HEREDITY AND EVOLUTION

Heredity and evolution The gene in populations In the study of heredity the first question that arises is how the genotype of an individual is formed from the constituents of the genotypes of his parents. This is the genetics of individuals or basic genetics. One may also inquire how the genotype in a fertilized egg cell influences the developmental pattern of the organism and thus realizes its potentialities. This is developmental genetics. An individual, at least an individual of a sexually reproducing species, is not, however, biologically complete in itself. Its biological role is actualized through its membership in a reproductive community, a Mendelian population. A Mendelian population consists of individuals among whom matings may or do occur. An individual is mortal and temporary; a Mendelian population has a continuity through time. The genetic processes in Mendelian populations are the subject matter of population genetics. The gene pool A Mendelian population is said to have a gene pool. The gene pool is the sum total of the genes carried by the individual members of the population. The gene pool also continues through time. The genes of the individuals of the generation now living come from a sample of the genes of the previous generation; if these individuals reproduce, their genes will pass into the gene pool of the following generations. The Mendelian population and its gene pool in humans have a very complex structure. Individuals born and living close together are more likely to meet and to mate than those living far apart. In a widely distributed species such as Homo sapiens, the likelihood of mating of individuals born on different continents was, until the development of modern means of travel, very small. The gene pool of the human species is, accordingly, divided into the smaller gene pools of races and populations living in different regions. Aside from the geographic divisions, there are also linguistic, religious, social, economic, and educational barriers that break the gene pools into further, often overlapping, subdivisions. The smallest subdivision is referred to as an isolate or panmictic unit; it consists of a relatively limited number of persons (or animals or plants) that may be regarded as potential mates. Few of these divisions may be sharp enough to decide where one gene pool subdivision ends and the other begins, and yet these subdivisions are biologically meaningful. A biological species, in sexually reproducing organisms, is defined as the most inclusive Mendelian population. The gene pool of Homo sapiens is an entity the limits of which are not in doubt, since no gene exchange between the human and any other related species takes place. Nor does the intraspecific differentiation impair the unity. There may never have been a marriage of, for example, an Eskimo and a Melanesian, but genetic communications between the Eskimo gene pool and the Melanesian gene pool occur through the chains of geographically intermediate populations. A genetic change arising anywhere in the world, if favourable, may spread throughout humanity. This is how genetic changes may have transformed the ancestral prehuman species into the present one. This genetic unity makes any genetic damage (e.g., that caused by exposure to high-energy radiation) a concern of all people, regardless of whether the damage is inflicted more heavily on one portion of the human population than on another. Heredity and evolution Selection as an agent of change Natural selection and Darwinian fitness Sexual reproduction under simple (Mendelian) inheritance is a conservative force that tends to maintain the genetic status quo in a population. If a gene frequency is 1 percent in a population, it tends to remain at 1 percent indefinitely unless some force acts to change it. Outside of the laboratory, the most powerful force for changing gene frequencies is natural selection. The carriers of some genes may survive more often or be more fecund than the carriers of other genes. When the carriers of different genes are not equally efficient in transmitting these genes to the succeeding generations, the result is natural selection. When the inequality of the transmission rates of the genes is imposed by human will, the result is artificial selection. In general, the genes that confer on their possessors a superior reproductive efficiency will increase in frequencies from generation to generation, and the reproductively inferior genes will become less frequent. Imagine a population that carries two alleles-A1 and A2-for a particular gene. Suppose that the relative numbers of the surviving progeny left by the carriers of the genotypes A1A1 and A2A2 are in the ratio 1 : 1 - s (the value s is called the selection coefficient). If for every 100 offspring of A1A1 parents only 90 surviving offspring are left by A2A2 parents, then s = 1/10 or 0.1. The heterozygotes, A1A2, may leave as many progeny as A1A1 (if A1 is dominant) or as many as A2A2 (if A2 is dominant) or an intermediate number (if neither is dominant). Alternatively, the heterozygotes may exhibit the quality of hybrid vigour (heterosis); that is, they may be reproductively superior to both homozygotes A1A1 and A2A2. And finally (though this is rare except in species hybrids) the heterozygotes may be at a disadvantage compared to both homozygotes. The situation is simplest if the heterozygotes (A1A2) are equal in reproductive efficiency to one of the homozygotes or intermediate between the two. Whichever gene, A1 or A2, confers a superior reproductive efficiency on its possessors will increase in frequency in the population. The increase will continue generation after generation; given enough time (i.e., enough generations) the more efficient gene will eliminate and supplant the less efficient one entirely. How rapid or slow the gene frequency changes will be depends on the magnitude of the selection coefficients. Table 4 gives several examples for a dominant allele, for a recessive allele, and for no dominance-i.e., for the case in which the fitness of the heterozygote is intermediate between the two homozygotes. The two alleles in the population with which the selection starts are assumed to be equally frequent, p = q = 0.5. The selection coefficients of one (lethal), 0.5, 0.1, and 0.01 are considered, and the gene frequencies after one, two, five, 10, 20, 50, and 100 generations of such selection are given. The homozygotes for a recessive gene allele may not reproduce at all. With natural selection this may happen because they are inviable (lethal) or sterile; with artificial selection the same result is accomplished if the breeder kills them or does not use them as parents. The recessive allele is then opposed by a selection s = 1. Table 4 shows that a gene with an initial frequency of 0.5 will decrease to 0.33 after one generation, to 0.25 after two, and to 0.01 after 100 generations. With weaker selection (s smaller than unity) the decrease of the recessive allele will, of course, be slower. Note, however, that in all cases the frequency change is more rapid when the gene is common than when it is rare. A dominant allele opposed by a selection s = 1 (a dominant lethal) disappears in a single generation, and even weaker selections against dominants are more efficient than similar selections against recessives. A selection in favour of a recessive is, of course, just the reverse of that against a dominant, and vice versa. The frequencies can be read from Table 4 by subtracting the frequencies given from unity. When the dominance is absent, the efficiency of selection is, as shown in Table 4, intermediate between those for recessives and for dominants. A most interesting, and at first sight paradoxical, outcome of selection arises if the heterozygote is superior to both homozygotes. Neither the gene A1 nor A2 is allowed to crowd the other out or to disappear entirely. Instead, a genetic equilibrium is reached, and the population attains the state of the so-called balanced polymorphism. All three genotypes continue to occur in the population, with frequencies dependent on the relative magnitudes of the selection coefficients s1 and s2. This will be true even if one of the homozygotes is seriously incapacitated, inviable, or sterile. The possible importance of this in human populations is considered below. Darwin's description of the process of natural selection as the survival of the fittest in the struggle for life is a metaphor. "Struggle" does not necessarily mean contention, strife, or combat; "survival" does not mean that ravages of death are needed to make the selection effective; and "fittest" is virtually never a single optimal genotype but rather an array of genotypes that collectively enhance population survival rather than extinction. All these considerations are most apposite to consideration of natural selection in humans. Decreasing infant and childhood mortality rates do not necessarily mean that natural selection in the human species no longer operates. Theoretically, natural selection could be very effective if all the children born reached maturity. Two conditions are needed to make this theoretical possibility realized: first, variation in the number of children per family and, second, variation correlated with the genetic properties of the parents. Neither of these conditions is farfetched. Darwinian fitness is sometimes referred to also as the adaptive value or the selective value; these terms are best treated as synonyms, although they may have somewhat different connotations. The Darwinian fitness of a genotype, or of a group of genotypes, is measured as the contribution of their carriers to the gene pool of the succeeding generation, relative to the contributions of other genotypes present in the same population. In the example given above, the Darwinian fitness of the genotype A1A1 was taken to be unity and the fitness of A2A2 as 1 - s or less than unity. The fitness is, of course, subject to change in different environments; the carriers of a genotype A1A1 may leave more surviving progeny than A2A2 in a certain environment, but the reverse may be the case in another environment. Darwinian fitness is reproductive fitness; bodily or mental vigour, health, and energy obviously contribute to this fitness but only insofar as they result in a superior reproductive capacity. Mules, no matter how strong and resistant, must be ranked zero in Darwinian fitness because they are sterile. The emphasis on reproductive success rather than on survival is characteristic of the modern concept of natural selection, as distinguished from the classical one. The difference is not, however, so great as it may seem at first glance; the carriers of a genotype evidently must survive in order to reproduce, and they must reproduce in order to survive in the next generation. Varieties of natural selection There are several kinds of natural selection, rather different in their biological consequences and in their importance to humans. The simplest of them is the normalizing selection, which was already known before Darwin but, of course, not under this name. Normalizing selection counteracts the accumulation in populations of hereditary diseases, malformations, and weaknesses. Suppose that a gene allele A1, the carriers of which have a high Darwinian fitness, mutates to a state A2, which lowers the fitness. If A2 is a dominant lethal or a gene that renders its carriers sterile, then (as shown in Table 4, column s = 1.0) all the A2 mutants will be eliminated in the same generation in which they arise. A new crop of mutants will, of course, appear in the next generation. If, however, the selection is not so completely efficient, some mutant genes will escape its dragnet and will be transmitted to the next generation. That generation will contain all the newly arisen mutants, a part of the mutants that arose in the preceding generation, a smaller part of those having arisen two generations ago, etc. How great a "genetic load" of uneliminated mutants a population can accumulate will depend principally on two factors-how often the mutation arises and how much it lowers the Darwinian fitness. Simple formulas have been worked out to describe the situations that arise. Suppose that a deleterious mutation from A1 (r) A2 occurs at a rate u per generation. Suppose further that the mutant is discriminated against by a selection coefficient s. If, then, the mutant A2 is dominant to the original state, A1, the frequency of A2 in the gene pool will be u/s. If A2 is recessive to A1, its accumulated frequency will be much higher, namely . The reason deleterious recessive mutants are allowed to attain higher frequencies than equally deleterious dominants is simple: a recessive mutant may be carried in many heterozygotes, in which it does not express itself, and is consequently protected, or sheltered, from the weeding-out action of natural selection. With mutants that are neither dominant nor recessive, the accumulation will be intermediate between u/s and . All human populations doubtless carry genetic loads consisting of harmful mutant genes. This cannot be blamed entirely on culture, civilization, or on any other specifically human attributes. Populations of Drosophila flies and of other sexually reproducing organisms also carry genetic loads. The accumulation of the genetic loads is a necessary consequence of the occurrence of mutations, most of which are harmful but not always harmful enough to be eliminated immediately after they are produced. Harmful mutations are accumulated until the numbers of the respective mutant genes become equal to the numbers eliminated by natural selection in the same population. The population is then said to be in the state of "genetic equilibrium." Muller has termed the elimination of harmful mutants "genetic death." Genetic "death" is sometimes cruel, sometimes rather benign. The death of a child from a severe hereditary disease and the genetically conditioned failure to have one more child are both genetic deaths. The higher the mutation rates, the more harmful are the mutants produced and the more frequent are the genetic deaths. In populations that have reached genetic equilibrium, the total number of genetic deaths will be equal to the total number of the mutations subject to normalizing natural selection. A very different form of natural selection is heterotic balancing selection. It occurs when the Darwinian fitness of a heterozygote exceeds the fitness of both homozygotes, a situation mentioned above. Heterotic balancing selection also leads to genetic equilibrium but not to an equilibrium between mutation and the normalizing selection. The balanced polymorphism that is established is due to the selection favouring the heterozygotes against the homozygotes. In a sexually reproducing population the heterozygotes tend, however, to produce a fresh crop of homozygotes in every generation. The maintenance in human populations of the grave hereditary disease sickle-cell anemia is apparently due to this form of selection. The sickling gene (HbS) produces a specific type of hemoglobin, while normal hemoglobin is related to another allele (HbA). Accordingly, the possible genotypes are HbAHbA, HbAHbS, and HbSHbS. The latter individuals are homozygous for the sickle-cell gene and will develop severe anemia. While the condition is not lethal before birth, such individuals rarely survive long enough to exhibit more than minimal fitness in the Darwinian sense of capacity to reproduce. On these grounds it might be concluded that natural selection eventually should drive the frequency of the HbS gene to complete elimination or at least down to the one in 100,000 level of the mutation rate. One would theorize a transient polymorphism; that is, one on the way out, toward fixation of the favoured allele. Evidence, however, seems to contradict theory since, in a number of African tribes living in their ancestral tropical lowlands where the falciparum form of malaria is widespread, the HbS (sickling) gene is very common indeed. On the other hand, the same gene is rare in genetically related but isolated tribes living in highlands that are free of mosquitoes that transmit this type of malaria. This discrepancy between lowland and highland people has led to the hypothesis that the HbAHbS heterozygote is fitter and capable of leaving more offspring than is the homozygous normal HbAHbA in a highly malarious environment. This extra measure of protection is evidently provided by the sickle-cell hemoglobin (HbS), which is detrimental to the malaria parasite. In malarial environments, populations that contain the sickle-cell gene, therefore, have advantages over populations free of this gene. The former populations are in less danger from the ravages of malaria, although they "pay" for this advantage by sacrificing in every generation some individuals who die of anemia. The lethal disease caused by homozygosity for the sickle-cell gene certainly brings about some genetic deaths; it is a part of the genetic load of the populations. But this genetic load, due to a disadvantage of being homozygous for certain genes, is very different from the mutational load controlled by the normalizing selection. The former is maintained by the heterotic balancing selection, while the latter is maintained by recurrent mutation. Another form of natural selection is diversifying, or disruptive, selection. In many discussions and mathematical analyses of selection this simplifying assumption is adopted: that the environment in which a population lives is uniform and that the selection advantages and disadvantages of different genotypes are independent of their frequencies in the population. This simplification, however, flies in the face of reality. Many animals can subsist on a variety of foods; many plants grow on different soils; humans have to fill many different employments, functions, professions, and social roles. It is most likely that some genotypes will be fitter in some environments than they are in other environments. Diversifying selection will then favour different genotypes in different subenvironments, or ecological niches, that occur in the population. A special form of selection occurs in mammals due to the incompatibility of certain maternal genotypes with those of their unborn children. The best studied case in humans is that of a Rhesus-positive fetus in a Rhesus-negative mother. This selection should, theoretically, make the entire population either Rhesus positive or Rhesus negative. For reasons that have not been clarified, it does not appear to be doing this. Another special kind of selection is that due to so-called meiotic drives, disturbances of the Mendelian segregation mechanisms, which result in sex cells carrying certain gene alleles being more or less frequent than expected on a random basis. The last to be mentioned, but in the long run possibly the most important form of selection, is directional selection. Suppose that the climate becomes warmer or cooler, that there appears a new source of food or a new predator or disease, or that there occur some other prominent environmental changes. Some genotypes will, then, become more favourable and others less favourable. Directional natural selection will operate to reconstruct the gene pool of the population in accord with the demands of the new environment. The physical basis of heredity Molecular genetics The data accumulated by the geneticists of the early 20th century provided compelling evidence that chromosomes are the carriers of the genes. But the nature of the genes themselves remained a mystery, as did the mechanism by which they exert their influence. Molecular genetics-the study of the molecular structure of the genes and the methods by which genes control the activities of the cell-provided the answers to these fundamental questions. Much of the information in molecular genetics has come from the study of microorganisms, particularly the bacterium Escherichia coli (a common inhabitant of the human intestine) and its interactions with various bacteriophages. Bacteria have many features that make them especially useful in genetics research. For example, they have an extremely short life cycle, so that many generations can be raised in a brief period of time. Equally important, bacteria have only one basic function-to reproduce. Consequently, their genome is relatively limited. Furthermore, unlike most higher organisms, bacteria are not diploid, so their genome does not include two alleles of each gene. This makes it easy to identify a bacterium that carries a mutant gene, as the effects of the mutation cannot be masked by a normal allele. Although they are not diploid, bacteria can and do occasionally exchange genetic information through a variety of processes. This genetic exchange feature has been important in certain lines of molecular genetics research. Viruses also have advantages in genetics studies. Although they can reproduce only in a living cell, they have the simplest form of genetic material and evidence both genetically controlled properties and the ability to mutate. Because of the relative simplicity of gene action in microorganisms, their study profoundly influenced early understanding of molecular genetics. Studies of the genetics of microorganisms involves the production of specific gene mutations and the examination of their biochemical effects. These studies have permitted the delineation of the metabolic pathways that produced the mutation in the experimental microorganism, as well as the isolation of the large molecules that contain the genetic information. Although there are virtues to bacteria as experimental subjects in genetics research, it should be pointed out that bacteria differ from higher organisms in some rather fundamental ways. In fact, bacteria (along with the cyanophytes, or blue-green algae) are sufficiently distinct as to constitute their own kingdom, the Monera. Monerans, unlike protists, plants, and animals, are procaryotic. This means that their cells lack a true, membrane-enclosed nucleus, the cellular structure that contains the chromosomes in all other organisms (which are known as eucaryotes). Perhaps more important in a discussion of genetics, the bacterial chromosome differs in composition from the chromosomes of eucaryotes, so much so that some authorities prefer to avoid the term chromosome in describing the genetic material of bacteria. In eucaryotes, the chromosomes consist primarily of deoxyribonucleic acid (DNA) and a variety of proteins. Bacterial chromosomes have little protein, which proved to be an important clue in determining the chemical nature of the hereditary substance. Finally, all the progeny of a bacterium are identical, whereas the cell progeny of the fertilized egg of a complex, multicellular organism gives rise to many different tissues and organs whose component cells display specific patterns of different gene activities. This latter process is called differentiation. Heredity and nucleic acids One of the most impressive and spectacular advances of biology in the 20th century was the discovery of the nature of the genetic material. The way information is encoded in the genes has been clarified and much has been learned about the mechanisms that translate this information into the developmental processes of the organism. In 1869 a substance containing nitrogen and phosphorus was extracted from cell nuclei. It was originally called nuclein, but is now known as DNA. DNA is the chemical component of the chromosomes that is chiefly responsible for their staining properties in microscopic preparations. As stated above, the chromosomes of eucaryotes contain a variety of proteins in addition to DNA. The question naturally arose whether the nucleic acids or the proteins, or both together, are the carriers of the genetic information, which makes the genes of the same organism and of different organisms specifically different. Until the early 1950s most biologists were inclined to believe that the proteins were the chief carriers of heredity. Nucleic acids contain only four different unitary building blocks, but proteins are made up of 20 different amino acids. Proteins therefore appeared to have a greater diversity of structure, and the diversity of the genes seemed at first likely to rest on the diversity of the proteins. The evidence that DNA acts as the carrier of the genetic information was first firmly demonstrated by exquisitely simple microbiological studies. One of these seminal studies was performed by the U.S. geneticist Oswald T. Avery and his coworkers in 1944. The background of Avery's study goes back to 1928, to research conducted by Fred Griffith of England, a bacteriologist who was studying the virulence of pneumococci, the bacteria that cause bacterial pneumonia. Griffith knew that the virulence of pneumococci-that is, their ability to cause infection-depends on the presence of an envelope composed of polysaccharides (sugar subunits) surrounding the bacterial cells. When grown on laboratory culture mediums, virulent pneumococci produced large colonies with a smooth, glistening surface. Bacteria from such cultures caused infection in mice. After many transfers to fresh laboratory mediums, however, some of the bacteria lost their polysaccharide envelopes and their ability to infect mice; correlated with these changes was the change to small colonies with rough outlines. The smooth and the rough variants were designated S and R, respectively. Griffith found that mice inoculated with either the living R pneumococci or with heat-killed S pneumococci remained free of infection, but mice inoculated with a mixture of living R and heat-killed S bacteria became infected. He further discovered that living S pneumococcal cultures could be obtained from such animals, proof that the virulent S cells were reconstituted from the mixture inoculated. Some material derived from the dead S bacteria had induced the transformation of R into S strains. Avery and his coworkers showed that the "transforming factor" which conferred the characteristic of virulence upon nonvirulent pneumococci consisted of DNA, which had been transferred from dead virulent pneumococci into living nonvirulent pneumococci. That this newly acquired characteristic was due to a specific genetic activity was demonstrated by its persistence in all of the offspring of the transformed bacteria. The DNA of the dead cells evidently accomplishes the transformation of the living ones by penetrating the wall of the living cell. Once a section of the transforming DNA strand is inside the recipient cell, there apparently occurs a pairing between homologous regions of the bacterial chromosome and the transforming DNA. There must follow breakage and subsequent reunion of the bacterial chromosome and the transforming DNA. Thus a portion of the transforming DNA becomes integrated into the bacterial chromosome. If this model is valid, one would expect that genes located near each other on the transforming DNA would appear together more often in a transformed cell than will genes relatively far apart in the transforming DNA. This expectation has been fulfilled, and the principle has been utilized as a means of mapping the donor-cell chromosome. In further research the principles involved in transformation have been confirmed in a more efficient process involving mammalian cells in culture. By means of a process called transfection, defined pieces of DNA enter the cell nucleus and are incorporated into the DNA, thus replacing a particular genetic deficiency formerly exhibited by the cell. In the early 1950s Alfred D. Hershey and Martha Chase obtained evidence confirming that DNA serves as the physical basis of heredity. In their experiment, Hershey and Chase used a bacteriophage that infects Escherichia coli, a colon bacteria. This bacteriophage (or simply phage) is an ultramicroscopic tadpole-shaped particle, with a hexagonal head, a cylindrical tail, and an end plate with six tail fibres (see virus). The entire outer surface consists of protein, but within the interior space of the head there is DNA. When a phage infects E. coli, it injects its own genetic material into the bacterial cell. The phage genes then subvert the metabolic machinery of the bacterium, causing the host cell to make phage DNA and phage protein. When a new generation of phage particle is ready inside the host, they destroy (lyse) the bacterium. This lysis releases the new phage particles into the medium, where they can attack other bacterial cells. Hershey and Chase prepared two populations of phage particles. In one population the phage protein was labelled with a radioactive isotope; in the other the phage DNA was radioactively labelled. After allowing both populations to attack E. coli cells, the experimenters analyzed the exterior and the interior of the infected cells for the presence of radioactive material. They discovered that the phage protein had remained outside of the host cell, while the phage DNA had been injected into the bacterium. This ingenious research demonstrated that the genetic material of the phage consists of DNA rather than protein. The evidence is now overwhelming that the basic material constituting the gene is fundamentally the same in all organisms: it consists of chainlike molecules of nucleic acids-DNA in most organisms and RNA (ribonucleic acid, a close chemical relative of DNA) in certain viruses. As will be discussed later, the gene no longer stands for a discrete unit of heredity of definite and invariable length but is thought of as an operational entity whose properties are more fluid and depend upon the mode of measurement. The physical basis of heredity When Mendel formulated his laws of heredity, he postulated a particulate nature for the units of inheritance. What exactly these particles were he did not know. Today scientists understand not only the physical location of hereditary units (i.e., the genes) but their molecular composition as well. The unravelling of the physical basis of heredity makes up one of the most fascinating chapters in the history of biology. Chromosomes and genes As has been discussed, each individual in a sexually reproducing species inherits two alleles for each gene, one from each parent. Furthermore, when such an individual forms sex cells, each of the resultant gametes receives one member of each allelic pair. The formation of gametes occurs through a process of cell division called meiosis; it is also known as reduction division, because the amount of hereditary material present in the gametes has been reduced by half. When gametes unite in fertilization, the double dose of hereditary material is restored, and a new individual is created. This individual, consisting at first of only one cell, grows via mitosis, a process of repeated cell divisions. Mitosis differs from meiosis in that each daughter cell receives a full copy of all the hereditary material found in the parent cell. It is apparent that the genes must physically reside in cellular structures that meet two criteria. First, these structures must be replicated and passed on to each generation of daughter cells during mitosis. Second, they must be organized into homologous pairs, one member of which is parcelled out to each gamete formed during meiosis. As early as 1848, biologists had observed that cell nuclei resolve themselves into small, rodlike bodies during mitosis; later these structures were found to absorb certain dyes and so came to be called chromosomes (coloured bodies). During the early years of the 20th century, cellular studies using ordinary light microscopes clarified the behaviour of chromosomes during mitosis and meiosis, which led to the conclusion that chromosomes are the carriers of genes.