look at some real life examples. Frizzle chickens have unusually shaped feathers: they curl backwards towards the bird's head instead of lying flat and close to the body. This phenotype is due to a single, dominant gene. Researchers Walter Landauer and Elizabeth Upham (1936) noticed that this gene not only modifies feather shape but also affects body temperature, metabolic rates, and the number of eggs produced.
In humans, many genetic diseases are pleiotropic. One of the best-known examples is phenylketonuria (PKU). A single gene codes for the enzyme phenylalanine hydroxylase, which is responsible for converting phenylalanine to tyrosine. Tyrosine plays key roles in various neural and metabolic processes, as well as in skin and hair pigmentation. Mutations at the phenylalanine hydroxylase locus often cause a spectrum of symptoms including mental retardation, seizures, and lighter skin. PKU is so common that babies are routinely screened for it shortly after birth and as long as they maintain a diet low in phenylalanine, they can avoid many of the symptoms (and now you know why there are phenylalanine warnings on bottles of diet sodas!)
Nature Vs. NurtureNot everything about an organism is determined by its genes. Your genes might say your skin is a certain shade, but if you go to the beach or the tanning salon (not that we recommend you do!) it will probably get darker. Not even identical twins, whose genetic make-up is the same, are truly identical. This is because the environment also plays a large role in determining a phenotype. The environment is not only about external factors such as humidity levels or temperature, but also includes internal factors such a pH levels and hormonal state. Some traits are not affected much by the environment: in these cases, a given genotype always produces the same phenotype, like the ABO blood types—if your genotype is IAIA or IAIO, or IOIO your blood type is O and so forth. Other characters are purely environmental: what language you speak, if you vote Republican or Democratic, is mostly a matter of the cultural environment you were raised in.
Most characters have both a genetic and an environmental component. Therefore, a given genotype might result in a different phenotype depending on environmental conditions. These conditions might be external (temperature, nutritional condition, and so on), or internal (sex of the individual, age). This is because the phenotype is the result of interactions among many different factors during development. Thus, small differences in how this process takes place might result in phenotypic variation even though the underlying genotype might be the same. This phenomenon can be described with the concepts of penetrance and expressivity.
Penetrance is a measure of how often the genotype and the associated phenotype occur together in a population. If all individuals in a group where the genotypes are known show the expected phenotype, then the trait is a 100% penetrant. Blood types for instance are 100% penetrant, while traits such as brachydactyly (an autosomal dominant trait that causes shortened index fingers and toes) shows 50 to 80% penetrance: only 50 to 80% of the people carrying the dominant allele present the disorder.
Expressivity describes variation in how a penetrant gene influences the phenotype. So even though the genotype and phenotype match closely, the same genotype generates a range of phenotypes. For example, the gene eyeless in Drosophila always generates eye malformations, but these range from a complete lack of eyes, to smaller, misshapen eyes, to normal eyes.
Coat color in the humble domestic pussy cat is an absolute riot of genetic effects, involving epistasis and pleiotropy (amongst many, many other things). To get a tabby cat, not only do you have to have the right allele for Tabby gene, but you also have to have the right allele for the Agouti gene (epistasis) and the alleles for the white coat color gene can cause congenital deafness and affect eye color (pleiotropy)!