CHAPTER 1

FUNDAMENTAL PROPERTIES OF GENES

Species share many traits in common from generation to generation. The bluebird nestlings in the box in my yard will look much like their parents when they are full-grown. The tomato plants that we set out will produce fruits that look, and hopefully taste, like those of their parents. Observable features of organisms, like color, size, and shape, comprise their phenotype. Adult male bluebirds share the phenotype of blue wings and a red breast.

A phenotype can be determined by inherited factors, by the environment, and often by both. For example, you are similar to your parents in many aspects of your appearance, your intelligence, and your susceptibility to some diseases, but you are not identical to them in all aspects of these traits. These three traits are clearly the product of both inherited and environmental factors. Considering appearance, I have crooked lower teeth and thinning gray hair, just like my father, but unlike me, neither of my parents has a scar on their knee from a childhood cut. The hair phenotype is inherited, whereas scars are from environmental influences. Quantitative studies show that intellectual capacity is about equally influenced by genetic and environmental factors. Susceptibility to diabetes is partially inherited, but a viral infection may trigger the autoimmune response at its core.

The genetic determinants of the inherited component of a phenotype are called genes. The set of genes that make up an organism is its genotype. In practice, we will consider only a small subset of the genes in an organism, which comprise a partial genotype. Likewise, an organism’s phenotype is all the traits it possesses, but we will only consider partial phenotypes, such as the blue wings of a bluebird or the color of the eyes of a fly.

This chapter will explore some of the basic characteristics of genes, and the experimental evidence for them. Some of the major points include the following.

§ Genes are the units of heredity

§ They are arranged in a linear fashion along chromosomes.

§ Recombination can occur both between and within genes.

§ Mutations in different genes required for a phenotype will complement each other in a diploid. This is the basis for genetic dissection of a pathway.

§ A gene is composed of a series of mutable sites that are also sites for recombination (now recognized as nucleotides).

§ One gene encodes one polypeptide.

§ The gene and the polypeptide are colinear.

§ Single amino acids are specified by a set of three adjacent mutable sites; this set is called a codon.

In considering experimental evidence for these points, some general genetic techniques as well as genetic techniques for bacteria and phage will be discussed.

Genes are mutable

We know that genes are mutable because they appear in different forms, called alleles. An allele that encodes a normal, functional product (found in nature or a standard laboratory stock) is called the wild type allele. Other alleles are altered in a way such that the encoded product differs in function from the wild type. This type of allele is mutated or mutant (adjective). The alteration in the gene is a mutation, and an organism showing the altered phenotype is a mutant (noun). Many mutated alleles encode a product that is nonfunctional or less functional than is the wild type, or normal, product; it is easier to break something than to improve it. A loss-of-function allele usually shows a recessive phenotype, which means that when it is present in the same cell as an allele that produces a different phenotype, the phenotype of the other allele is obtained. If no functional product is made, this loss-of-function allele is a null mutation; this can result from no expression or expression of a completely nonfunctional product. Other loss-of-function mutants make less than the normal amount of product, these are called hypomorphs. Another class of mutated allele encodes a product that provides an altered or new function. These gain-of-function mutations usually show a dominant phenotype; e.g. when the gain-of-function allele is in the same cell as a loss-of-function allele, the phenotype of the gain-of-function allele is observed. Another class of gain-of-function mutants makes more than the normal amount of product; these are called hypermorphs.

Within a population, the number of alleles at a given locus can vary considerably. Mutant alleles that cause a loss or detrimental change in the function of a gene are selected against, and they are rare in a wild population. In the laboratory, one can utilize growth conditions that select for certain mutants or that maintain mutants, so mutant organisms that would be rare or non-existent in the wild are encountered quite frequently in the laboratory. In many cases, however, alternate forms of genes, i.e. different alleles, have no particular effect on gene function. These variants can be found quite frequently in a population. One common examples of such genetically determined, apparently neutral variation is the ability of some persons to "roll" their tongue. In general, these common alleles are roughly equivalent in function to the wild type allele. Thus they are not providing a strong selective advantage or disadvantage. All the common alleles can be considered the wild type allele. Variant alleles that occur in greater than 5% of population are called polymorphisms. The term variant includes all alternative forms of a gene, whether they have an effect on function or not. The term mutant allele sometimes implies an altered function for the gene.

As will become clearer when we study the fine structure of genes, it is possible to change the structure of the gene (the nucleotide sequence in DNA) without changing the structure of the encoded polypeptide (the amino acid sequence). These silent substitutions also generate different alleles, but they can only be detected by examining the structure of the gene; the phenotypes of alleles that differ by silent substitutions are usually identical.

Another possibility is that a mutant allele not only causes a loss of function of the encoded protein, but this altered protein interferes with the activity of other proteins. One way this can happen is by the polypeptide product forming a complex with other polypeptides (e.g. in a heteromultimeric enzyme complex). Sometimes the mutant polypeptide will prevent formation of an active complex with the partner, even in the presence of wild-type polypeptide, thereby leading to a dominant negative phenotype. These are of considerable utility now in designing mutant genes and proteins to try to disrupt some cellular function. They are most commonly made in a vector that will drive a high level of expression of the mutant gene, and usually over-expression is needed to generate the dominant negative phenotype.

Genes are the units of heredity: Mendel's Laws

First Law: Alleles segregate equally

The original experiments by Gregor Mendel involved phenotypic traits (physical, observable characteristics) controlled by single genes. The first one we'll consider is seed color, which can be yellow or green. The dominant allele, denoted Y, generates yellow peas in either the homozygous (YY) or heterozygous (Yy) state, whereas the recessive allele, denoted y, generates green peas only in the homozygous state (yy). (In plants and flies, the dominant allele is denoted by a capitalized abbreviation and the recessive allele is denoted by a lower case abbreviation.) In a cross between two parents, one homozygous for the dominant allele (YY) and the other homozygous for the recessive allele (yy), Mendel showed that the F1 progeny were all yellow, i.e. they had had the same phenotype as the parent with the dominant allele. The recessive allele was not contributing to the phenotype.

Had it been lost during the cross? No, when the F1 is crossed with itself, both parental phenotypes were seen in the F2 progeny. The effect of the recessive allele reappeared in the second cross, showing that it was still present in the F1 hybrids, but was having no effect. In the F2 progeny, the dominant phenotype (yellow) was observed in 75% of the progeny and the recessive (green) appeared in only 25% of the progeny.

Note that discrete phenotypes were obtained (yellow or green), not a continuum of phenotypes. The genes are behaving as units, not as some continuous function.

The results can be explained by hypothesizing that each parent has two copies of the gene (i.e., two alleles) that segregate equally, one per gamete. Since they are homozygous, each parent can form only type of gamete (Y or y, respectively). When the gametes join in the zygotes of the F1 generation, each individual receives one dominant allele and one recessive allele (Yy), and thus all of the F1 generation shows the dominant phenotype (e.g. yellow peas). This is the uniform phenotype observed for the F1 generation.

The two alleles did not alter one another when present together in the F1 generation, because when F1 is crossed with F1, the two parental phenotypes are obtained in the F2 generation.

The ratio of 3:1 dominant: recessive observed in the F2 is expected for the equal segregation of the alleles from the F1 (Y and y) and their random rejoining in the zygotes of the F2, producing the genotypes 1 YY, 2 Yy, and 1 yy. Again the genes are behaving as discrete units. These precise mathematical ratios (3:1 for phenotypes in this cross, or 1:2:1 for the genotype) provide the evidence that genes, units of heredity, are determining the phenotypes observed.

Figure 1.1. Mendel’s First Law: Equal segregation of alleles.

Not all loci show the property of complete dominance illustrated by the Y locus in peas. Sometimes partial dominance is observed, in which an intermediate phenotype seen in a heterozygote. An example is the pink color of snapdragons obtained when white and red are crossed. However, the parental phenotypes reappear in the F2 generation, showing that the alleles were not altered in the heterozygote. In this case, gene dosage is important in determining the phenotype; two wild-type alleles produce a red flower, but only one wild-type allele produces a pink flower. Sometimes co‑dominance is observed, in which both alleles contribute equally to the phenotype. An example is the ABO blood group locus. Heterozygotes have both the A and B form of the glycoprotein that is encoded by the different alleles of the gene.

Second Law: Different genes assort independently

Now that we have some understanding of the behavior of the different alleles of a single gene, let's consider how two different genes behave during a cross. Do they tend to stay together, or do they assort independently?

Mendel examined two different traits, seed color (as described in the previous section) and seed shape. Two alleles at the locus controlling seed shape were studied, the dominant round (R) and recessive wrinkled (r) alleles. Mendel crossed one parent that was homozygous for the dominant alleles of these two different genes (round yellow RRYY) with another parent that was homozygous for the recessive alleles of those two genes (wrinkled green rryy) (see Fig. 1.2).

Re-stating the basic question, do the alleles at each locus always stay together (i.e. round with yellow, wrinkled with green) or do they appear in new combinations in the progeny? As expected from the 1st law, the F1 generation shows a uniform round yellow phenotype, since one dominant and one recessive allele was inherited from the parents. When the F2 progeny are obtained by crossing the F1 generation, the parental phenotypes reappear (as expected from the first law), but two nonparental phenotypes also appear that differ from the parents: wrinkled yellow and round green!

The results can be explained by the alleles of each different gene assorting into gametes independently. For example, in the gametes from the F1 generation, R can assort with Y or y, and r can assort with Y or y, so that four types of gametes form: RY, Ry, rY, and ry. These can rejoin randomly with other gametes from the F1 generation, producing the results in the grid shown in Fig. 1.2. The alternative, that R always assorted with Y, etc. was not observed.

Again, the genes are behaving as units, and the gene for one trait (e.g. color) does not affect a gene for another trait (e.g. shape). Further breeding shows that many nonparental genotypes are present, some of which give a parental phenotype (e.g. RrYy).

These results are obtained for genes that are not linked on chromosomes. Linkage can lead to deviations from these expected ratios in a mating, and this can be used to map the locations of genes on chromosomes, as discussed in the next section.

Figure 1.2. Mendel’s Second Law: Independent assortment of different genes.

Genes are on chromosomes

In 1902, Sutton and Boveri independently realized that the behavior of genes in Mendelian crosses mimics the movement of chromosomes during meiosis and fertilization. They surmised that the two alleles of each gene correlated with the homologous pair of chromosomes. The equal segregation of alleles could be explained by the separation of homologous chromosomes at anaphase I of meiosis. As diagrammed in Fig. 1.3, the chromosome with the R allele would go to a different cell than its homolog with the r allele at the end of meiosis I, and likewise for the Y and y alleles. The rejoining of alleles corresponded to the joining of chromosomes, one from each parent, at fertilization. The independent assortment of different genes mimics the independent separation of homologs of different chromosomes in meiosis. For instance, the paternal copy of chromosome 1 may assort with the maternal copy of chromosome 21 in formation of a gamete. Figure 1.3 shows the dark blue chromosome with the R allele assorting with the light red chromosome with the y allele, but it is equally likely that it will assort with the dark red chromosome with the Y allele. As shown in Fig. 1.4, the completion of meiosis results in 4 germ cells for each cell that entered meiosis. All the combinations of alleles of different genes diagrammed in Fig. 1.2 can be formed in this process.

This correlation of the behavior of alleles in matings and the movement of chromosomes during meiosis and fertilization produced the chromosomal theory of inheritance. One could think of the alleles discerned in genetic crosses as being located at the same locus on the different homologs of a chromosome.

Legend for Figure 1.3. [NEXT PAGE] Movement of chromosomes during meiosis I, the first divisional process of meiosis. The chromosomes are drawn starting after the synthesis of a copy of each homologous chromosome, so there are two copies of each homolog of a chromosome pair. The two DNA duplexes for each homolog are joined at a single centromere. Meiosis is the process of segregating these four copies of each chromosome (4 alleles for each gene) into four germ cells with one copy of each chromosome. In this diagram, two different chromosome pairs are displayed with each homolog colored a different shade (dark or light red for the shorter chromosome, dark or light blue for the longer chromosome). Each line is a duplex DNA molecule. The R locus is on the longer blue chromosome, with distinctive alleles for each homolog, and the Y locus is on the shorter red chromosome, again with distinctive alleles for each homolog. Meiosis begins with the leptotene, when the chromosomes become visible as long filaments. The two homologous chromosomes undergo synapsis during zygotene, in which they align along their lengths. The chromosomes become shorter and thicker during pachytene, and crossovers between chromatids of the two different homologs form. The chromosomes start to pull apart in diplotene, at which point the crossovers in chiasmata are visible. The chromosomes shorten further during diakinesis. During metaphase, the chromosomes align along the equatorial plane of the cell, i.e. the plane in which cell division will occur. The nuclear membrane is disassembled at this point. The members of a homologous pair move to opposite poles of the cell during anaphase. This is the cytological even that accounts for the equal segregation of alleles. Note that the centromeres do not separate during anaphase I, and the two sister chromatids stay together. The crossovers are also resolved at this stage. In some organisms, the nuclear membrane reforms during a telophase of meiosis I, followed by cell division and an interphase I.

Figure 1.3. Movement of chromosomes during meiosis I.

Figure 1.4. Movement of chromosomes during meiosis II, the second divisional process of meiosis. The chromosomes, each with two sister chromatids linked at the centromere, contract and become visible during prophase II. The nuclear membrane disassembles and chromosomes align along the equatorial plane during metaphase II. The centromeres divide and the chromosomes separate during anaphase II. The nuclear membrane reforms during telophase II, and after cell division, a tetrad with one of each chromosome is produced. If the dark blue chromosome had assorted with the dark red chromosome during anaphase I, the resulting spores would be R Y and r y.

Linked genes lie along chromosomes in a linear array

The proponents of the chromosome theory of heredity realized that that the number of genes would probably greatly exceed the number of chromosomes. However, many early genetic studies showed independent assortment between genes with no evidence of linkage. This led to a proposal that a chromosome broke down during meiosis into smaller parts consisting only of individual genes, but such disassembly of chromosomes during meiosis was never observed. Evidence for linkage did eventually come from a demonstration of the absence of independent assortment between different genes. In complementary work, McClintock and Creighton demonstrated an association between different genes and a particular chromosome in 1931.

The behavior of two genes carried on the same chromosome may deviate from the predictions of Mendel's 2nd law. The proportion of parental genotypes in the F2 may be greater than expected because of a reduction in nonparental genotypes. This propensity of some characters to remain associated instead of assorting independently is called linkage. When deduced from studies of a population, it is called linkage disequilibrium.

Fig. 1.5. illustrates a cross that shows linkage.

(1) An F1 heterozygote (AaBb) is made by crossing a homozygous dominant parent (AABB) with a homozygous recessive parent (aabb). A backcross is then made between the F1 heterozygote (AaBb) and a recessive homozygote (aabb), so that the alleles of the recessive parent make no contribution to the phenotype of the progeny. (This is a fairly common cross in genetics, since the genotype of an individual can be ascertained by crossing with such an individual, homozygous recessive at both loci.)

(2) As shown in part A of Fig. 1.5, if there is no linkage, one expects 50% parental phenotypes (from genotypes AaBb and aabb) and 50% nonparental phenotypes (from genotypes Aabb and aaBb). This fits with the expectations of Mendel's law of independent assortment of different genes for this backcross. (Sometimes the nonparental phenotypes are called "recombinant" but that confuses this reassortment with events that involve crossovers in the DNA.)

(3) If the two genes are linked and there is no recombination between them, then all progeny will have a parental phenotype. In particular, if genes A and B are linked, then the backcross AB/ab x ab/ab yields AB/ab progeny 50% of the time and ab/ab progeny 50% of the time, in the absence of recombination. [In this notation, the alleles to the left of the slash (/) are linked on one chromosome and the alleles to the right of the slash are linked on the homologous chromosome.] Thus only the parental phenotypes are found in the progeny of this cross (i.e. the progeny will show either the dominant characters at each locus or the recessive characters at each locus). Another way of looking at this is that, in the absence of recombination between the homologous chromosomes, all the progeny of this cross will be one of the first two types shown in panel B of Fig. 1.5.

Note that the dominant alleles can be in the opposite phase, with the dominant A allele linked to the recessive b allele. For instance, the F1 heterozygote could be formed by a cross between the parents Ab/Ab and aB/aB to generate Ab/aB. In this case, the backcross Ab/aB x ab/ab will still generate only progeny with parental phenotypes but a new, nonparental genotype (i.e. Ab/ab and aB/ab; these look like the parents Ab/Ab and aB/aB). The phase with both dominant alleles on the same chromosome is called the "coupling conformation”, whereas the opposite phase is called the "repulsing conformation."

Figure 1.5. Linkage and recombination between genes on the same chromosome.

(4) But in most cases, recombination can occur between linked genes. In part B of Fig. 1.5, there is an increase in parental types (from the 50% expected for unlinked genes to the observed 70%) and a decrease in nonparental types (30%), showing that allele A tends to stay with allele B, in contrast to the prediction of the 2nd law. Thus these genes are not assorting independently, and one concludes there is linkage between genes A and B.

The frequency of parental types is not as high as expected for linkage without recombination (which would have been 100%, as discussed above). Indeed, the nonparental types in this experiment result from a physical crossover (breaking and rejoining) between the two homologous chromosomes during meiosis in the AB/ab parent. This is a recombination event in the DNA.

(5) We conclude that genes A and B are linked, and have a recombination frequency of 30%.

map distance = x 100

1 map unit = 1 centiMorgan = 1% recombination

1 centiMorgan = 1 cM = about 1 Mb for human chromosomes

Question 1.1.

In their genetic studies of the fruitfly Drosophila melanogaster, Thomas Hunt Morgan and his co-workers found many examples of genes that associated together in groups. One example is the gene for purple eye color (the mutant allele is abbreviated pr) that is recessive to the allele for normal red eyes (pr+) and the gene for vestigial, or shortened, wings (the mutant allele is abbreviated vg) that is recessive to the normal allele for long wings (vg+). When a homozygous purple vestigial fly is crossed to a homozygous red-eyed long-winged fly, the heterozygous F1 generation shows a normal phenotype. When male heterozygotes are backcrossed to females that are homozygous purple vestigial (i.e. homozygous recessive at both loci), only two phenotypes appear in the progeny: the homozygous recessive purple vestigial flies and the normal flies.

a) What are the predictions of the backcross if the two genes are not linked?

b) What do the results of the backcross tell you?

c) If the heterozygotes F1 in the backcross are female, then purple long-winged and red-eyed vestigial flies appear in the progeny. The combined frequency of these recombinant types is 15.2 %. What does this tell you about the arrangement of the genes?

Question 1.5 provides some practice in calculating recombination frequencies.

Individual map distances are (roughly) additive.

A‑‑10‑‑B‑5‑C

‑‑‑-‑‑‑-‑15‑‑‑‑

The recombination distances are not strictly additive if multiple crossovers can occur (see questions 1.6 and 1.7.)

Recombination between linked genes occurs by the process of crossing over between chromosomes, at chiasma during meiosis. The mechanism of recombination is considered in Chapter 8.

Genetic dissection by complementation

Genes are the hereditary units that when altered change a phenotype; genes are classically defined by their effects on phenotype. But in many cases more than one gene affects a phenotype. Metabolic pathways, such as synthesis of DNA, repair of DNA, synthesis of leucine, or breakdown of starch occur in multiple steps catalyzed by enzymes. Each subunit of each enzyme is encoded in a gene, and all those genes are needed for the efficient running of the pathway. Multiple genes also determine complex traits, such as susceptibility to substance abuse, diabetes, and other diseases, and probably less pressing concerns, such as retaining a healthy head of hair after you are 40.

Many pathways have been elucidated by finding many mutants that are defective in that process, hopefully enough to sample every gene in the organism (saturation mutagenesis), and grouping them according to the gene that is mutated. All the mutations in the same gene fall into the same complementation group. Two mutants complement each other if they restore the normal phenotype when together in a diploid. This occurs when the mutants have mutations in different genes. If one is examining mutants with a similar phenotype (e.g. inability to grow on leucine or inability to make DNA), then tests of all pairwise combinations of the mutants will place them into complementation group, which complement between groups but not within groups. The complementation groups then define the genes in the process under study. This is a powerful method of genetic dissection of a pathway. We will encounter it over and over in this textbook. In this section, we will look at complementation in detail, and contrast it with recombination.

Complementation

Dominance observed in heterozygotes reflects the ability of wild-type alleles to complement loss-of-function alleles. You know that a dominant allele will determine the phenotype of a heterozygote composed of a dominant and a recessive allele. Often, recessive alleles are loss-of-function mutations, whereas the dominant allele is the wild type, encoding a functional enzyme. Using the example that led to Mendel's First Law, a cross between YY (yellow) peas and yy (green) peas yielded yellow peas in the F1 heterozygote (Yy). In this case the chromosome carrying the Y allele encodes the enzymatic function missing in the product of the recessive y allele, and the pathway for pigment biosynthesis continues on to make a yellow product. Thus you could say that the dominant Y allele complements the recessive y allele - it provides the missing function.

We can continue the analogy to the classic cross for Mendel's Second Law. Let's look at the same genes, but a different arrangement of alleles. Consider a cross between round green (RRyy) and wrinkled yellow (rrYY) peas; in this case each parent is providing a dominant allele of one gene and a recessive allele of the other. The F1 heterozygote is round yellow (RrYy), i.e., the phenotypes of the dominant alleles are seen. But you could also describe this situation as the chromosomes from rrYY peas complementing the deficiency in the RRyy chromosomes, and vice versa. In particular, the Y allele from the rrYY parent provides the function missing in the y allele from the RRyy parent, and the R allele from the RRyy parent provides the function missing in the r allele from the rrYY parent. If the phenotype you are looking for is a round yellow pea, you could conclude that mutants in the R-gene complement mutants in the Y-gene. Since in a heterozygote, the functional allele will provide the activity missing in the mutant allele (if the mutation is a loss-of-function), one could say that dominant alleles complement recessive alleles. Thus dominant alleles determine the phenotype in a heterozygote with both dominant and recessive alleles.

A general definition of complementation is the ability of two mutants in combination to restore a normal phenotype. Consider two genes, A with wild-type allele A1 and loss-of-function allele A2, and B with wild-type allele B1 and loss-of-function allele B2. A cross between two mutant organisms, one homozygous for mutations in A and the other homozygous for mutations in B, produces wild-type progeny:

A2A2 B1B1 ´ A1A1 B2B2 parents

¯

A2A1 B1B2 F1 progeny

Note that one wild type allele is present for each locus, A1 for gene A and B1 for gene B. Thus the F1 progeny, what was missing in each mutant parent is restored in the heterozygous progeny. We say that the two mutants complement each other.

Complementation distinguishes between mutations in the same gene or in different genes

The ability of complementation analysis to determine whether mutations are in the same or different genes is the basis for genetic dissection. In this process, one finds the genes whose products are required in a pathway. In the examples from peas used above, the metabolic pathway to yellow pigments is distinctly different from the pathway to round peas, which is the starch biosynthesis pathway. Complementation analysis is useful in dissecting the steps in a pathway, starting with many mutants that generate the same phenotype. This is a more conventional example of complementation.

Many fungi can propagate as haploids but can also mate to form diploids prior to sporulation. Thus one can screen for mutants in haploids and obtain recessive mutants, and then test their behavior in combination with other mutants in the diploid state. Let's say that a haploid strain of a fungus was mutagenized and screened for arginine auxotrophs, i.e. mutants that require arginine to grow. Six of the mutants were mated to form all the possible diploid combinations, and tested for the ability of the diploids to grow in the absence of arginine (prototrophy). The results are tabulated below, with a + designating growth in the absence of arginine, and a - designating no growth.

Table 1.1. Growth of the diploids in the absence of arginine

Mutant number

Mutant number	1	2	3	4	5	6
1	-	+	+	-	+	+
2		-	-	+	+	+
3			-	+	+	+
4				-	+	+
5					-	+
6						-

As you would expect, when mutant 1 is mated with itself, the resulting diploid is still an auxotroph; this is the same as being homozygous for the defective allele of a gene. But when mutant 1 is mated with mutant 2 (so their chromosomes are combined), the resulting diploid has prototrophy restored, i.e. it can make its own arginine. This is true for all the progeny. We conclude that mutant 1 will complement mutant 2. If we say that mutant 1 has a mutation in gene 1 of the pathway for arginine biosynthesis, and mutant 2 has a mutation in gene 2 of this pathway, then the diagram in Fig. 1.6 describes the situation in the haploids and the diploid. (Note that if the organism has more than one chromosome, then genes 1 and 2 need not be on the same chromosome.) Since the enzymes encoded by genes 1 and 2 are needed for arginine biosynthesis, neither mutant in the haploid state can make arginine. But when these chromosomes are combined in the diploid state, the chromosome from mutant 1 will provide a normal product of gene 2, and the chromosome from mutant 2 will provide a normal product of gene 1. Since each provides what is missing in the other, they complement. Just like Jack Spratt and his wife. Mutant 1 will also complement mutant 3, and one concludes that these strains are carrying mutations in different genes required for arginine biosynthesis.

Figure 1.6. Complementation between two haploid mutants when combined in a diploid.

In contrast, the diploid resulting from mating mutant 1 with mutant 4 is still an auxotroph; it will not grow in the absence of arginine. Assuming that both these mutants are recessive (i.e. contain loss-of-function alleles), then we conclude that the mutations are in the same gene (gene 1 in the above diagram). We place these mutants in the same complementation group. Likewise, mutants 2 and 3 fail to complement, and they are in the same complementation group. Thus mutant 2 and mutant 3 are carrying different mutant alleles of the same gene (gene 2).

Mutant 5 will complement all the other mutants, so it is in a different gene, and the same is true for mutant 6. Thus this mutation and complementation analysis shows that this fungus has at least 4 genes involved in arginine biosynthesis: gene 1 (defined by mutants alleles in strains 1 and 4), gene 2 (defined by mutants alleles in strains 2 and 3), and two other genes, one mutated in strain 5 and the other mutated in strain 6.

Genetic dissection by complementation is very powerful. An investigator can start with a large number of mutants, all of which have the same phenotype, and then group them into sets of mutant alleles of different genes. Groups of mutations that do not complement each other constitute a complementation group, which is equivalent to a gene. Each mutation in a given complementation group is a mutant allele of the gene. The product of each gene, whether a polypeptide or RNA, is needed for the cellular function that, when altered, generates the phenotype that was the basis for the initial screen. The number of different complementation groups, or genes, gives an approximation of the number of polypeptides or RNA molecules utilized in generating the cellular function.

Question 1.2.

Consider the following complementation analysis. Five mutations in a biosynthetic pathway (producing auxotrophs in a haploid state) were placed pairwise in a cell in trans (diploid analysis). The diploid cells were then assayed for reconstitution of the biosynthetic pathway; complementing mutations were able to grow in the absence of the end product of the pathway (i.e. they now had a functional biosynthetic pathway). A + indicates a complementing pair of mutations; a - means that the two mutations did not complement.

Mutation number

                                                              &n