This brief review of basic genetic concepts in a context of how we learn, both as scientists and as students, is intended to focus the scope of what the following experiments are intended to teach.
The two fundamental aspects of genetics are gene action and heredity. The genes, of course, are the elemental units of genetic information; their actions, or expression, result in all the physiological processes that make up an organism. Although the genes can also be considered as the elemental units of heredity, this is not precisely true. In a sense, the individual base pairs of the DNA are the elemental units of genetic transmission, and by sophisticated techniques one can demonstrate the inheritance of portions of genes. But for the basic concepts of heredity that everyone should understand, the gene is the fundamental element. It is the identity of the unit of function and a unit of heredity that is important.
The most useful working definition (or perhaps "talking" definition) of a gene is that it is the unit of genetic information, encoded in DNA (or RNA in the case of some viruses, including those that cause AIDS), that specifies the composition of either a protein or an RNA molecule. A comprehensive, formal definition is difficult and even counter-productive to agree upon. If you understand how the genetic processes work, then you understand what genes are. Dealing in concrete terms with important examples of different kinds of genes and what they have in common can guide students toward an appreciation of the abstract concept. Starting with a comprehensive, and necessarily vague, definition can only obscure the simplicity and usefulness of the concept: a gene denotes a piece of information, coded in nucleic acid, that can be inherited. Understanding this requires understanding the basic processes by which the information encoded in nucleic acids is transmitted from one generation to the next, and how that information is expressed as physiological processes. When students learn this they learn the basis of the scientific discipline of genetics.
Genetics is unique among the sciences because in no other case is the scientific method so integral to the way science is taught. Usually, we teach only the results or findings of science, because we cannot teach how those results were achieved. In other scientific disciplines the experimental methods are so complex, the data are so extensive, and reducing and analyzing them is so tedious, that teaching them at the junior high and secondary school levels is neither practical nor desirable. Genetics is different. The study of genetics is the study of how scientists -- geneticists -- know what they know. In teaching genetics, we can really teach scientific method. In genetics, the results and the way the results are achieved need not -- indeed, cannot -- be separated.
The key is to realize that basic genetic concepts are operational definitions, meaning that genetic facts are defined by the operations we use to observe and demonstrate them. The experiments described in this book are designed to allow students to observe these operations for themselves and thereby to develop genetic concepts first-hand.
So, how do we define "the gene" operationally? In fact, we don't. We define specific, individual genes, operationally, by their genotypes (heredity) and phenotypic traits (physiological expression). The genotype is the information coded in the DNA, while phenotype refers to the describable, heritable characteristics of an organism, such as the color of some part of it. We could use virtually any describable characteristic, but some are more useful than others. We define the gene or genes that determine a phenotypic trait through controlled experiments.
The mainstay of genetics is the controlled experiment, where we change only one variable at a time, so that we can identify the change in the outcome with a change in a variable. But in genetics one can often conduct several controlled experiments simultaneously, in parallel, which is much more fun than doing them one at a time. We use experimental organisms that differ by a small number of well-defined, single-gene traits. By using mutants, we generate strains that have one or the other of two alternative forms, or alleles of a gene. For example, we will use two mutant genes (ade1 and ade2) that result in the cells turning red under some conditions (Roman 1956). The alternative alleles of these genes are the non-mutant forms (ADE1 and ADE2), which result in cream-colored cells. The appearance of the red color defines the presence of the mutant allele in a particular strain, and the absence of red color usually defines the absence of the mutant allele, and therefore, the presence of the non-mutant allele.
When we have succeeded, through controlled experiments, in identifying a unit of heredity with a phenotype, we have operationally defined the gene for that trait. The genotype seems more abstract, and therefore more vague, than the phenotype. Experimentally, observing the genotype directly has always been more difficult. However, since molecular biologists can now isolate genes and determine their exact nucleotide sequences, the genotype has become more concrete. These methods are too complicated to allow students to observe the genotype first hand, so at this point we are forced to revert to teaching the results, or using models and simulations. From experiments on inheritance, in which we observe patterns of transmission of traits from one generation to another, we can develop operational definitions of organizational units of genetic information, which we call genes and chromosomes. These, too, can then be identified with physical structures.
We mark individual genes of interest with mutations to distinguish them from the thousands of genes that make up the genome of an organism. Although hundreds of genes have been defined by mutations in yeast, we need only a few of them to teach the fundamentals of genetics, so we have chosen some that are particularly instructive and easy to work with.
The genes that determine the mating type of haploid strains are especially important and are easy to study. In this case we don't need to worry about getting mutants because two alternative alleles of the MAT gene -- MATa (or just a ) and MATà> (or just à>) -- determine the two opposite mating types (Herskowitz & Oshima 1981). They are defined, operationally, by their ability to mate with each other. Therefore, we can determine the mating type of a haploid strain by a mating-type test -- determining if it can mate with one or the other of two strains whose mating types we know. Strains used to define particular alleles of genes in this manner are called "tester" strains.
When haploids mate they produce a diploid that is heterozygous for mating type. This a/à> diploid cannot mate with either mating type, but it can sporulate, which is something that haploids cannot do. When a diploid does sporulate it produces two spores of each mating type, which demonstrates that a and à> are indeed determined by two alternative alleles of a single genetic locus.
The mating-type alleles are unique in that neither is simply a mutant form of the other. Each is a functional gene, but normally each haploid cell expresses only one of the two. In this way they behave as alleles. The phenotype of the heterozygous diploid is different from either of the two haploids. From extensive genetic, physiological, and biochemical studies we know that the mating type alleles are control genes that determine whether or not other genes are expressed. Another advantage of yeast is the ease with which these regulatory genes can be studied.
The easiest genes to study by mutation are those involved in biosynthesis of small molecules such as amino acids, vitamins, and the purines and pyrimidines that are the building blocks of nucleic acids. The fact that yeast do not require any organic molecules for growth except a carbon source and one vitamin means that they can manufacture these other compounds from the carbon and nitrogen sources. Thousands of enzymes, each one a highly specialized catalyst responsible for a single chemical transformation, carry out this process of biosynthesis. They act cooperatively in production-line processes called pathways. Step by step the raw materials are converted from one compound to another to form each of the building-block compounds that the cell requires. The cell uses these small molecules (monomers) to build the large macromolecules (biopolymers) such as proteins, nucleic acids, and complex lipids and carbohydrates that are the structural and functional components of the cell (Broach 1981; Jones & Fink 1982; Mortimer & Schild 1981a; Mortimer & Schild 1981b).
Many of these small molecules are essential for growth, but they need not be made by the cell itself. If they are present in the growth medium the cell will take them in and use them instead of making its own. If a mutation occurs in one of the genes that codes for an enzyme involved in the biosynthesis of a particular small molecule, then that enzyme may be inactive, and the cell will be unable to make that product. That mutant cell, and all of its progeny, will be able to grow only on medium that contains the required product as a nutrient. A mutant strain that requires a new nutrient for growth is called an auxotroph; the strain that does not require that nutrient is a prototroph. In practice, we call a particular strain auxotrophic or prototrophic for a particular compound, as a shorthand way of saying it has, or doesn't have, a particular growth requirement. Once an auxotrophic mutant has been isolated and characterized, the presence or absence of the mutant gene can be defined by whether or not the cell requires that nutrient in the growth medium. However, a requirement for a particular nutrient does not necessarily define a particular gene because many genes are involved in the biosynthesis of each compound. Clearly, mutations in any of the genes in the same pathway will result in the same requirement. Using additional tests to distinguish mutations in one gene from another is necessary in this case. Furthermore, some genes are involved in the biosynthesis of many different compounds, while others are specific for only one. The latter type are obviously easiest to work with because each mutation will result in a single new growth requirement.
In our experiments we will use two types of auxotrophic mutants: mutants that require adenine and mutants that require tryptophan.
One of the first groups of auxotrophic mutants isolated from yeast were those that required adenine (Roman 1956). This compound is one of the bases which form the building blocks of nucleic acids, but it is also a component of adenosine triphosphate (ATP), which plays a central role in energy metabolism. There are several genes in yeast that are necessary for cells to make their own adenine. A mutation in any one of these gives the cell a requirement for adenine in its growth medium. Mutations in two of these genes are especially useful, because, in addition to requiring adenine, their colonies develop a pink or red color. These were the first two adenine-requiring mutants discovered, so they are called ade1 and ade2. Many of the genetics experiments in this guide make use of the properties of these mutants, so you may wish to read more about them in A Closer Look at...Adenine-Requiring Mutants.
We use these three terms to denote different ways that genes interact to produce a phenotypic trait, such as red or cream colony color. When a haploid strain that carries the mutation responsible for the red color is crossed to a haploid strain that carries the non-mutant form of that gene, the diploid formed does not produce the red color. This experiment operationally defines the non-mutant allele as the dominant form and the mutant allele, which results in red color in the haploid, as the recessive form. At the molecular level, we can understand dominance as the production of a functional gene product by the non-mutant allele and recessiveness as the failure of the mutant allele to produce a functional product or any product at all. One should note that dominance and recessiveness describe relationships between alleles -- different forms of the same gene. Epistasis, on the other hand, denotes a relationship between different genes. If a cell carries a functional allele of the ADE1 gene, for example, and a mutant allele of the ade2 gene, that cell will have the red phenotype, characteristic of the mutant ade2 allele, rather than the white phenotype characteristic of the ADE1 allele. We say that the ade2 allele is epistatic to the ADE1 allele. It would not be correct to say that ade2 is "dominant" to ADE1, because they are not allelic. On the other hand, we can also say that respiration-deficient mutants are epistatic to ade1 and ade2 mutants with respect to red pigment formation.
Some alleles are neither dominant nor recessive. If the phenotype of a heterozygous diploid -- one that carries different alleles of a gene --is intermediate between the respective homozygous phenotypes, we call them codominant. In some cases, however, such as with mating type, the phenotype of the heterozygous diploid is entirely different from either of the homozygous phenotypes. The concept of dominance does not apply at all here. We call this type of interaction complementary, to denote that each gene is producing a different functional product, and the phenotype results from the interaction of the two products. This is analogous to the case we have already discussed, in which a strain that has a mutation in one of the red adenine genes (for example, ADE1 ade2) was crossed with one having a mutation in the other red adenine gene (ade1 ADE2). The diploid has the normal cream-colored, adenine-independent phenotype because the two mutants have complementary genotypes. In this case, the complementation is between different genes. Many forms of complementary gene expression occur in cells. The term complementation denotes a type of interaction that is operationally defined, rather than a particular molecular mechanism.
In the process of deriving operational definitions of the genes we study, we give them useful names (Sherman 1981; Mortimer & Schild 1981a; Mortimer & Schild 1981b). The form of a gene's name tells us some of its characteristics. Each name begins with a three-letter symbol that describes its phenotype. The dominant alleles are always written in all-capital letters and recessive alleles are written in lower-case letters. Different genes that have the same phenotypes are distinguished by a number that follows the three-letter symbol without a space. If we need to distinguish between different recessive alleles, we use yet another number separated by a dash. All the genes that code for enzymes in the AMP biosynthetic pathway are given the symbol ade or ADE, corresponding to the recessive and dominant alleles, respectively. We call mutant alleles of the two "red" genes ade1 and ade2, and their wild-type forms ADE1 and ADE2. The alternative mating-type alleles are called MATa and MATà>, but they are usually abbreviated to simply a and à>. In printing or with word processors it is conventional to put gene symbols in italics to distinguish them from phenotype descriptions. In typing, gene symbols are underlined.
Up to this point we have defined the genes as units of function, but by observing how they are transmitted from one generation to another, we can also study genes as units of inheritance. From extensive genetic and molecular studies we know that the genes of yeast, and most other organisms, are DNA: sequences of nucleotide pairs that code the information carried by the gene. We also know that the genes are strung together in long chains, called chromosomes. In our yeast, Saccharomyces cerevisiae, there are 16 different chromosomes.
One of the most fundamental properties of living organisms is their ability to reproduce. Since the phenotype is determined by the genotype, reproduction of the organism requires reproduction of all the genes that determine the phenotype. This total set of genes is called the genome. The genome of each cell is duplicated prior to each cell division and the two copies are then divided between the progeny cells. This happens both in mitotic and meiotic cell divisions, but in the two cases the outcome is different because the function of the cell division is different.
Mitotic cell division occurs during assimilative growth (commonly called vegetative growth, a term which is not technically correct for fungi), wherein the cells reproduce to form nearly identical progeny. In yeast, and other undifferentiated organisms, every cell is virtually identical; they are all products of simple mitotic replication. In higher, differentiated plants and animals, the cells not only divide, but they differentiate into the myriad of different types of cells that make up the organism. A highly simplified example of this process of differentiation occurs, reversibly, in yeast during mating. The pheromone-induced change of a cell from a dividing unit to a gamete (shmoo) is analogous to hormone-induced differentiation of cells in higher plants and animals.
Whereas mitosis produces nearly identical progeny cells, meiosis is guaranteed to yield progeny that are different from the parent cell and usually different from each other. While mitosis is the mode of asexual division in both haploid and diploid cells, meiosis occurs only in diploids and yields haploid progeny. Both processes begin with a round of chromosome replication, in which the DNA is replicated and distributed between two replicate copies of each of the chromosomes. Since these replicate copies do not separate from each other immediately, it is useful to give them a name. They are called chromatids, or simply strands. They are separate structures, but remain joined together at one short region, called the centromere. (Note that each chromatid -- or strand -- consists of a DNA double helix, which, in turn, consists of two polynucleotide strands. In other words, what we call a strand at the cytological level consists of two strands at the molecular level.) So, before replication, in the single-strand stage, each chromosome consists of a single chromatid. After replication, but before the chromatids disjoin (separate at the centromere), they are at the double-strand (two chromatid) stage. Both mitosis and meiosis are extremely complicated in their detailed events, but simple in terms of the information flow -- the way various genes are transmitted from parent to progeny. We will concentrate on this informational level of the process.
Figure 4: Mitosis Without Crossing Over
Information Flow at Mitosis:
The usual mitotic division guarantees that each progeny cell receives a set of chromosomes -- genome -- that is identical to the parent cell (not counting occasional mutations, which normally occur at the time of DNA replication). Figure 4 shows how a diploid heterozygous for ADE2/ade2 produces two identical heterozygous progeny. At the onset of cell division each cell becomes polarized by a structure called the spindle. The chromosomes, at the two-strand stage, congregate at the center of this structure, separate into two independent single-stranded chromatids, and move along the fibers of the spindle to its poles, becoming organized within the cell so that when it divides the set of chromosomes at one pole ends up in one cell and those at the other pole end up in the other cell. This separation of the chromatids, called disjunction, occurs independently for each chromosome, so that each cell receives one copy of each of the chromosomes of its parent. If the parent is haploid then each progeny cell is haploid; if the parent is diploid, the progeny are diploid.
Information Flow at Meiosis:
Meiotic division (Figure 5) is a little more complicated. Since the net effect is to go from a diploid state to a haploid state, there are two divisions. In the first meiotic division, the centromeres do not disjoin. Instead, they pair: the two copies of each chromosome in the diploid nucleus come together, side-by-side, with their centromeres together.
Figure 5: Meiosis Without Crossing Over
The two copies are not necessarily identical, because they may be heterozygous at any number of genes; instead, they are said to be homologous and are sometimes referred to as homologs. Since each of the paired homologous chromosomes has two-strands, we call this the four-strand stage. The paired chromosomes separate and move along the spindle with each double-stranded member going to an opposite pole. This step may be described as centromere separation. The second meiotic division is analogous to mitosis in a haploid cell: the chromosomes, still in the two-strand stage, become organized at the center of a new spindle, and then the centromeres disjoin and the chromatids migrate to the spindle poles, yielding a total of four haploid sets of chromosomes at the single-strand stage. These become incorporated into the four nuclei of the meiotic products, in the case of yeast, the ascospores. Figure 6 shows how a diploid heterozygous for ADE2/ade2 (represented as +/-) produces four haploid spores, two ADE2 (+), and two ade2 (-).
Obviously, if the homologous chromosomes are not identical -- that is, some of the genes are heterozygous -- then the four product genomes will not be identical. Since the segregation of the chromatids at the first meiotic division is independent, from one chromosome to another, the segregation of genes on different chromosomes will be independent. Indeed, this is the pattern of assortment Mendel made famous. For each heterozygous pair of alleles, there will be two products -- ascospores -- of one type, and two of the other. If there are two heterozygous pairs on different chromosomes, they will segregate independently, so that all possible combinations will be represented among a sample of the progeny spores. The combinations that are the same as the parent-cell combinations are termed parental types, while those that are not like the parent-cell combinations are termed non-parental or recombinant, which is why the process is called recombination.
But chromosome segregation is not the only way that recombination can occur. According to this model, all the genes on the same chromosome would segregate together, and would never recombine, but we do not observe this. In fact, nearly all genes segregate from each other to some extent, although not always at the frequency predicted by random assortment. Recombination of genes on the same chromosome occurs when the homologous chromosomes exchange portions of their chromatids while paired at the four-strand stage in preparation for the first meiotic division (Figure 6). These exchanges -- crossovers -- occur at random along the chromatids, so the chance of two heterozygous genes on the same chromosome recombining depends on how far apart they are. The closer together they are, the less often they will recombine. The tendency of alleles of genes on the same chromosome to segregate together more often than expected by chance alone, is called linkage. The less frequently they recombine, the more tightly they are linked.
In yeast, crossing over occurs quite frequently, so even genes on the same chromosome appear to segregate independently unless they are very close together. The three specific genes that we are using in these experiments, mating-type and the two red adenine genes, are on three different chromosomes, so we will not observe any cases of linkage.
Figure 7: Meiosis With Crossing Over
We usually think of mitosis as a process in which there is no segregation of heterozygous characteristics. Each mitotic product is considered identical to the parent. We can easily demonstrate that mitotic segregation, or recombination, does occur at a low frequency in yeast (Figure 7) (Mortimer & Hawthorne 1969). It also occurs in the somatic cells of higher organisms, but is more difficult to demonstrate. Mitotic segregation, as we shall demonstrate, makes genes that are heterozygous become homozygous, resulting in the expression of recessive alleles that otherwise would not be expressed. Since most deleterious mutations are recessive, and are carried unexpressed in the heterozygous condition, this has important consequences. In higher organisms, including people, it may be a mechanism for development of some diseases, including cancer. In experimental organisms, especially in yeast, it provides an alternative to meiotic segregation for studying the genotype of diploid strains.
Figure 8: Mitosis With Crossing Over
We can understand mitotic recombination if we assume that at mitosis the homologous chromosomes actually do pair, as they do in meiosis. Since we cannot observe this under the microscope, it probably only happens occasionally. If we then assume that crossing over occurs between the centromere and the position of a pair of heterozygous alleles at this four-strand stage, then, when the chromatids disjoin, segregation could indeed occur. In fact, if the disjoined chromatids are randomly distributed to the poles, then in half of the divisions in which such a crossover occurs, the heterozygous alleles would become homozygous in the progeny cells, resulting in two clones of homozygous cells growing side-by-side. When it occurs in a developing colony, the colony becomes sectored. Since cells that are heterozygous for one of the red adenine mutations (such as ADE1/ade1 or ADE2/ade2) are white and those that are homozygous for one of the recessive alleles (ade1/ade1 or ade2/ade2) are red, mitotic segregation results in the appearance of red sectors in the normally white colonies (Figure 8).
Figure 9: Formation of Sectored Colonies
Homozygous ade2/ade2 clone produces red sector in otherwise cream-colored colony. Segregation at first division produces half-red and half-cream-colored colony (right), while segregation at later division produces smaller red sector (left).
Mitotic recombination can be induced by radiation, such as ultraviolet or x-rays. If we irradiate single cells that are heterozygous for ADE2/ade2 with a dose that kills about half of them, then several percent of the survivors will form colonies with red sectors. When radiation induces this effect, the segregation occurs early in the development of the colony, so the sectors are large, sometimes half of the colony. If the segregation results from a reciprocal crossover, then the cream-colored half of the colony should be homozygous for the dominant ADE2 allele. However, the cream-colored sector often turns out to still be heterozygous ADE2/ade2. This is easy to detect, because if it is heterozygous it will segregate red sectors, but if homozygous, it won't. This type of sectored colonies appears to arise from some sort of nonreciprocal event. We frequently observe nonreciprocal segregation or recombination events in yeast, and other fungi. They usually occur less frequently than the reciprocal events, but not always. We now consider them a distinct type of genetic event called gene conversion, and this process has gained more general interest recently as an attractive model for explaining the events occuring at the genetic sequences associated with the production of antibodies in higher animals (Fogel, Mortimer & Lusnak 1981).
Return to contents
Last updated Friday August 19 2005