Direction of migration
Figure 8.1. Protein electrophoresis of genotypic variation associated with A and S alleles at hemoglobin ,6-chain locus in humans. The buffer environment is such that the hemoglobin molecule is disassociated into its component a and , chains and the , chains have a charge difference depending upon the animo acid they have at the sixth position of the amino acid chain.
tetramer consisting of two a-globin chains and two ,-globin chains. Under the appropriate buffer conditions, the hemoglobin tetramer can be disassociated into its component polypep-tide chains. When protein electrophoresis is performed on blood samples in the proper pH and buffer environment, the ,-hemoglobin chains display different electrophoretic mobility phenotypes in individuals differing in their genotypes with respect to the S and Aalleles, as shown in Figure 8.1 (the a-globin chains, which have a distinct electrophoretic mobility from both types of , chains, are not shown). As can be seen in that figure, the SS and AA homozygous genotypes each produce a single band, but with distinct phenotypes of electrophoretic mobility. However, note that the electrophoretic phenotype of the AS heterozygotes is the sum of the phenotypes of the two homozygotes. Thus, the phenotype associated with each allele in homozygous conditions is fully expressed in the heterozygote as well. In this case, we would say that the A and S alleles are codominant.
Sickle cell anemia gets its name from the fact that the red blood cells (the cells that carry the hemoglobin molecules) will distort their shape from their normally disk-shaped form to a sickle shape (Figure 8.2) under the environmental conditions of a low partial pressure of molecular oxygen (O2). Hemoglobin is the molecule that transports oxygen from the lungs to the tissues throughout our bodies. When the oxygen is released by the hemoglobin molecule due to a low partial pressure of oxygen in the ambient environment, an allosteric change occurs in the hemoglobin molecule. This three-dimensional change causes the valine in the 6S-globin to protrude outward, where it can stick into a pocket in the three-dimensional structure of an a chain on an adjacent hemoglobin molecule. The hemoglobin molecules are tightly packed together in the red blood cells, making such a joining of ,S-globin to a-globin likely. Indeed, long strings of these joined hemoglobin molecules can assemble, and these strings in turn distort the shape of the red blood cell, leading to the trait of sickling
under environmental conditions of a low partial pressure of oxygen. Such environmental conditions occur in the capillaries when oxygen is taken up by a peripheral tissue. Such conditions also occur at high altitude, or during pregnancy (fetal hemoglobin has a higher oxygen affinity than adult hemoglobin, allowing the fetal blood to take oxygen from the mothers blood across the placenta). Because both AS and SS genotypes have jS-globin chains in their red blood cells, both of these genotypes show the sickling trait under the appropriate environmental conditions. Therefore, with respect to whether red blood cells sickle or not, the S allele is dominant.
In SS homozygotes, there are no j A-globin chains to disrupt the strings of joined hemoglobins, so the strings tend to be longer in SS individuals than in AS individuals. As a consequence, the distortion of the normal shape of the red blood cell tends to be more severe in SS individuals than in AS individuals under the environmental conditions of a low partial pressure of oxygen. Indeed, the distortion can be so severe in SS individuals that the red blood cell ruptures, losing its hemoglobin molecules and leading to anemia. The highly distorted red blood cells also pass poorly through the narrow capillaries, which is one of the places with the appropriate environmental conditions to induce sickling. The anemia and the inability of the distorted cells to move easily through the capillaries can lead to wide spectrum of phenotypic effects (Figure 8.3) known collectively as sickle cell anemia.
As Figure 8.3 shows, sickle cell anemia is actually a complex clinical syndrome with multiple phenotypic effects (pleiotropy) and much variation in expression from one individual
Low partial pressure of O2 Abnormal hemoglobin
Sickling of red blood cells
Sickling of red blood cells
to the next. Indeed, the symptoms vary from early-childhood death to no clinical symptoms at all (the absence of clinical symptoms in some individuals will be discussed in Chapter 12 but is due in part to epistasis with other loci). However, regardless of the exact degree of expression, the clinical syndrome of sickle cell anemia is found only in SS individuals. Therefore, with respect to the phenotype of sickle cell anemia, the S allele is recessive.
The AS and SS genotypes show resistance to falciparum malaria (Friedman and Trager 1981), one of the most lethal forms of malaria in humans. The malarial parasite enters the red blood cells of its host. A cell infected by the falciparum but not by the other malarial parasites develops knobs on its surface, which leads to its sticking to the endothelium of small blood vessels. In such sequestered sites, sickling takes place because of the low oxygen concentration in AS and SS individuals. The infected red cell is also more acidic than the uninfected cell, a pH environment that enhances the rate of sickling. The spleen generally removes the red blood cells with the distorted sickle morphology before the parasite can complete its life cycle, leading to the phenotype of malarial resistance. Obviously, this phenotype can only be expressed under the environmental conditions of malarial infection. Because the genotypes AS and SS show this type of resistance to malaria, the S allele is dominant for the phenotype of malarial resistance.
Phenotype of Health (Viability)
The A and S alleles influence how healthy an individual is, particularly with regard to viability, the ability of the individual to stay alive in the environment. First consider an environment that does not have falciparum malaria. In such an environment, SS individuals have a substantial chance of dying before adulthood due to hemolytic anemia and other complications, as shown in Figure 8.3. This is particularly true in areas that have poor health care. Because AA and AS individuals do not suffer from sickle cell anemia, the S allele is recessive for the phenotype of health in a nonmalarial environment.
Now consider health in a malarial environment. The SS individuals have poor health because they suffer from sickle cell anemia, but the AA individuals also have poor health because of falciparum malaria and have a high probability of childhood death. However, the AS individuals do not suffer from sickle cell anemia, and they have some resistance to malaria. Therefore, in an environment with falciparum malaria, the S allele is overdominant with respect to the phenotype of health because the AS heterozygotes have superior viability to either homozygote class.
Note that the S allele can be dominant, recessive, codominant, or overdominant depending upon which phenotype is being measured and the environment in which the measurement is made. Although we frequently use such expressions as a "dominant allele" or "recessive allele," such expressions are merely a linguistic short-hand for describing the genotype-phenotype relationship in a particular environmental context. Dominance, recessiveness, and so on, are NOT intrinsic properties of an allele. Context is always important when dealing with the relationship between genotype and phenotype.
NATURE VERSUS NURTURE?
Does nature (the genotype) or nurture (the environment) play the dominant role in shaping an individual's phenotype? From premise 3, we can see that this is a false issue. Phenotypes emerge from the interaction of genotype and environment. It is this interaction that is the true causation of an individual's phenotype, and it is meaningless to try to separate genotype and environment as distinct causes for the individual's phenotype. However, in population genetics, we are often concerned with a population of individuals with much phenotypic variability. Accordingly, in much of population genetics our concern centers on causes of phenotypic variation among individuals within the deme rather than the causation of any single individual's phenotype. Causes of phenotypic variation in a population are quite distinct from causes of individual phenotypes, and the nature/nurture issue is limited only to causes of variation. We will illustrate these statements by considering yet another simple Mendelian genetic disease: phenylketonuria (PKU).
The enzyme phenylalanine hydroxylase catalyzes the amino acid phenylalanine to tyro-sine and is coded for by an autosomal locus in humans. Several loss-of-function mutations have occurred at this locus (Scriver and Waters 1999), and homozygosity for loss-of-function alleles is associated with the clinical syndrome known as phenylketonuria. Let k designate the set of loss-of-function alleles and K be the set of functional alleles at this locus. Because kk homozygotes cannot catalyze phenylalanine, they have a buildup of phenylalanine, a common amino acid in most foods. The degradation products of phenylalanine, such as phenylketones, also build up in kk homozygotes. The phenylketones are typically found at high levels in the urine of the kk homozygotes, an easily scored phenotype that gives the syndrome its name. However, there are other phenotypes associated with this syndrome. For example, kk homozygotes tend to have a lighter skin color than most individuals that share their ethnic background because one of the main pigments in our skin, melanin, is synthesized from tyrosine, which cannot be produced from phenylalanine in kk homozygotes. However, the reason why PKU has attracted much attention is the tendency for kk homozygotes to suffer from mental retardation. As with sickle cell anemia, there is tremendous heterogeneity in the phenotype of mental ability among kk homozygotes that in part is due to epistasis with other loci (Scriver and Waters 1999). However, we will ignore epistasis for now and just treat PKU as a single-locus, autosomal recessive genetic disease.
The primary source of phenylalanine is our diet. The kk homozygotes typically have normal mental abilities at birth. While in utero, the kk homozygote is not eating but is obtaining its nutrients directly from the mother. Typically, the mother is a carrier of PKU with the genotype Kk, which means that she can catalyze phenylalanine to tyrosine. After birth, the kk homozygote cannot metabolize the phenylalanine found in a normal diet, and mental retardation will likely soon develop. If a baby with the kk genotype is identified soon after birth and placed on a diet with low phenylalanine, the baby will usually develop a normal level of intelligence. Thus, the same kk genotype can give radically different phenotypes depending upon the dietary environment. Because of the responsiveness of the kk genotype to environmental intervention, many countries require genetic screening of all newborns through a simple urine test to detect the kk homozygotes (Levy and Albers 2000). The PKU screening program has been successful in greatly reducing the incidence of mental retardation due to kk genotypes.
Individuals who are kk are generally advised to maintain a low-phenylalanine diet throughout their life. However, phenylalanine is such a common component of most protein-bearing foods that such diets are highly restrictive and more expensive than normal diets. Moreover, the beneficial effects of the low-phenylalanine diet are strongest in children. Once the brain has fully developed, kk individuals often do not perceive much of an impact of diet on their mental abilities. As a result, compliance with the diet tends to drop off with age. Note that the nature of the interaction of genotype (kk) with environment (the amount of phenylalanine in the diet) shifts with ontogeny (development) of the organism. Thus, genotype-by-environment interactions are not static even at the individual level.
Prior to the successful screening program, few kk women reproduced due to their severe retardation, but with treatment, many kk women married and had children. However, many of these women were now eating a normal diet and hence had high levels of phenylalanine and its degradation products in their blood. The developing fetus, usually with the genotype Kk, which typically develops normally, was now exposed to an in utero environment that inhibited normal brain development. Such Kk children of kk mothers on a normal diet were born with irreversible mental retardation. Is the phenotype of mental retardation due to nature or nurture in this case? Obviously, both are important. One cannot predict the phenotype of an individual on the basis of genotype alone; the genotype must be placed in an environmental context before prediction of phenotype is possible. Thus, what is inherited here is not the trait of mental retardation but rather the response to the dietary and maternal environments.
Now consider the disease of scurvy. Ascorbic acid (vitamin C) is essential for collagen synthesis in mammals. Most mammals can synthesize ascorbic acid, but all humans are homozygous for a nonfunctional allele that prevents us from synthesizing ascorbic acid. As a result, when humans eat a diet lacking vitamin C, they begin to suffer from a collagen deficiency, which leads to skin lesions, fragile blood vessels, poor wound healing, loss of teeth, and eventually death if the vitamin-deficient diet persists too long. Thus, humans have an inherited response to the dietary environment that can lead to the disease of scurvy.
Both scurvy and PKU therefore have a similar biological causation at the individual level. Both diseases result from the way in which an individual homozygous for a loss-of-function allele responds to a dietary environment. Yet, PKU is typically said to be a "genetic" disease, whereas scurvy is said to be an "environmental" disease. Phenylketonuria is considered a genetic disease because although the disease arises from the interaction of both genes and environment, the environmental component of the interaction is nearly universal (phenylalanine is in all normal diets) whereas the genetic component of the interaction, the kk genotype, is rare. As a consequence, when PKU occurs in a human population, it is because the person has the kk genotype since virtually all of us have a diet that would allow the PKU response given a kk genotype (at least until the screening program). Hence, the phenotype of PKU is strongly associated with the kk genotype in human populations. The condition of scurvy is also the result of an interaction between genes and environment, but in this case the genetic component of the interaction is universal in humans. However, the environmental component of the interaction of having a diet without sufficient amounts of ascorbic acid is rare. Therefore, the phenotype of scurvy is associated with a diet deficient in vitamin C in human populations.
When we ask the question, what causes PKU or scurvy, our answer is that an interaction of genes with environments is the cause of these diseases. However, when we ask the question, what causes some people to have PKU or scurvy and others not, we conclude that genetic variation is the cause of phenotypic variation for PKU whereas environmental variation is the cause of phenotypic variation for scurvy. As the PKU/scurvy example illustrates, the interaction of genes with environment creates a confoundment between frequency and apparent causation in a population of phenotypically variable individuals. When causation at the individual level arises from an interaction of components, then the rarer component at the level of the population is the one with the stronger association with phenotypic variation. Scurvy is an environmental disease because the dietary environment is rare but the genotypic component is common; PKU is a genetic disease because the dietary environment is common but the genotypic component is rare.
The dependency of causation of variation upon frequency in a population is illustrated by a hypothetical example in Table 8.1 in which a disease arises from the interaction of two independently varying components. In particular, the disease only occurs when the first component has state A1 and the second component has state B1. In this population, state A1 is relatively common, having a frequency of 0.9, and state B1 is relatively rare, having a frequency of 0.1. As shown in Table 8.1, the frequency of the disease in the population is given by the product (0.9)(0.1) = 0.09, and the remaining 91% of the population has no disease. Now suppose that a survey is done in this population on component A. Given that an individual has state A1, then that individual will only show the disease when the
Was this article helpful?
Many women who have recently given birth are always interested in attempting to lose some of that extra weight that traditionally accompanies having a baby. What many of these women do not entirely realize is the fact that breast-feeding can not only help provide the baby with essential vitamins and nutrients, but can also help in the weight-loss process.