5

CHAPTER

Gene Interaction

The colors of peppers are determined

by the interaction of several genes. An
allele Y promotes the early elimination of
chlorophyll (a green pigment), whereas y
does not. Allele R determines red and r
CHAPTER OUTLINE AND LEARNING OBJECTIVES determines yellow carotenoid pigments.
Alleles c1 and c2 of two different
5.1 INTERACTIONS BETWEEN THE ALLELES OF A SINGLE GENE: genes down-regulate the amounts of
carotenoids, causing the lighter shades.
VARIATIONS ON DOMINANCE Orange is down-regulated red. Brown
LO 5.1 Distinguish between the various types of dominance, based on the phenotypes is green plus red. Pale yellow is down-
of heterozygotes. regulated yellow. [Anthony Griffiths.]
LO 5.2 Recognize phenotypic ratios diagnostic of the presence of a lethal allele.
LO 5.3 Give some possible reasons why incomplete penetrance and variable expressivity
may occur in a population of individuals with identical genotypes at a locus under
study.

5.2 INTERACTION OF GENES IN PATHWAYS

LO 5.4 Describe the hypotheses proposed to explain various types of gene interaction at
the molecular level.

5.3 INFERRING GENE INTERACTIONS

LO 5.5 Determine whether two mutations are in the same gene or in different genes,
using progeny ratios or using complementation tests.
LO 5.6 Infer how two genes may be interacting, based on modified Mendelian ratios.
LO 5.7 For known cases of gene interaction, predict progeny ratios in crosses.

153

07_GriffitITGA12e_11478_Ch05_153_192.indd 153 03/10/19 12:01 PM

 Reflecting the fact that the thousands of genes in the genome must clearly
interact at the cellular level, many genes are observed to interact at the phe-
CHAPTER OBJECTIVE notypic level, resulting in modified inheritance ratios. Our broad objective in
this chapter is to catalog the inheritance patterns that reveal various types of
gene interaction.

T 5.1 INTERACTIONS BETWEEN THE

he thrust of our presentation in the book so far has
been to show how geneticists identify a gene that
affects some biological property of interest. We have
ALLELES OF A SINGLE GENE:
seen how the approaches of forward genetics can be used VARIATIONS ON DOMINANCE
to identify individual genes. The researcher begins with a
set of mutants and then crosses each mutant with the wild LO 5.1 Distinguish between the various types of dominance,
type to see if the mutant shows single-gene inheritance. The based on the phenotypes of heterozygotes.
cumulative data from such a research program would reveal
a set of genes that all have roles in the development of the LO 5.2 Recognize phenotypic ratios diagnostic of the
property under investigation. In some cases, the researcher presence of a lethal allele.
may be able to identify specific biochemical functions for
many of the genes by comparing gene sequences with those LO 5.3 Give some possible reasons why incomplete
of other organisms. The next step, which is a greater chal- penetrance and variable expressivity may occur in a
population of individuals with identical genotypes at
lenge, is to deduce how the genes in a set interact to influ- a locus under study.
ence phenotype.
How are the gene interactions underlying a prop- There are thousands of different ways to alter the sequence
erty deduced? One molecular approach is to analyze pro- of a gene, each producing a mutant allele, although only
tein interactions directly in vitro by using one protein as some of these mutant alleles will appear in a real popula-
“bait” and observing which other cellular proteins attach tion. The known mutant alleles of a gene and its wild-type
to it. Proteins that are found to bind to the bait are candi- allele are referred to as multiple alleles, or an allelic series.
dates for interaction in the living cell. Another molecular One of the tests routinely performed on a new mutant
approach is to analyze mRNA transcripts. The genes that allele is to see if it is dominant or recessive. Basic infor-
collaborate in some specific developmental process can be mation about dominance and recessiveness is useful in
defined by the set of RNA transcripts present when that working with the new mutation and can be a source of
process is going on, a type of analysis now carried out with insight into the way the gene functions, as we will see in
a technique called RNA-seq, as we will see in Chapter 14. the examples. Dominance is a manifestation of how the
Finally, gene interactions and their significance in shaping alleles of a single gene interact in a heterozygote. In any
phenotype can be deduced by genetic analysis, which is the experiment, the interacting pair of alleles may be wild-type
focus of this chapter. and mutant alleles (+/m), or two different mutant alleles
Gene interactions can be classified broadly into two (m1 /m2 ). Several types of dominance have been discovered,
categories. The first category consists of interactions each representing a different type of interaction between a
between alleles of a single gene (a single locus). These pair of alleles.
types of interactions can be thought of, broadly speaking,
as variations on dominance. In earlier chapters we dealt
with alleles displaying full dominance or full recessiveness, Complete dominance and recessiveness
but as we shall see in this chapter, there are other types The simplest type of dominance is full dominance, also
of dominance, each with its own underlying cell biology. called complete dominance, which we examined in Chapter 2.
Although this information does not address the range of The phenotype of a fully dominant allele will be displayed
genes affecting a function, a great deal can be learned of a when only one copy is present, such as in a heterozygote
gene’s role by considering allelic interactions. The second individual; in a heterozygote, the other allele whose phe-
category consists of interactions between two or more loci. notype is not displayed is the fully recessive allele. In full
These interactions reveal the number and types of genes dominance, the homozygous dominant cannot be distin-
in the overall program underlying a particular biological guished from the heterozygote; that is, at the phenotypic
function. level, genotype A /A cannot be distinguished from genotype

154

07_GriffitITGA12e_11478_Ch05_153_192.indd 154 03/10/19 12:01 PM

 5.1 Interactions Between the Alleles of a Single Gene: Variations on Dominance 155

Mutations of haplosufficient genes are recessive

Homozygous Heterozygote Homozygous
wild type recessive mutant
+/+ +/m m/m

Protein Functional Functional Nonfunctional

mRNA

Chromosome
+ + m
+ m m
Chromosome

mRNA

Protein

Functional Nonfunctional Nonfunctional

FIGURE 5-1 In the heterozygote, even though the mutated copy of the gene produces
nonfunctional protein, the wild-type copy generates enough functional protein to produce
the wild-type phenotype.

A /a . As we saw in Chapter 2, the alleles that result in PAH to break down the phenylalanine entering the body.
phenylketonuria (PKU) and many other single-gene human One “dose” of functional PAH protein, produced by one
diseases are fully recessive, whereas their wild-type alleles P allele, results in the wild-type phenotype. Thus, the PAH
are dominant. Other single-gene diseases such as pseudo- wild-type allele is said to be haplosufficient. Haplo means
achondroplasia result from alleles that are fully dominant, a haploid dose (one), and sufficient refers to the ability of
whereas, in those cases, the wild-type allele is recessive. that single dose to produce the wild-type phenotype. Hence,
How can these dominance relations be interpreted at the both P /P (two doses) and P /p (one dose) have enough PAH
cellular level? activity to result in the normal cellular chemistry. People
The disease PKU is a good general model for recessive with p /p have zero doses of PAH activity. Figure 5-1 illus-
mutations. Recall from Chapter 2 that PKU is caused by a trates this general notion.
defective allele of the gene encoding the enzyme phenylal- How can we explain fully dominant mutations? There
anine hydroxylase (PAH). In the absence of normal PAH, are several molecular mechanisms for dominance. A regu-
the phenylalanine entering the body in food is not broken larly encountered mechanism is that the wild-type allele of
down and hence accumulates. Under such conditions, phe- a gene is haploinsufficient. In haploinsufficiency, one wild-
nylalanine is converted into phenylpyruvic acid, which is type dose is not enough to achieve normal levels of func-
transported to the brain through the bloodstream and there tion. Assume that 16 units of a gene’s product are needed
impedes normal development, leading to intellectual dis- for normal chemistry and that each wild-type allele can
abilities. The reason that the defective allele is recessive is make 10 units. Two wild-type alleles will produce 20!units
that just one copy of the wild-type allele P produces enough of product, well over the minimum. But consider what

07_GriffitITGA12e_11478_Ch05_153_192.indd 155 03/10/19 12:01 PM

 156 C H A P T E R 5 ! ! Gene Interaction

Two models forModified

dominance of a mutation connective-tissue protein formed of three monomers inter-
Some F2 Ratios
twined (a trimer). In the mutant heterozygote, the abnormal
Model 1: Model 2: Phenotype protein wraps around one or two normal ones and distorts
Haploinsufficiency Dominant negative the trimer, leading to malfunction. In this way, the defective
collagen acts as a spoiler. The difference between haploin-
sufficiency and the action of a dominant negative as causes
+/+ of dominance is illustrated in Figure 5-2.

KEY CONCEPT For most genes, a single wild-type copy is

2 “doses” of product Dimer Wild type
adequate for full expression (such genes are haplosufficient),
and their null mutations are fully recessive. Harmful mutations
of haploinsufficient genes are often dominant. Mutations in
M/M Mutant genes that encode units in homo- or heterodimers can behave
as dominant negatives, acting through “spoiler” proteins.
0 “doses”
Incomplete dominance
Snapdragons (Antirrhinums) are one of the favorite plant
+/M Mutant
species for genetic analysis. When a pure-breeding wild-
type snapdragon line having red petals is crossed with a
1 “dose” (inadequate) pure line having white petals, the F1 has pink petals. If an F2
is produced by selfing the F1, the result is
FIGURE 5-2 A mutation may be dominant because (left) a single
1 of the plants have red petals
wild-type gene does not produce enough protein product for proper 4
function or (right) the mutant allele acts as a dominant negative that
1 of the plants have pink petals
produces a “spoiler” protein product. 2
1 of the plants have white petals
4
happens if one of the alleles is a null mutation, which pro-
duces a nonfunctional protein (or no protein at all). A!null Figure 5-3 shows these phenotypes. From this 1: 2:1
mutation in combination with a single wild-type allele ratio in the F2, we can deduce that the inheritance pattern is
would produce 10 + 0 = 10 units, well below the minimum. based on two alleles of a single gene. However, the hetero-
Hence, the heterozygote (wild type/null) is mutant, and the zygotes (the F1 and half the F2 ) are intermediate in pheno-
mutant allele is, by definition, dominant. In mice, the gene type. By inventing allele symbols, we can list the genotypes
Tbx1 is haploinsufficient. This gene encodes a transcription- in this experiment as c+ /c+ (red), c /c (white), and c+ /c (pink).
regulating protein (a transcription factor) that acts on genes The occurrence of the intermediate phenotype suggests an
responsible for the development of the pharynx. A!knock- incomplete dominance, the term used to describe the gen-
out of one wild-type allele results in an inadequate con- eral case in which the phenotype of a heterozygote is inter-
centration of the regulatory protein, which results in mediate between those of the two homozygotes, on some
defects in the development of the pharyngeal arteries. The quantitative scale of measurement.
same haploinsufficiency is thought to be responsible for How do we explain incomplete dominance at the
DiGeorge syndrome in humans, a condition with cardio- molecular level? In incomplete dominance, each wild-type
vascular and craniofacial abnormalities. allele generally produces a set dose of its protein product.
Another important type of dominant mutation is called a The number of doses of a wild-type allele determines the
dominant negative. Polypeptides with this type of mutation concentration of a chemical made by the protein, such as
act as “spoilers” or “rogues.” In some cases, the gene product pigment. In the four-o’clock plant, two doses (c+ /c+ ) pro-
is a unit of a homodimeric protein, a protein composed of duce the most copies of transcript, thus producing the
two units of the same type. In the heterozygote (+/M) , the greatest amount of protein and, hence, the greatest amount
mutant polypeptide binds to the wild-type polypeptide and of pigment, enough to make the flower petals red. One dose
acts as a spoiler by distorting it or otherwise interfering with (c+ /c) produces less pigment, and so the petals are pink.
its function. The same type of spoiling can also hinder the A!zero dose (c /c) produces no pigment.
functioning of a heterodimer composed of polypeptides from
different genes. In other cases, the gene product is a mono- Codominance
mer, and, in these situations, the mutant protein binds the Another variation on the theme of dominance is codom-
substrate, and it acts as a spoiler by hindering the ability of inance, the expression of both alleles of a heterozygote.
the wild-type protein to bind to the substrate. A clear example is seen in the human ABO blood groups,
An example of a mutation that can act as a dominant where there is codominance of antigen alleles. The ABO
negative is found in the gene for collagen protein. Some blood groups are determined by three alleles of one gene.
mutations in this gene give rise to the human phenotype These three alleles interact in several ways to produce the four
osteogenesis imperfecta (brittle-bone disease). Collagen is a blood types of the ABO system. The three most important

07_GriffitITGA12e_11478_Ch05_153_192.indd 156 03/10/19 12:01 PM

 5.1 Interactions Between the Alleles of a Single Gene: Variations on Dominance 157

Incomplete dominance Sickled and normal red blood cells

FIGURE 5-3 In snapdragons, a heterozygote is pink (right),

intermediate between the two homozygotes red (middle) and white
(left). The pink heterozygote demonstrates incomplete dominance.
[John Kaprielian/Science Source.]

alleles are i, I A , and I B , but a person can have only two of the
three alleles or two copies of one of them. The pairwise com- FIGURE 5-4 The sickle-shaped cell is caused by a single mutation in
the gene for hemoglobin. [Eye of Science/Science Source.]
binations result in six different genotypes: the three homozy-
gotes and three different types of heterozygotes, as follows: Figure 5-4 shows an electron micrograph of blood cells
Genotype Blood type including some sickled cells. In regard to the presence or
A A A absence of anemia, the Hb A allele is dominant. In the het-
I / I , I /i A
erozygote, a single Hb A allele produces enough functioning
B B B
I /I , I / i B hemoglobin to prevent anemia. In regard to blood-cell shape,
A
I /I B
AB however, there is incomplete dominance, as shown by the fact
i/ i O
that, in the heterozygote, many of the cells have a slight sickle
shape. Finally, in regard to hemoglobin itself, there is codom-
In this allelic series, the alleles determine the presence inance. The alleles Hb A and HbS encode two different forms
and form of a complex sugar molecule present on the sur- of hemoglobin that differ by a single amino acid, and both
face of red blood cells. This sugar molecule is an antigen, a forms are synthesized in the heterozygote. The A and S forms
cell-surface molecule that can be recognized by the immune of hemoglobin can be separated by electrophoresis because it
system. The alleles I A and I B determine two different forms happens that they have different charges (Figure 5-5). We see
of this cell-surface molecule. However, the allele i results in that homozygous Hb A /Hb A people have one type of hemo-
no cell-surface molecule of this type (it is a null allele). In globin (A), and anemics have another (type S), which moves
the genotypes I A /i and I B /i , the alleles I A and I B are fully more slowly in the electric field. The heterozygotes have both
dominant over i. However, in the genotype I A /I B, each of types, A and S. In other words, there is codominance at the
the alleles produces its own form of the cell-surface mole- molecular level. The fascinating population genetics of the
cule, and so the A and B alleles are codominant. Hb A and HbS alleles will be considered in Chapter 20.
The human disease sickle-cell anemia illustrates the Sickle-cell anemia illustrates the arbitrariness of the
somewhat arbitrary ways in which we classify dominance. terms dominance, incomplete dominance, and codomi-
The gene concerned encodes the molecule hemoglobin, nance. The type of dominance inferred depends on the
which is responsible for transporting oxygen in blood ves- phenotypic level at which the assay is made—organismal,
sels and is the major constituent of red blood cells. There cellular, or molecular. Indeed, caution should be applied to
are two main alleles, Hb A and HbS, and the three possible many of the categories that scientists use to classify struc-
genotypes have different phenotypes, as follows: tures and processes; these categories are devised by humans
for the convenience of analysis.
Hb A /Hb A : normal; red blood cells never sickle
HbS /HbS : severe, often fatal anemia; abnormal KEY CONCEPT In general, three main types of dominance
hemoglobin causes red blood cells to can be distinguished: full dominance, incomplete dominance,
have sickle shape and codominance. The type of dominance is determined by
the molecular functions of the alleles of a gene and by the
Hb A /HbS : no anemia; red blood cells sickle only
investigative level of analysis.
under low oxygen concentrations

07_GriffitITGA12e_11478_Ch05_153_192.indd 157 03/10/19 12:01 PM

 158 C H A P T E R 5 ! ! Gene Interaction

Heterozygotes can express the protein product of both alleles Seven alleles and their interactions
in leaf patterning of clover
Sickle-cell Sickle-cell
Phenotype Unaffected anemia trait
Genotype HbA / HbA HbS / HbS HbS / HbA

Origin vv

Positions to which Migration V lV l

hemoglobins have
migrated

Hemoglobin types A S S and A V hV h V l Vh

present

FIGURE 5-5 The electrophoresis of normal and mutant hemoglobins.

Shown are results produced by hemoglobin from a person with
V fV f V lV f V hV f
sickle-cell trait (a heterozygote), a person with sickle-cell anemia, and
a healthy person. The green bands show the positions to which the
hemoglobins migrate on the starch gel.

V baV ba V lV ba V hV ba V fV ba
The leaves of clover plants show several variations on
the dominance theme. Clover is the common name for
plants of the genus Trifolium. There are many species.
Some are native to North America, whereas others grow
there as introduced weeds. Much genetic research has been V bV b V lV b V hV b V fV b V ba V b
done with white clover, which shows considerable varia-
tion among individual plants in the curious V, or chevron,
pattern on the leaves. The different chevron forms (and the
absence of chevrons) are determined by a series of seven V byV by V lV by V hV by V fV by V b a V by V b V by
alleles, as seen in Figure 5-6, which shows the many different
types of interactions possible for even one allele. In most FIGURE 5-6 Multiple alleles determine the chevron pattern on the
leaves of white clover. The genotype of each plant is shown below
practical cases, many alleles of a gene can be found together
it. There is a variety of dominance interactions. [Research by W. Ellis
in a population, constituting an allelic series. The pheno- Davies.]
types shown by the allelic combinations are many and var-
ied, reflecting the relative nature of dominance: an allele can
show dominance with one partner but not with another. function) is essential to the organism’s operation. Essential
Hence, the complexity illustrated by the ABO blood type genes are those without which an organism dies. (An exam-
system is small compared with that in a case such as clover ple of an essential gene might be a ribosomal gene without
chevrons. which no protein would be made.) Indeed, with the use of
modern DNA technology, a null mutant allele of a gene of
KEY CONCEPT A gene can take on many forms, called interest can now be made intentionally and made homo-
alleles, each caused by various mutations of the DNA zygous to see if it is lethal and under which environmental
sequence. Some mutant alleles have phenotypic impact; conditions. Lethal alleles are also useful in determining the
others do not. developmental stage at which the gene normally acts. In
this case, geneticists look for whether death from a lethal
mutant allele occurs early or late in the development of a
zygote. The phenotype associated with death can also be
Recessive lethal alleles informative in regard to gene function; for example, if a
An allele that is capable of causing the death of an organ- certain organ appears to be abnormal, the gene is likely to
ism is called a lethal allele. In the characterization of a set be expressed in that organ.
of newly discovered mutant alleles, a recessive mutation What is the diagnostic test for lethality? The test is well
(a mutation in the homozygous state) is sometimes found illustrated by one of the prototypic examples of a lethal
to be lethal. This information is potentially useful in that allele, a coat-color allele in mice (see the Model Organism
it shows that the newly discovered gene (of yet unknown box on page 159). Normal wild-type mice have coats with a

07_GriffitITGA12e_11478_Ch05_153_192.indd 158 03/10/19 12:01 PM

 MODEL ORGANISM Mus musculus
arranged in blocks in exactly the same positions as those
of humans.
Research on the Mendelian genetics of mice began early
in the twentieth century. One of the most important early con-
tributions was the elucidation of the genes that control coat
color and pattern. Genetic control of the mouse coat has pro-
vided a model for all mammals, including cats, dogs, horses,
and cattle. A great deal of work was also done on mutations
induced by radiation and chemicals. Mouse genetics has been
of great significance in medicine. A large proportion of human
genetic diseases have mouse counterparts useful for exper-
imental study (they are called “mouse models”). The mouse
has played a particularly important role in the development of
Green-glowing genetically modified mice. The jellyfish gene for green fluorescent our current understanding of the genes underlying cancer.
protein has been inserted into the chromosomes of the glowing mice. The other
The mouse genome can be modified by the insertion
mice are normal. [Eye of Science/Science Source.]
of specific fragments of DNA into a fertilized egg or into
somatic cells. The mice in the photograph have received a

T he laboratory mouse is descended from the house

mouse Mus musculus. The pure lines used today as
standards are derived from mice bred in past centuries by
jellyfish gene for green fluorescent protein (GFP) that makes
them glow green under special lights. Gene knockouts and
replacements also are possible.
mouse “fanciers.” Among model organisms, it is the one A major limitation of mouse genetics is its cost. Whereas
whose genome most closely resembles the human genome. working with a million individuals of E. coli or S. cerevisiae is
Its diploid chromosome number is 40 (compared with 46 a trivial matter, working with a million mice requires a factory-
in humans), and the genome is slightly smaller than that of size building. Furthermore, although mice do breed rapidly
humans (the human genome being 3000 Mb) and contains compared with humans, they cannot compete with microor-
approximately the same number of genes (current esti- ganisms for speedy life cycle. Hence, the large-scale selec-
mate 25,000). Furthermore, all mouse genes seem to have tions and screens necessary to detect rare genetic events
counterparts in humans. A large proportion of genes are are not possible.

rather dark overall pigmentation. A mutation called yellow The expected monohybrid ratio of 1: 2:1 would be
(a lighter coat color) shows a curious inheritance pattern. If found among the zygotes, but it is altered to a 2:1 ratio in
any yellow mouse is mated with a homozygous wild-type the progeny actually seen at birth because zygotes with a
mouse, a 1:1 ratio of yellow to wild-type mice is always lethal AY /AY genotype do not survive to be counted. This
observed in the progeny. This result suggests that a yellow hypothesis is supported by the removal of uteri from preg-
mouse is always heterozygous for the yellow allele and that nant females of the yellow ! yellow cross; one-fourth of the
the yellow allele is dominant over wild type. However, if embryos are found to be dead.
any two yellow mice are crossed with each other, the result
is always as follows: A recessive lethal allele, yellow coat
yellow ! yellow � 2 yellow, 1 wild type
3 3

Figure 5-7 shows a typical litter from a cross between

yellow mice.
How can the 2:1 ratio be explained? The results make
sense if the yellow allele is assumed to be lethal when
homozygous. The yellow allele is known to be of a coat-
color gene called A. Let’s call it AY . Hence, the results of
crossing two yellow mice are
AY /A ! AY /A
Progeny 1
4
AY /AY lethal
1 Y FIGURE 5-7 A litter from a cross between two mice heterozygous
2
A /A yellow
for the dominant yellow coat-color allele. The allele is lethal in a double
1 A /A wild type dose. Not all progeny are visible. [Anthony Griffiths.]
4

159

07_GriffitITGA12e_11478_Ch05_153_192.indd 159 03/10/19 12:01 PM

 160 C H A P T E R 5 ! ! Gene Interaction

Tailless, a recessive lethal allele in cats Geneticists commonly encounter situations in which
expected phenotypic ratios are consistently skewed in one
direction because a mutant allele reduces viability. For
example, in the cross A /a ! a /a , we predict a progeny ratio
of 50 percent A /a and 50 percent a/a, but we might con-
sistently observe a ratio such as 55 percent:45 percent or
60 percent:40 percent. In such a case, the recessive allele
is said to be sublethal because the lethality is expressed in
only some but not all of the homozygous individuals. Thus,
lethality may range from 0 to 100 percent, depending on
the gene itself, the rest of the genome, and the environment.
We have seen that lethal alleles are useful in diagnosing
the time at which a gene acts and the nature of the phenotypic
defect that kills. However, maintaining stocks bearing lethal
alleles for laboratory use is a challenge. In diploids, recessive
lethal alleles can be maintained as heterozygotes. In haploids,
heat-sensitive lethal alleles are useful. They are members of a
general class of temperature-sensitive (ts) mutations. Their
phenotype is wild type at the permissive temperature (often
FIGURE 5-8 A Manx cat. A dominant allele causing taillessness is room temperature) but mutant at some higher restrictive
lethal in the homozygous state. The phenotype of two eye colors is
unrelated to taillessness. [Gerard Lacz/NHPA/Photoshot.]
temperature. Temperature-sensitive alleles are thought to be
caused by mutations that make the protein prone to twist or
bend its shape to an inactive conformation at the restrictive
The AY allele produces effects on two characters: coat temperature. Research stocks can be maintained easily under
color and survival. In general, the term pleiotropic is used permissive conditions, and the mutant phenotype can be
for any allele that affects several properties of an organism. assayed in a subset of individuals by a switch to the restrictive
The tailless Manx phenotype in cats (Figure 5-8) also is conditions. In diploids, temperature-sensitive dominant lethal
produced by an allele that is lethal in the homozygous state. mutations also are useful. This type of mutation is lethal even
A single copy of the Manx allele, M L , severely interferes when present in a single dose, but only when the experimenter
with normal spinal development, resulting in the absence switches the organism to the restrictive temperature.
of a tail in the M L /M heterozygote. But in the M L /M L Null alleles for genes identified through genomic
homozygote, two copies of the Manx allele produces such sequencing can be made by using a variety of “reverse
an extreme abnormality in spinal development that the genetic” procedures that specifically knock out the function
embryo does not survive. of that gene. These will be described in Chapter 14.

KEY CONCEPT Some mutant alleles are lethal; that is, they KEY CONCEPT To see if a gene is essential, a null allele is
can result in the death of the organism. Lethality is most often tested for lethality.
recessive.

The yellow and M L alleles have their own phenotypes Penetrance and expressivity
in a heterozygote, but most recessive lethals are silent in the
In the analysis of single-gene inheritance, there is a natural
heterozygote. In such a situation, recessive lethality is diag-
tendency to choose mutants that produce clear Mendelian
nosed by observing the death of 25 percent of the progeny
ratios. In such cases, we can use the phenotype to distin-
at some stage of development.
guish mutant and wild-type genotypes with almost 100 per-
Whether an allele is lethal or not often depends on the
cent certainty. In these cases, we say that the mutation is
environment in which the organism develops. Whereas
100 percent penetrant into the phenotype. However, many
certain alleles are lethal in virtually any environment,
mutations show incomplete penetrance; that is, not every
others are viable in one environment but lethal in another.
individual with the genotype expresses the corresponding
Human hereditary diseases provide some examples. Cystic
phenotype. Thus, penetrance is defined as the percentage of
fibrosis and sickle-cell anemia are diseases that would be
individuals with a given allele who exhibit the phenotype
lethal without treatment. Furthermore, many of the alleles
associated with that allele.
favored and selected by animal and plant breeders would
Why would an organism have a particular genotype
almost certainly be eliminated in nature as a result of
and yet not express the corresponding phenotype? There
competition with the members of the natural population.
are several possible reasons:
The dwarf mutant varieties of grain, which are very high
yielding, provide good examples; only careful nurturing by 1. The influence of the environment. Individuals with
farmers has maintained such alleles for our benefit. the same genotype may show a range of phenotypes,

07_GriffitITGA12e_11478_Ch05_153_192.indd 160 03/10/19 12:01 PM

 5.1 Interactions Between the Alleles of a Single Gene: Variations on Dominance 161

depending on the environment. The range of phenotypes Penetrance and expressivity contrasted
for mutant and wild-type individuals may overlap: the
phenotype of a mutant individual raised in one set of Phenotypic expression
(each oval represents an individual)
circumstances may match the phenotype of a wild-type
individual raised in a different set of circumstances.
Should this matching happen, the mutant cannot be dis-
tinguished from the wild type. Variable penetrance
2. The influence of other interacting genes. Uncharacter-
ized modifiers, epistatic genes, or suppressors in the rest
of the genome (all discussed shortly) may act to prevent
Variable expressivity
the expression of the typical phenotype.
3. The subtlety of the mutant phenotype. The subtle effects
brought about by the absence of a gene function may be
difficult to measure in a laboratory situation. Variable penetrance and expressivity

A typical encounter with incomplete penetrance is FIGURE 5-10 Assume that all the individuals shown have the same
shown in Figure 5-9. In this human pedigree, we see a nor- pigment allele (P) and possess the same potential to produce pigment.
mally dominantly Effects from the rest of the genome and the environment may suppress
or modify pigment production in any one individual. The color indicates
inherited phenotype
Inferring incomplete penetrance the level of expression.
disappearing in the sec-
ond generation only to (genotype b /b) from different stocks might show very dif-
1 2 reappear in
3 the next. ferent4intensities of brown pigment
5 from light to dark. As
Another measure for penetrance, variable expressivity may be due to varia-
Q for describing the tion in the allelic constitution of the rest of the genome or to
range of phenotypic environmental factors. Figure 5-10 illustrates the distinction
expression is called between penetrance and expressivity. An example of vari-
expressivity. Expressiv- able expressivity in dogs is found in Figure 5-11.
R
ity measures the degree The phenomena of incomplete penetrance and variable
FIGURE 5-9 In this human pedigree to which a given allele expressivity can make any kind of genetic analysis substan-
6 of a dominant allele that is7not fully is expressed
8 at the phe- tially9 more difficult, including
10human pedigree analysis and
penetrant, person Q does not display notypic level; that is, predictions in genetic counseling. For example, it is often
the phenotype but passed the dominant expressivity measures the case that a disease-causing allele is not fully penetrant.
allele to at least two progeny. Because
the allele is not fully penetrant, the other
the intensity of the Thus, someone could have the allele but not show any signs
progeny (for example, R) may or may not phenotype. For exam- of the disease. If that is the case, it is difficult to give a clean
have inherited the dominant allele. ple, “brown” animals genetic bill of health to any person in a disease pedigree

FIGURE 5-11 Ten grades of piebald

Variable expressivity
spotting in beagles. Each of these dogs
has the allele SP , the allele responsible
for piebald spots in dogs. The variation
is caused by variation at other loci.

1 2 3 4

5 6 7 8

9 10

07_GriffitITGA12e_11478_Ch05_153_192.indd 161 03/10/19 12:01 PM

 162 C H A P T E R 5 ! ! Gene Interaction

(for example, person R in Figure 5-9). On the other hand, maleylacetoacetic acid; so he proposed that, in AKU, there
pedigree analysis can sometimes identify persons who do is a defect in this conversion. Consequently, homogentisic
not express but almost certainly do have a disease geno- acid builds up and is excreted. Garrod’s observations raised
type (for example, individual Q in Figure 5-9). Similarly, the possibility that the cell’s chemical pathways were under
variable expressivity can complicate counseling because the control of a large set of interacting genes. However, the
persons with low expressivity might be misdiagnosed. direct demonstration of this control was provided by the
Even though penetrance and expressivity can be quanti- later work of Beadle and Tatum on the fungus Neurospora.
fied, they nevertheless represent “fuzzy” situations because
rarely is it possible to identify the specific factors causing Biosynthetic pathways in Neurospora
variation without substantial extra research.
The landmark study by George Beadle and Edward
Tatum in the 1940s not only clarified the role of genes but
KEY CONCEPT The terms penetrance and expressivity
quantify the modification of a gene’s effect by varying environ- also demonstrated the interaction of genes in biochemi-
ment and genetic background; they measure, respectively, the cal pathways. They later received a Nobel Prize for their
percentage of cases in which the phenotype is observed and study, which marks the beginning of all molecular biology.
its severity. Beadle and Tatum did their work on the haploid fungus
Neurospora, which we have met in earlier chapters. Their
plan was to investigate the genetic control of cellular chem-
We now turn to the approaches that can be used to istry. In what has become the standard forward genetic
detect the interaction between two or more loci. approach, they first irradiated Neurospora cells to produce
mutations and then tested cultures grown from ascospores
for interesting mutant phenotypes relevant to biochemical
5.2 INTERACTION OF GENES function. They found numerous mutants that had defec-
IN!PATHWAYS tive nutrition. Specifically, these mutants were auxotrophic
mutants, meaning that the mutants would not grow unless
their medium contained one or more specific cellular build-
LO 5.4 Describe the hypotheses proposed to explain various
types of gene interaction at the molecular level. ing blocks. Whereas wild-type Neurospora can use its cellular
biochemistry to synthesize virtually all its cellular compo-
Genes act by controlling cellular chemistry. Early in the nents from the inorganic nutrients and a carbon source in the
twentieth century, Archibald Garrod, an English physician medium, auxotrophic mutants cannot. In order to grow, such
( Figure 5-12), made the first observation supporting this mutants require a nutrient to be supplied (a nutrient that a
insight. Garrod noted that several recessive human diseases wild-type fungus is able to synthesize for itself), suggesting
show defects in what is called metabolism, the general set that the mutant is defective for some normal synthetic step.
of chemical reactions taking place in an organism. This As their first step, Beadle and Tatum confirmed that each
observation led to the notion that such genetic diseases are mutation that generated a nutrient requirement was inherited
“inborn errors of metabolism.” Garrod worked on a dis- as a single-gene mutation because each gave a 1:1 ratio when
ease called alkaptonuria (AKU), or black urine disease. He crossed with a wild type (remember, Neurospora is a haploid
discovered that the substance responsible for black urine organism). Letting aux represent an auxotrophic mutation,
was homogentisic acid, which is present in high amounts + ! aux
and secreted into the urine in AKU patients. He knew that,
#
in unaffected people, homogentisic acid is converted into
progeny: 1 + and 1 aux
2 2

Discoverer of inborn errors Their second step was to classify the specific nutri-
of metabolism tional requirement of each auxotroph. Some would grow
only if proline was supplied, others methionine, others
pyridoxine, others arginine, and so on. Beadle and Tatum
decided to focus on arginine auxotrophs. They found that
the genes that mutated to give arginine auxotrophs mapped
to three different loci on three separate chromosomes. (To
determine whether their collection of arginine auxotrophs
resulted from mutation in the same gene or multiple differ-
ent genes, Beadle and Tatum methodically analyzed pairs
of mutants using the complementation test, discussed in
Section 5.3.) Let’s call the genes at the three loci the arg-1,
arg-2, and arg-3 genes. A key breakthrough was Beadle and
Tatum’s discovery that the auxotrophs for each of the three
FIGURE 5-12 British physician Archibald loci differed in their response to the structurally related
Garrod (1857–1936). [SPL/Science Source.] compounds ornithine and citrulline (Figure 5-13). The arg-1

07_GriffitITGA12e_11478_Ch05_153_192.indd 162 03/10/19 12:01 PM

 5.2 Interaction of Genes in!Pathways 163

Arginine and its chemical relatives This brilliant model, which was initially known as
the one-gene–one-enzyme hypothesis, was the source of
NH2 NH2 the first exciting insight into the functions of genes: genes
C"O C " NH somehow were responsible for the function of enzymes,
NH2 NH NH and each gene apparently controlled one specific enzyme
in a series of interconnected steps in a biochemical path-
(CH2)3 (CH2)3 (CH2)3
way. Other researchers obtained similar results for other
CHNH2 CHNH2 CHNH2 biosynthetic pathways, and the hypothesis soon achieved
COOH COOH COOH general acceptance. All proteins, whether or not they are
Ornithine Citrulline Arginine enzymes, also were found to be encoded by genes, and
so the phrase was refined to become the one-gene–one-
FIGURE 5-13 The chemical structures polypeptide hypothesis. (Recall that a polypeptide is the
of arginine and the structurally related
simplest type of protein, a single chain of amino acids.)
compounds citrulline and ornithine.
It soon became clear that a gene encodes the physical
mutants grew when supplied with any one of the chemi- structure of a protein, which in turn dictates its function.
cals ornithine, citrulline, or arginine. The arg-2 mutants Beadle and Tatum’s hypothesis became one of the great
grew when given arginine or citrulline but not ornithine. unifying concepts in biology because it provided a bridge
The arg-3 mutants grew only when arginine was supplied. that brought together the two major research areas of
These results are summarized in Table 5-1. genetics and biochemistry.
Cellular enzymes were already known to interconvert We must add parenthetically that, although the great
such related compounds. On the basis of the properties of the majority of genes encode proteins, some are known to
arg mutants, Beadle and Tatum and their colleagues proposed encode RNAs that have special functions. All genes are
a biochemical pathway for such conversions in Neurospora: transcribed to make RNA. Protein-encoding genes are tran-
scribed to messenger RNA (mRNA), which is then!trans-
enzyme X enzyme Y
precursor � ���� � ornithine � ���� � lated into protein. However, the RNA encoded by a
enzyme Z
citrulline � ���� � arginine minority of genes is never translated into protein because
the RNA itself has a unique function. These are called
This pathway nicely explains the three classes of mutants functional RNAs. Some examples are transfer RNAs, ribo-
shown in Table 5-1. Under the model, the arg-1 mutants somal RNAs, and small cytoplasmic RNAs; more about
have a defective enzyme X, and so they are unable to con- them will be covered in later chapters.
vert the precursor into ornithine as the first step in produc-
ing arginine. However, they have normal enzymes Y and Z, KEY CONCEPT Chemical synthesis in cells is by pathways of
and so the arg-1 mutants are able to produce arginine if sup- sequential steps catalyzed by enzymes. The genes encoding
plied with either ornithine or citrulline. Similarly, the arg-2 the enzymes of a specific pathway constitute a functionally
mutants lack enzyme Y, and the arg-3 mutants lack enzyme interacting subset of the genome.
Z. Thus, a mutation at a particular gene is assumed to inter-
fere with the production of a single enzyme. The defective
enzyme creates a block in some biosynthetic pathway. The Gene interaction in other types
block can be circumvented by supplying to the cells any com- of pathways
pound that normally comes after the block in the pathway. The notion that genes interact through pathways is a
We can now diagram a more complete biochemical model: powerful one that finds application in all organisms. The
arg-1+ arg-2+ Neurospora arginine pathway is an example of a synthetic
# # pathway, a chain of enzymatic conversions that synthe-
precursor � ���� enzyme X
� ornithine � ����
� enzyme Y sizes essential molecules. We can extend the idea again to
a human case already introduced, the disease phenylke-
arg-3+
tonuria (PKU), which is caused by an autosomal recessive
#
allele. This disease results from an inability to convert phe-
enzyme Z
citrulline � ���� � arginine nylalanine into tyrosine. As a result of the block, phenylal-
anine accumulates and is spontaneously converted into a
TABLE 5-1 Growth of arg Mutants in Response to toxic compound, phenylpyruvic acid. The PKU gene is part
Supplements of a metabolic pathway like the Neurospora arginine path-
Supplements way, a section of which is shown in Figure 5-14. The illustra-
Mutant Ornithine Citrulline Arginine tion includes several other diseases caused by blockages in
steps in this pathway (including alkaptonuria, the disease
arg-1 + + +
investigated by Garrod).
arg-2 % + +
Another type of pathway is a signal-transduction
arg-3 % % + pathway. This type of pathway is a chain of complex sig-
Note: A plus sign means growth; a minus sign means no growth. nals, from the environment to the internal components

07_GriffitITGA12e_11478_Ch05_153_192.indd 163 03/10/19 12:01 PM

 164 C H A P T E R 5 ! ! Gene Interaction

A synthetic pathway and associated diseases analysis of mutants with aberrant mating response, and
the steps were pieced together by using the approaches in
Dietary protein the next section. A mutation at any one of these steps may
disrupt the mating process.
Developmental pathways comprise the steps by which
a zygote becomes an adult organism. This process involves
Phenylalanine
(Phe)
If [Phe] high many genetically controlled steps, including establishment of
the anterior-posterior and dorsal-ventral axes, laying down
Phenylpyruvic
acid the basic body plan of organs, and tissue differentiation and
Phe hydroxylase PKU
movement. These steps can require gene regulation and sig-
nal transduction. Developmental pathways will be taken up
Albinism Cretinism in detail in Chapter 13, but the interaction of genes in these
Tyrosine pathways is analyzed in the same way, as we will see next.
(Tyr)
Melanin Thyroxine
KEY CONCEPT Gene interaction occurs in any cellular
pathway, such as biosynthetic, signal transduction, and
Tyr aminotransferase developmental.

Hydroxyphenylpyruvic acid
(HPA)
5.3 INFERRING GENE
HPA oxidase Tyrosinosis INTERACTIONS
LO 5.2 Recognize phenotypic ratios diagnostic of the
Homogentisic acid presence of a lethal allele.
(HA)

HA oxidase Alkaptonuria LO 5.5 Determine whether two mutations are in the same
gene or in different genes, using progeny ratios or
using complementation tests.
Maleylacetoacetic
acid
LO 5.6 Infer how two genes may be interacting, based on
modified Mendelian ratios.

CO2 + H2O
LO 5.7 For known cases of gene interaction, predict
progeny ratios in crosses.
FIGURE 5-14 A section of the phenylalanine metabolic pathway in
humans, including diseases associated with enzyme blockages. The
The genetic approach that reveals the interacting genes for
disease PKU is produced when the enzyme phenylalanine hydroxylase
malfunctions. Accumulation of phenylalanine results in an increase a particular biological property is briefly as follows:
in phenylpyruvic acid, which interferes with the development of the
Step 1. Obtain many single-gene mutants and test for
nervous system.
dominance.
Step 2. Test the mutants for allelism—are they at one or
of the cell, that result in activation of cellular responses. several loci?
These pathways are crucial to the proper function of an
Step 3. Combine the mutants in pairs to form double
organism. One of the best-understood signal-transduction
mutants to see if the genes interact.
pathways was worked out from a genetic analysis of the
mating response in baker’s yeast. Two mating types, deter- Gene interaction is inferred from the phenotype of the dou-
mined by the alleles MATa and MAT&, are necessary for ble mutant: if the genes interact, then the phenotype differs
yeast mating to occur. When a cell is in the presence of from the simple combination of both single-gene mutant
another cell of opposite mating type, it undergoes a series phenotypes. If mutant alleles from different genes interact,
of changes in shape and behavior to prepare for mating. then we infer that the wild-type genes interact normally as
This mating response is triggered when a mating pher- well. In cases in which the two mutants interact, a modified
omone (hormone) is released by a cell of the opposite 9:3:3:1 Mendelian ratio will often result.
mating type and binds to a membrane receptor on the A procedure that must be carried out before testing
receiving cell. This signal promotes the sequential action of interactions is to determine whether each mutation is of
a set of genes, which ultimately activates the transcription a different locus (step 2 above). The mutant screen could
of mating-specific genes that enable the cell to mate. This have unintentionally favored certain genes. Thus, the set of
set of genes was discovered through a standard interaction gene loci needs to be defined, as shown in the next section.

07_GriffitITGA12e_11478_Ch05_153_192.indd 164 03/10/19 12:01 PM

 5.3 Inferring Gene Interactions 165

KEY CONCEPT The complementation test is a standard way Let’s illustrate the complementation test with an example
of determining whether or not two recessive mutations are in from harebell plants (genus Campanula). The wild-type flower
the same gene. The mutations are united in one cell, and if the color of this plant is blue. Let’s assume that, from a mutant
cell shows the wild-type phenotype, the mutations have com- hunt, we have obtained three white-petaled mutants and that
plemented and must be in different genes. they are available as homozygous pure-breeding strains. They
all look the same, and so we do not know a priori whether
they are genetically identical. We will call the mutant strains
Sorting mutants using the $, £, and ¥ to avoid any symbolism using letters, which might
complementation test imply dominance. When crossed with wild type, each mutant
gives the same results in the F1 and F2 as follows:
How is it possible to decide whether two mutations belong to
3 1
the same gene? There are several ways. First, each mutant allele white $ ! blue � F1 , all blue � F2 , 4
blue, 4
white
could be mapped. Then, if two mutations map to two different 3 1
white £ ! blue � F1 , all blue � F2 , 4
blue, 4
white
chromosomal loci, they are likely of different genes. However,
3 1 !white
this approach is time consuming on a large set of mutations. A white ¥ ! blue � F1 , all blue � F2 ,! 4
blue, 4
quicker approach often used is the complementation test.
In each case, the results show that the mutant condition
In a diploid, the complementation test is performed by
is determined by the recessive allele of a single gene. How-
intercrossing two individuals that are homozygous for differ-
ever, are they three alleles of one gene, of two genes, or of
ent recessive mutations. The next step is to observe whether
three genes? Because the mutants are recessive, the question
the progeny have the wild-type phenotype. If the progeny
can be answered by the complementation test, which asks if
are wild type, the two recessive mutations must be in dif-
the mutants complement one another.
ferent genes because the respective wild-type alleles pro-
Let us intercross the mutants to test for complemen-
vide wild-type function. In this case, the two mutations are
tation. Assume that the results of intercrossing mutants
said to have complemented. Consider two genes a1 and a2,
$, £, and ¥ are as follows:
named after their mutant alleles. We can represent the het-
erozygotes as follows, depending on whether the genes are white $ ! white £ � F1 , all white
on the same chromosome or are on different chromosomes: white $ ! white ¥ � F1 , all blue
Same chromosome: white £ ! white ¥ � F1 , all blue
a1 1 From this set of results, we can conclude that mutants $ and
£ must be caused by alleles of one gene (say, w1) because they
1 a2 do not complement, but ¥ must be caused by a mutant allele of
another gene (w2) because ¥ complements both $ and £.
Different chromosomes:
a1 1 KEY CONCEPT When two independently derived recessive
mutant alleles producing similar recessive phenotypes fail to
complement, they must be alleles of the same gene.
1 a2

You can see that each locus has one wild-type allele to pro- How does complementation work at the molecular
vide wild-type function, resulting in wild-type progeny. level? The normal blue color of the harebell flower is caused
However, if the progeny are not wild type, then the by a blue pigment called anthocyanin. Pigments are chem-
recessive mutations must be alleles of the same gene. icals that absorb certain colors of light; in regard to the
Because both alleles of the gene are mutants, there is no
wild-type allele to provide wild-type function. These alleles Harebell plant
could have different mutant sites within the same gene, but
they would both be nonfunctional. Consider two recessive
mutations, a� and a��, of a gene whose wild type allele is a+.
The heterozygote a� /a�� would be
a!

a"
= mutation
+
Since there is no a allele to provide wild-type function, the
progeny will not be wild type.
At the operational level, complementation is defined as
the production of a wild-type phenotype when two haploid
genomes bearing different recessive mutations are united in Flowers of the harebell plant (Campanula species).
the same cell. [Gregory G. Dimijian, M.D./Science Source.]

07_GriffitITGA12e_11478_Ch05_153_192.indd 165 03/10/19 12:02 PM

 166 C H A P T E R 5 ! ! Gene Interaction

harebell, the anthocyanin absorbs all wavelengths except The!nuclei in a heterokaryon do not generally fuse. In one
blue, which is reflected into the eye of the observer. How- sense, this condition is a “mimic” diploid.
ever, this anthocyanin is made from chemical precursors Assume that, in different strains, there are mutations
that are not pigments; that is, they do not absorb light of in two different genes conferring the same mutant pheno-
any specific wavelength and simply reflect back the white type—for example, an arginine requirement. We will call
light of the sun to the observer, giving a white appearance. these genes arg-1 and arg-2. The genotypes of the two strains
The blue pigment is the end product of a series of biochem- can be represented as arg -1 ! arg -2+ and arg -1+ ! arg -2 .
ical conversions of nonpigments. Each step is catalyzed by a These two strains can be fused to form a heterokaryon with
specific enzyme encoded by a specific gene. We can explain the two nuclei in a shared cytoplasm:
the results with a pathway as follows:
Nucleus 1 is arg -1 ! arg -2+
gene w1! gene w2! Nucleus 2 is arg -1+ ! arg -2
Because gene products are made in a common cytoplasm,
the two wild-type alleles can exert their dominant effect and
enzyme 1 enzyme 2
cooperate to produce a heterokaryon of wild-type pheno-
type. In other words, the two mutations complement, just as
precursor 1 precursor 2 blue anthocyanin they would in a diploid. If the mutations had been alleles of
the same gene, there would have been no complementation.
A homozygous mutation in either of the genes will lead
to the accumulation of a precursor that will simply make
Analyzing double mutants of random
the plant white. Now the mutant designations could be
written as follows:
mutations
Recall that, to learn whether two genes interact, we need
$ w1$ /w1$ ! w2+ /w2+
to assess the phenotype of the double mutant to see if it is
£ w1£ /w1£ ! w2+ /w2+ different from the combination of both single mutations.
¥ w1+ /w1+ ! w2¥ /w2¥ The double mutant is obtained by intercrossing. The F1 is
obtained as part of the complementation test; so with the
However, in practice, the subscript symbols would be
assumption that complementation has been observed, sug-
dropped and the genotypes would be written as follows:
gesting different genes, the F1 is selfed or intercrossed to
$ w1/w1 ! w2+ /w2+ obtain an F2 homozygous for both mutations. This double
£ w1/w1 ! w2+ /w2+ mutant may then be identified by looking for Mendelian
ratios. For example, if a standard 9:3:3:1 Mendelian ratio
¥ w1+ /w1+ ! w2 /w2
is obtained, the phenotype present in only 1/16 of the prog-
Hence, an F1 from $ ! £ will be eny represents the double mutant (the “1” in 9:3:3:1). In
cases of gene interaction, however, the phenotype of the
w1/w1 ! w2+ /w2+
double mutant may not be distinct but will match that of
These F1 plants will have two defective alleles for w1 and one of the single mutants. In this case, a modified Mende-
will therefore be blocked at step 1. Even though enzyme 2 is lian ratio will result, such as 9:3: 4 or 9: 7.
fully functional, it has no substrate on which to act; so no blue The standard 9:3:3:1 Mendelian ratio is the simplest
pigment will be produced, and the phenotype will be white. case, expected if there is no gene interaction and if the two
The F1 plants from the other crosses, however, will have the mutations under test are on different chromosomes. This
wild-type alleles for both of the enzymes needed to take the 9:3:3:1 ratio is the null hypothesis: any modified Mende-
intermediates to the final blue product. Their genotypes will be lian ratio representing a departure from this null hypothesis
would be informative, as the following examples will show.
w1+ /w1 ! w2+ /w2
KEY CONCEPT A range of modified 9: 3: 3:1 F1 ratios can
Hence, we see that complementation is actually a result
reveal specific types of gene interaction.
of the cooperative interaction of the wild-type alleles of the
two genes. Figure 5-15 summarizes the interaction of the
complementing and noncomplementing white mutants at The 9 : 3 : 3 :1 ratio: no gene interaction As a base-
the genetic and cellular levels. line, let’s start with the case in which two mutated genes
In a haploid organism, the complementation test can- do not interact, a situation where we expect the 9:3:3:1
not be performed by intercrossing. In fungi, an alternative ratio. Let’s look at the inheritance of skin coloration
method brings mutant alleles together to test complemen- in corn snakes. The snake’s natural color is a repeat-
tation: fusion resulting in a heterokaryon ( Figure 5-16). ing black-and-orange camouflage pattern, as shown in
Fungal cells fuse readily. When two different strains fuse, Figure 5-17a. The phenotype is produced by two separate
the haploid nuclei from the different strains occupy one pigments, both of which are under genetic control. One
cell, which is the heterokaryon (Greek; different kernels). gene determines the orange pigment, and the alleles that

07_GriffitITGA12e_11478_Ch05_153_192.indd 166 03/10/19 12:02 PM

 5.3 Inferring Gene Interactions 167

FIGURE 5-15 Three phenotypically

The molecular basis of genetic complementation
identical white harebell mutants—
$, £, and ¥—are intercrossed.
Wild type
Mutations in the same gene (such as
$ and £) cannot complement because
the F1 has one gene with two mutant
alleles. The pathway is blocked and
the flowers are white. When the
mutations are in different genes (such
+ + as £ and ¥), there is complementation
by the wild-type alleles of each gene
in the F1 heterozygote. Pigment is
+ +
synthesized and the flowers are blue.
w1 w2
gene gene

Mutant "$" Mutant "£" Mutant "¥"

"$" + "£" + + "¥"

"$" + "£" + + "¥"

w1 w2 w1 w2 w1 w2
gene gene gene gene gene gene

P White $ × White £ White £ × White ¥

F1

No
complementation Complementation
$ + + ¥

£ + £ +

Enzyme 1 Enzyme 2
No substrate Enzyme 2
!
Colorless No Colorless Colorless
White Blue
precursor 1 precursor 2 precursor 1 precursor 2

Block (no enzyme 1)

Mutation in the same gene Mutation in different genes

we will consider are o+ (presence of orange pigment) and black pigment (Figure 5-17c). The double homozygous
o (absence of orange pigment). Another gene determines recessive o /o ; b /b is albino (Figure 5-17d). Notice, how-
the black pigment, and its alleles are b+ (presence of black ever, that the faint pink color of the albino is from yet
pigment) and b (absence of black pigment). These two another pigment, the hemoglobin of the blood that is visi-
genes are unlinked. The natural pattern is produced by ble through this snake’s skin when the other pigments are
the genotype o+ /% ; b+ /% . (The dash represents the pres- absent. The!albino snake also clearly shows that there is
ence of either allele.) A snake that is o /o ; b+ /% is black another element to the skin-pigmentation pattern in addi-
because it lacks the orange pigment (Figure 5-17b), and tion to pigment: the repeating motif in and around which
a snake that is o+ /% ; b /b is orange because it lacks the pigment is deposited.

07_GriffitITGA12e_11478_Ch05_153_192.indd 167 03/10/19 12:02 PM

 168 C H A P T E R 5 ! ! Gene Interaction

Testing complementation by using a heterokaryon

arg-1 cells, defective for arg-2 cells, defective for

one specific enzyme in a different enzyme in
arginine synthetic pathway arginine synthetic pathway

Fusion

Heterokaryon grows
without arginine

FIGURE 5-16 A heterokaryon of Neurospora and similar fungi mimics a diploid state. When
vegetative cells fuse, haploid nuclei share the same cytoplasm in a heterokaryon. In this example,
haploid nuclei with mutations in different genes in the arginine synthetic pathway complement to
produce a Neurospora culture that no longer requires arginine. Functional enzyme produced by
arg-1+ is shown in purple, and functional enzyme produced by arg-2+ is shown in red.

If a homozygous orange and a homozygous black snake Here, however, an F2 shows a standard 9:3:3:1 ratio:
are crossed, the F1 is wild type (camouflaged), demonstrat-
� o+ /o ; b+ /b ! � o+ /o ; b+ /b
ing complementation:
(camouflaged) (camouflaged)
� o+ /o+ ; b /b ! � o /o ; b+ /b+ #
(orange) (black) F2 9 o+ /– ; b+ / – (camouflaged)
# +
3 o /– ; b /b (orange)
F1 o+ /o ; b+ /b
(camouflaged) 3 o /o ; b+ / – (black)
1 o /o ; b /b (albino)
The 9:3:3:1 ratio is produced because the two pigment
Independently synthesized and inherited pigments genes act independently at the cellular level.

}
(a) (b) b +
precursor ���� black pigment
camouflaged
+
o
precursor ���� orange pigment
If the presence of one mutant makes one pathway fail,
the other pathway is still active, producing the other pig-
ment color. Only when both mutants are present do both
pathways fail, and no pigment of any color is produced.
The 9:7 ratio: genes in the same pathway The F2 ratio
from the harebell dihybrid cross shows both blue and white
(c) (d) plants in a ratio of 9: 7. How can such results be explained?
Introduction to Genetic Analysis, 11e The 9: 7 ratio is clearly a modification of the dihybrid 9:3:3:1
Figure 06.13 #619
04/29/14 ratio with the 3:3:1 combined to make 7; hence, some kind
Dragonfly Media Group of interaction is inferred. The cross of the two white lines and
subsequent generations can be represented as follows:
w1/w1 ; w2+ /w2+ (white) ! w1+ /w1+ ; w2 /w2 (white)
#
F1 w1 /w1 ; w2+ /w2 (blue)
+

w1 /w1 ; w2+ /w2 ! w1+ /w1 ; w2+ /w2

+

#
FIGURE 5-17 In corn snakes, combinations of orange and black
F2 9 w1+ /– ; w2+ /– (blue) 9
pigments determine the four phenotypes shown. (a) A wild-type black-

}
and-orange camouflaged snake synthesizes both black and orange 3 w1+ / – ; w2 /w2 (white)
pigments. (b) A black snake does not synthesize orange pigment. +
3 w1/w1 ; w2 / – (white) 7
(c)!An orange snake does not synthesize black pigment. (d) An albino
snake synthesizes neither black nor orange pigment. [Anthony Griffiths.] 1 w1/w1 ; w2 /w2 (white)

07_GriffitITGA12e_11478_Ch05_153_192.indd 168 03/10/19 12:02 PM

 5.3 Inferring Gene Interactions 169

Clearly, in this case, the only way in which a 9: 7 ratio the F1 dihybrid will also result in a 9: 7 phenotypic ratio in
is possible is if the double mutant has the same phenotype the F2:
as the two single mutants. Hence, the modified ratio consti-
tutes a way of identifying the double mutant’s phenotype. Functional a +
Furthermore, the identical phenotypes of the single and Proportion Genotype protein Ratio
double mutants suggest that each mutant allele controls a 9 + +
r /% ; a /% Yes 9
16
different step in the same pathway. The results show that

}
a plant will have white petals if it is homozygous for the 3
16
r + /% ; a /a No
recessive mutant allele of either gene or both genes. To have 3 r /r ; a+ /% No 7
the blue phenotype, a plant must have at least one copy of 16

the dominant, wild-type allele of both genes because both 1

16
r /r ; a /a No
are needed to complete the sequential steps in the pathway.
No matter which is absent, the same pathway fails, pro-
KEY CONCEPT A 9:7 F2 ratio suggests interacting genes in
ducing the same phenotype. Thus, three of the genotypic
the same pathway; absence of either gene function leads to
classes will produce the same phenotype, and so, overall,
absence of the end product of the pathway.
only two phenotypes result.
The example in harebells entailed different steps in a
synthetic pathway. Similar results can come from gene reg- The 9:3:4 ratio: recessive epistasis A 9:3: 4 ratio
ulation. A regulatory gene often functions by producing a in!the F2 suggests a type of gene interaction called epistasis.
protein that binds to a regulatory site upstream of a target This word means “stand upon,” referring to the situation
gene, facilitating the transcription of the gene (Figure!5-18). in which a double mutant shows the phenotype of one
In the absence of the regulatory protein, the target gene mutation but not the other. The overriding mutation is epi-
would be transcribed at very low levels, inadequate for cel- static, whereas the overridden one is hypostatic. Epistasis
lular needs. Let’s cross a pure line r /r defective for the reg- also results from genes being in the same pathway. In a
ulatory protein to a pure line a /a defective for the target simple synthetic pathway, the epistatic mutation is carried
protein. The cross is r /r ; a+ /a+ ! r + /r + ; a /a. The r + /r ; a+ /a by a gene that is farther upstream (earlier in the pathway)
dihybrid will show complementation between the mutant than the gene of the overridden mutation (Figure 5-19). The
genotypes because both r + and a+ are present, permitting mutant phenotype of the upstream gene takes precedence,
normal transcription of the wild-type allele. When selfed, no matter what is taking place later in the pathway.

FIGURE 5-18 The r + gene

Interaction between a regulatory encodes a regulatory protein, and
protein and its target the a+ gene encodes a structural
protein. Both must be normal for
Regulatory gene Gene for protein A Protein product of gene a a functional (“active”) structural
protein to be synthesized.
r+ a+

(a) Wild-type protein A

Normal produced

(b) r a+
Mutation in
the gene that No protein A
encodes the produced
regulatory Nonfunctional
protein regulatory protein

(c) r+ a
Mutation in
the gene that Mutant protein A
encodes the produced
structural
protein

r a
(d)
Mutation in No protein A
both genes produced

07_GriffitITGA12e_11478_Ch05_153_192.indd 169 03/10/19 12:02 PM

 170 C H A P T E R 5 ! ! Gene Interaction

A model for recessive epistasis In the F2 , the 9:3: 4 phenotypic ratio is diagnostic of
recessive epistasis. As in the preceding case, we see, again,
that the ratio tells us what the phenotype of the double
must be, because the 16 4 component of the ratio must be a

grouping of one single mutant class ( 16 3

) plus the double
mutant class ( 16 ). Hence, the double mutant expresses only
1

Dihybrid w +/ w ; m + /m
one of the two mutant phenotypes; so, by definition, white
must be epistatic to magenta. (To find the double mutant
within the group, white F2 plants would have to be individ-
ually testcrossed.) This interaction is called recessive epis-
Selfed tasis because a recessive phenotype (white) overrides the
other phenotype. Dominant epistasis will be considered in
the next section.
9
16
w + / – ; m + / – Both enzymes active At the cellular level, we can account for the recessive
w+ m+ epistasis in Collinsia by the following type of pathway (see
Enzyme 1 Enzyme 2 9
also Figure 5-19).
+ +
gene w gene m
colorless ����� � magenta ����� � blue
Notice that the epistatic mutation occurs in a step in the
3
w +/ – ; m/m Blocked at second enzyme pathway leading to blue pigment; this step is upstream of
16
w+ the step that is blocked by the masked mutation.
Another informative case of recessive epistasis is the
Enzyme 1 3 yellow coat color of some Labrador retriever dogs. Two
alleles, B and b, stand for black and brown coats, respec-
tively. The two alleles produce black and brown melanin.
3 The allele e of another gene is epistatic on these alleles,
16
w/w ; m + / – Blocked at first enzyme
giving a yellow coat ( Figure 5-20). Therefore, the geno-
m+
types B /% ; e /e and b /b ; e /e both produce a yellow phe-
Enzyme 2 notype, whereas B /% ; E / % and b /b ; E /% are black and
No substrate brown, respectively. This case of epistasis is not caused by
an upstream block in a pathway leading to dark pigment.
4
1 Yellow dogs can make black or brown pigment, as can be
w/w ; m/m Blocked at first enzyme
16 seen in their noses and lips. The action of the allele e is
to prevent the deposition of the pigment in hairs. In this
case, the epistatic gene is developmentally downstream; it
represents a kind of developmental target that must be of
FIGURE 5-19 Wild-type alleles of two genes (w + and m+ ) encode E!genotype before pigment can be deposited.
enzymes catalyzing successive steps in the synthesis of a blue petal
pigment. Homozygous m /m plants produce magenta flowers, and
KEY CONCEPT Epistasis is inferred when a mutant allele of
homozygous w /w plants produce white flowers. The double mutant
w /w ; m /m also produces white flowers, indicating that white is epistatic to one gene masks the expression of a mutant allele of another
magenta. gene and expresses its own phenotype instead.

Let’s look at an example concerning petal-pigment synthe-

In fungi, tetrad analysis is useful in identifying a double
sis in the plant blue-eyed Mary (Collinsia parviflora). From the
mutant. For example, an ascus containing half its products
blue wild type, we’ll start with two pure mutant lines, one with
as wild type must contain double mutants. Consider the
white (w /w) and the other with magenta petals (m /m). The w
cross
and m genes are not linked. The F1 and F2 are as follows:
a ! b+ ! a+ ! b
w /w ; m+ /m+ (white) ! w+ /w+ ; m /m (magenta)
F1 w+ /w ; m+ /m (blue) In some proportion of progeny, the alleles a and b will
# segregate together (a nonparental ditype ascus). Such a tet-
w+ /w ; m+ /m ! w+ /w ; m+ /m rad will show the following phenotypes:
# wild type a+ ! b+! double mutant a!b
F2 9 w+ /– ; m+ /– (blue) 9 wild type +
a !b +!
double mutant a!b
3 w+ /– ; m /m (magenta) 3
Hence, the double mutant must be the non-wild-type
3 w /w ; m+ / –
1 w /w ; m /m
(white)
(white) } 4 genotype and can be assessed accordingly. If the pheno-
type is the a phenotype, then b is being overridden; if the

07_GriffitITGA12e_11478_Ch05_153_192.indd 170 03/10/19 12:02 PM

 5.3 Inferring Gene Interactions 171

Recessive epistasis due to the yellow coat mutation

(a) (b) (c)

FIGURE 5-20 Three different coat colors in Labrador retrievers. Two alleles B and b of a pigment
gene determine (a) black and (b) brown, respectively. At a separate gene, E allows color deposition in
the coat, and e /e prevents deposition, resulting in (c) the gold phenotype. Part c illustrates recessive
epistasis. [Anthony Griffiths.]

phenotype is the b phenotype, then a is being overridden. If Dominant epistasis due to a white mutation
both phenotypes are present, then there is no epistasis.

The 12 : 3 :1 ratio: dominant epistasis In foxgloves

(Digitalis purpurea), two genes interact in the pathway that
determines petal coloration. The two genes are unlinked.
One gene affects the intensity of the red pigment in the
petal; allele d results in the light red color seen in natural
populations of foxgloves, whereas D is a mutant allele
that produces dark red color (Figure 5-21). The other gene
determines in which cells the pigment is synthesized: allele
w allows synthesis of the pigment throughout the petals as
in the wild type, but the mutant allele W confines pigment
synthesis to the small throat spots. If we self a dihybrid D/ d ;
W / w, then the F2 ratio is as follows:
9 D / – ; W /– (white with spots)
3 d /d ; W /– (white with spots)
} 12 FIGURE 5-21 In foxgloves, D and d cause dark and light pigments,
respectively, whereas the epistatic W restricts pigment to the throat
spots. [Anthony Griffiths.]
3 D / – ; w /w (dark red) 3
1 d /d ; w /w (light red) 1
built up by the laborious combination of candidate muta-
The ratio tells us that the dominant allele W is epistatic, tions two at a time. However, for our next type of gene
producing the 12:3:1 ratio. The 12 16
component of the ratio interaction, the experimenter can readily select interesting
must include the double mutant class ( 16 9
), which is clearly mutant alleles. A suppressor is a mutant allele of a gene
white in phenotype, establishing the epistasis of the domi- that reverses the effect of a mutation of another gene,
nant allele W. The two genes act in a common developmen- resulting in a wild-type or near-wild-type phenotype.
tal pathway: W prevents the synthesis of red pigment but Suppression implies that the target gene and the suppres-
only in a special class of cells constituting the main area sor gene normally interact at some functional level in their
of the petal; synthesis is allowed in the throat spots. When wild-type states. For example, assume that an allele a+ pro-
synthesis is allowed, the pigment can be produced in either duces the normal phenotype, whereas a recessive mutant
high or low concentrations. allele a results in abnormality. A recessive mutant allele s
at another gene suppresses the effect of a, and so the gen-
KEY CONCEPT Genetic analysis of gene interaction works in otype a /a ! s /s will have the wild-type (a+ -like) phenotype.
both directions. (1) Specific progeny ratios can be used to infer Suppressor alleles sometimes have no effect in the absence
gene interaction. (2) In a known case of gene interaction, ensuing
of the other mutation; in such a case, the phenotype of
progeny ratios can be predicted.
a+ /a+ ! s /s would be wild type. In other cases, the suppres-
sor allele produces its own abnormal phenotype.
Suppressors It is not easy to specifically select or screen Screening for suppressors is quite straightforward.
for epistatic interactions, and cases of epistasis have to be Start with a mutant in some process of interest, expose this

07_GriffitITGA12e_11478_Ch05_153_192.indd 171 03/10/19 12:02 PM

 172 C H A P T E R 5 ! ! Gene Interaction

mutant to mutation-causing agents such as high-energy radi- A molecular mechanism for suppression
ation, and screen the descendants for wild types. In haploids
such as fungi, screening is accomplished by simply plating
mutagenized cells and looking for colonies with wild-type
phenotypes. Most wild types arising in this way are merely m+ s+
reversals of the original mutational event and are called
revertants. However, some will be “pseudorevertants,” dou- Wild
type Active protein
ble mutants in which one of the mutations is a suppressor. complex
Revertant and suppressed states can be distinguished by
appropriate crossing. For example, in yeast, the two results
m s+
would be distinguished as follows:
First
true revertant a+ ! standard wild-type a+ mutation
Inactive
#
Progeny all a+
m s
suppressed mutant a ! s ! standard wild-type a+ ! s+ Second
# mutation
acting as
Progeny a+ ! s+ wild type suppressor
Active protein
complex
a+ ! s wild type
a ! s+ original mutant m+ s
a!s wild type (suppressed) Suppressor
mutation
The appearance of the original mutant phenotype iden- alone Inactive
tifies the parent as a suppressed mutant.
In diploids, suppressors produce various modified F2
ratios, which are useful in confirming suppression. Let’s
look at a real-life example from Drosophila. The recessive FIGURE 5-22 A first mutation alters the binding site of one protein so
allele pd results in purple eye color when unsuppressed. that it can no longer bind to a partner. A suppressor mutation in the partner
A recessive allele su has no detectable phenotype itself but alters the binding site so that both proteins are able to bind once again.
suppresses the unlinked recessive allele pd. Hence, pd /pd ;
suppression is based on the physical binding of gene prod-
su /su is wild type in appearance and has red eyes. The
ucts in the cell—for example, protein–protein binding.
following analysis illustrates the inheritance pattern.
Assume that two proteins normally fit together to provide
A!homozygous purple-eyed fly is crossed with a homozy-
some type of cellular function. When a mutation causes a
gous red-eyed stock carrying the suppressor.
shape change in one protein, it no longer fits together with
pd /pd ; su+ /su+ (purple) ! pd + /pd + ; su /su (red) the other; hence, the function is lost (Figure 5-22). How-
# ever, a suppressor mutation that causes a compensatory
shape change in the second protein can restore fit and hence
F1 all pd + /pd ; su+ /su (red)
normal function. In this figure, if the genotypes were dip-
Self pd + /pd ; su+ /su (red) ! pd + /pd ; su+ /su (red) loids representing an F2 from a dihybrid, then a 14: 2 ratio
# would result because the only mutant genotypes would be

}
F2 + + m/m ( s+ /s+ (1 16) and m+ /m+ ( s/s (1 16), totaling (2 16). If this
9 pd / – ; su /– red
were a haploid dihybrid cross (such as m+ s+ ! m s), a 1:1
+
3 pd / – ; su /su red 13 ratio would result. From suppressor ratios generally, interact-
1 pd /pd ; su /su red ing proteins often can be deduced.
3 pd /pd ; su / – +
purple 3 Alternatively, in situations in which a mutation
causes a block in a metabolic pathway, the suppressor
13
The overall ratio in the F2 is 13 red:3 purple. The 16 com- finds some!way of bypassing the block—for example, by
ponent must include the double mutant, which is clearly wild rerouting into the blocked pathway intermediates similar
type in phenotype. This ratio is expected from a recessive to those beyond the block. In the following example, the
suppressor that itself has no detectable phenotype. suppressor provides an intermediate B to circumvent the
Suppression is sometimes confused with epistasis. How- block.
ever, the key difference is that a suppressor cancels the
expression of a mutant allele and restores the correspond- No suppressor
ing wild-type phenotype. Furthermore, often only two phe- A B product
notypes segregate (as in the preceding examples) rather
With suppressor
than three, as in epistasis.
How do suppressors work at the molecular level? There A B product
are many possible mechanisms. A particularly useful type of B

07_GriffitITGA12e_11478_Ch05_153_192.indd 172 03/10/19 12:02 PM

 5.3 Inferring Gene Interactions 173

In several organisms, nonsense suppressors have been A model for synthetic lethality
found—mutations in tRNA genes resulting in an anticodon
that will bind to a premature stop codon within a mutant
coding sequence. Hence, the suppressor allows translation A+ B+ Wild type
full binding;
to proceed past the former block and make a complete
fully functional
protein rather than a truncated one. Such suppressor muta-
tions often have little effect on the phenotype other than in DNA
suppression.

A– B+ Mutant A
KEY CONCEPT Mutant alleles called suppressors cancel the partial binding;
effect of a mutant allele of another gene, resulting in wild-type functional
phenotype.

Modifiers As the name suggests, a modifier mutation at a

Mutant B
second locus changes the degree of expression of a mutated A+ B– partial binding;
gene at the first locus. Regulatory genes provide a simple functional
illustration. As in an earlier example, regulatory proteins
bind to the sequence of the DNA upstream of the start site
for transcription. These proteins regulate the level of tran-
scription. In the discussion of complementation, we con- Double mutant
sidered a null mutation of a regulatory gene that almost A– B– binding impossible;
nonfunctional
completely prevented transcription. However, some reg-
ulatory mutations change the level of transcription of the
target gene so that either more or less protein is produced.
In other words, a mutation in a regulatory protein can
FIGURE 5-23 Two interacting A N IMATED
down-regulate or up-regulate the transcribed gene. Let’s
proteins perform some essential A RT
look at an example using a down-regulating regulatory function on some substrate such
mutation b, affecting a gene A in a fungus such as yeast. as DNA but must first bind to it. A model for synthetic
We look at the effect of b on a leaky mutation of gene A. Reduced binding of either protein lethality
A leaky mutation is one with some low level of gene allows some functions to remain,
but reduced binding of both is lethal.
function. We cross a leaky mutation a with the regulatory
mutation b:
leaky mutant a ! b+ ! inefficient regulator a+ ! b
have no backup, and the individual will lack essential func-
Progeny Phenotype tion and die. In another instance, a leaky mutation in one
wild type
step of a pathway may cause the pathway to slow down, but
a+ ! b+ leave enough function for life. However, if double mutants
a+ ! b defective (low transcription) combine, each with a leaky mutation in a different step, the
defective (defective protein A) whole pathway grinds to a halt. One version of the latter
a ! b+
interaction is two mutations in a protein machine, as shown
a!b extremely defective (low transcription of
in Figure 5-23.
defective protein)
In the earlier discussions of modified Mendelian
Hence, the action of the modifier is seen in the appear- ratios, all the crosses were dihybrid selfs. As an exercise,
ance of two grades of mutant phenotypes within the a you might want to calculate the ratios that would be pro-
progeny. duced in the same systems if testcrosses were made instead
of!selfs.
Synthetic lethals In some cases, when two viable single
mutants are intercrossed, the resulting double mutants are KEY CONCEPT Two mutations that are individually benign
lethal. In a diploid F2 , this result would be manifested as can become lethal when united in the same genotype. Such
a 9:3:3 ratio because the double mutant (which would be synthetic lethals can point to some type of normal gene inter-
action in the wild type.
the “1” component of the ratio) would be absent. These
synthetic lethals can be considered a special category of
gene interaction. They can point to specific types of inter- KEY CONCEPT Genetic analysis of gene interaction makes
actions of gene products. For instance, genome analysis use of mutant alleles, but the gene interaction revealed is one
has revealed that evolution has produced many duplicate that is taking place normally in the wild type.
systems within the cell. One advantage of these duplicates
might be to provide “backups.” If there are null mutations A summary of some of the ratios that reveal gene inter-
in genes in both duplicate systems, then a faulty system will action is shown in Table 5-2.

07_GriffitITGA12e_11478_Ch05_153_192.indd 173 03/10/19 12:02 PM

 174 C H A P T E R 5 ! ! Gene Interaction

TABLE 5-2 Some Modified F2 Ratios

9 : 3 : 3 : 1 No interaction
9 : 7 Genes in same pathway
9 : 3 : 4 Recessive epistasis
12 : 3 : 1 Dominant epistasis
13 : 3 Suppressor has no phenotype
14 : 2 Suppressor is like mutant
Note: Some of these ratios can be produced with other mechanisms of interaction.

SUMMARY

A gene does not act alone; rather, it acts in concert with The interaction of different genes can be detected by
many other genes in the genome. In forward genetic analy- testing double mutants because allele interaction implies
sis, deducing these complex interactions is an important stage interaction of gene products at the functional level. Some
of the research. Individual mutations are first tested for their key types of interaction are epistasis, suppression, and
dominance relations, a type of allelic interaction. Recessive synthetic lethality. Epistasis is the replacement of a mutant
mutations are often a result of haplosufficiency of the wild- phenotype produced by one mutation with a mutant pheno-
type allele, whereas dominant mutations are often the result type produced by mutation of another gene. The observation
either of haploinsufficiency of the wild type or of the mutant of epistasis suggests a common developmental or chemical
acting as a dominant negative (a rogue polypeptide). Some pathway. A suppressor is a mutation of one gene that can
mutations cause severe effects or even death (lethal muta- restore wild-type phenotype to a mutation at another gene.
tions). Lethality of a homozygous recessive mutation is a way Suppressors often reveal physically interacting proteins or
to assess if a gene is essential in the genome. nucleic acids. Some combinations of viable mutants are
The interaction of different genes is a result of their par- lethal, a result known as synthetic lethality. Synthetic lethals
ticipation in the same or connecting pathways of various can reveal a variety of interactions, depending on the nature
kinds—synthetic, signal transduction, or developmental. of the mutations.
Genetic dissection of gene interactions begins by the exper- The different types of gene interactions produce F2 dihy-
imenter amassing mutants affecting a character of interest. brid ratios that are modifications of the standard 9:3:3:1.
The complementation test determines whether two distinct For example, recessive epistasis results in a 9:3: 4 ratio.
recessive mutations are of one gene or of two different In more general terms, gene interaction and gene-
genes. The mutant genotypes are brought together in an F1 environment interaction are revealed by incomplete
individual, and if the phenotype is mutant, then no comple- penetrance (the ability of a genotype to express itself in
mentation has occurred and the two alleles must be of the the phenotype) and variable expressivity (the quantitative
same gene. If the phenotype is wild type, then complementa- degree of phenotypic manifestation of a genotype).
tion has occurred, and the alleles must be of different genes.

KEY TERMS

allelic series (multiple alleles) (p.!154) functional RNA (p. 163) permissive temperature (p. 160)
codominance (p. 156) heterokaryon (p. 166) pleiotropic allele (p. 160)
complementation (p. 165) incomplete dominance (p. 156) restrictive temperature (p. 160)
complementation test (p. 165) lethal allele (p. 158) revertant (p. 172)
dominant negative mutation (p.!156) modifier (p. 173) suppressor (p. 171)
double mutants (p. 164) multiple alleles (p. 154) synthetic lethal (p. 173)
epistasis (p. 169) null mutation (p. 156) temperature-sensitive (ts) mutations
essential gene (p. 158) one-gene–one-polypeptide hypothesis (p. 160)
expressivity (p. 161) (p. 163)
full (complete) dominance (p. 154) penetrance (p. 160)

07_GriffitITGA12e_11478_Ch05_153_192.indd 174 03/10/19 12:02 PM

 CHAPTER 5 REVIEW 175

SOLVED PROBLEMS

SOLVED PROBLEM 1 a. What irregularity does this pedigree show?

Most pedigrees show polydactyly (see Figure 2-25) inher- b. What genetic phenomenon does this pedigree illustrate?
ited as a rare autosomal dominant, but the pedigrees of c. Suggest a specific gene-interaction mechanism that
some families do not fully conform to the patterns expected could produce such a pedigree, showing genotypes of
for such inheritance. Such a pedigree is shown here. (The pertinent family members.
unshaded diamonds stand for the specified number of unaf-
fected persons of unknown sex.)

I
1 2

II
1 2 3 4 5 6 7 8 9 10 11

III 4
5 6 7 8 9 10 11 12 13 14 15 16 17

IV 4
5 6 7 8 9

SOLUTION polydactyly. So the only person from whom a suppressor

a. The normal expectation for an autosomal dominant is for could come is I-2. Furthermore, I-2 must be heterozygous
each affected individual to have an affected parent, but this for the suppressor allele because at least one of her chil-
expectation is not seen in this pedigree, which constitutes dren does express polydactyly. Therefore, the suppressor
the irregularity. What are some possible explanations? allele must be dominant. We have thus formulated the
Could some cases of polydactyly be caused by a hypothesis that the mating in generation I must have been
different gene, one that is an X-linked dominant gene? (I-1) P /p ! s /s ! (I-2) p /p ! S /s
This suggestion is not useful, because we still have to
explain the absence of the condition in persons II-6 where S is the suppressor and P is the allele responsible
and II-10. Furthermore, postulating recessive inheri- for polydactyly. From this hypothesis, we predict that
tance, whether autosomal or sex-linked, requires many the progeny will comprise the following four types if
people in the pedigree to be heterozygotes, which is the genes assort:
inappropriate because polydactyly is a rare condition.
Genotype Phenotype Example
b. Thus, we are left with the conclusion that polydac-
tyly must sometimes be incompletely penetrant. As P /p ! S /s normal (suppressed) II-6, II-10
described in this chapter, some individuals who have P /p ! s /s polydactylous II-1
the genotype for a particular phenotype do not express p /p ! S /s normal
it. In this pedigree, II-6 and II-10 seem to belong in this
p /p ! s /s normal
category; they must carry the polydactyly gene inher-
ited from I-1 because they transmit it to their progeny.
If S is rare, the progeny of II-6 and II-10 are:
c. As discussed in this chapter, environmental suppression
of gene expression can cause incomplete penetrance, as Progeny genotype Example
can suppression by another gene. To give the requested P /p ! S /s III-13
genetic explanation, we must come up with a genetic
P /p ! s /s III-8
hypothesis. What do we need to explain? The key is that
I-1 passes the mutation on to two types of progeny, rep- p /p ! S /s
resented by II-1, who expresses the mutant phenotype, p /p ! s /s
and by II-6 and II-10, who do not. (From the pedigree,
we cannot tell whether the other children of I-1 have We cannot rule out the possibilities that II-2 and II-4
the mutant allele.) Is genetic suppression at work? I-1 have the genotype P /p ! S /s and that by chance none of
does not have a suppressor allele because he expresses their descendants are affected.

07_GriffitITGA12e_11478_Ch05_153_192.indd 175 03/10/19 12:02 PM

 176 C H A P T E R 5 ! ! Gene Interaction

SOLVED PROBLEM 2 b. Now it is just a matter of deducing the specific geno-

Beetles of a certain species may have green, blue, or tur- types. Notice that the question states that the parents
quoise wing covers. Virgin beetles were selected from a were taken from a polymorphic population, which
polymorphic laboratory population and mated to determine means that they could be either homozygous or het-
the inheritance of wing-cover color. The crosses and results erozygous. A parent with blue wing covers, for exam-
were as given in the following table: ple, might be homozygous (wb /wb ) or heterozygous
(wb /w g or wb /w t ) . Here, a little trial and error and
Cross Parents Progeny common sense are called for, but, by this stage, the
1 blue ! green all blue question has essentially been answered, and all that
2 remains is to “cross the t’s and dot the i’s.” The follow-
blue ! blue 3
blue: 41 turquoise
4 ing genotypes explain the results. A dash indicates that
3 green ! green 3
green: 41 turquoise the genotype may be either homozygous or heterozy-
4
4 blue ! turquoise 1 blue: 12 turquoise gous in having a second allele farther down the allelic
2
series.
5 blue ! blue 3
blue: 41 green
4
Cross Parents Progeny
6 blue ! green 1 blue: 12 green
2 b b g
1 w /w ! w /– wb /w g or wb /–
7 blue ! green 1 blue: 41 green :
2
1 2 wb /w t ! wb /w t 3
4
! wb /– : 41 w t /w t
4
turquoise
3 w g /w t ! w g /w t 3
4
! w g /– : 41 w t /w t
8 turquoise ! turquoise all turquoise
4 wb /w t ! w t /w t 1 ! w b /w t : 1 w t /w t
2 2
a. Deduce the genetic basis of wing-cover color in this 5 wb /w g ! wb /w g 3 ! w b /– : 1 w g /w g
4 4
species.
6 wb /w g ! w g /w g 1 w b /w g : 1 w g /w g
2 2
b. Write the genotypes of all parents and progeny as com-
pletely as possible. 7 wb /w t ! w g /w t 1 w b /– : 1 w g /w t : 1
2 4 4
w t /w t

SOLUTION 8 w t /w t ! w t /w t t
all w /w t

a. These data seem complex at first, but the inheritance

pattern becomes clear if we consider the crosses one at SOLVED PROBLEM 3
a time. A general principle of solving such problems, as The leaves of pineapples can be classified into three types:
we have seen, is to begin by looking over all the crosses spiny (S), spiny tip (ST), and piping (nonspiny; P). In crosses
and by grouping the data to bring out the patterns. between pure strains followed by intercrosses of the F1, the
One clue that emerges from an overview of the data following results appeared:
is that all the ratios are one-gene ratios: there is no evi-
dence of two separate genes taking part at all. How Phenotypes
can such variation be explained with a single gene? Cross Parental F1 F2
The answer is that there is variation for the single gene 1 ST ! S ST 99 ST : 34 S
itself—that is, multiple allelism. Perhaps there are three
2 P ! ST P 120 P : 39 ST
alleles of one gene; let’s call the gene w (for wing-cover
color) and represent the alleles as w g, wb, and w t. Now 3 P!S P 95 P : 25 ST : 8 S
we have an additional problem, which is to determine
the dominance of these alleles. a. Assign gene symbols. Explain these results in regard to
Cross 1 tells us something about dominance the genotypes produced and their ratios.
because all of the progeny of a blue ! green cross are b. Using the model from part a, give the phenotypic ratios
blue; hence, blue appears to be dominant over green. that you would expect if you crossed (1) the F1 progeny
This conclusion is supported by cross 5, because the from piping ! spiny with the spiny parental stock and
green determinant must have been present in the paren- (2) the F1 progeny of piping ! spiny with the F1 progeny
tal stock to appear in the progeny. Cross 3 informs us of spiny ! spiny tip.
about the turquoise determinants, which must have
SOLUTION
been present, although unexpressed, in the parental
stock because there are turquoise wing covers in the a. First, let’s look at the F2 ratios. We have clear 3:1 ratios
progeny. So green must be dominant over turquoise. in crosses 1 and 2, indicating single-gene segrega-
Hence, we have formed a model in which the domi- tions. Cross 3, however, shows a ratio that is almost
nance is wb > w g > w t. Indeed, the inferred position certainly a 12:3:1 ratio. How do we know this ratio?
of the w t allele at the bottom of the dominance series Well, there are simply not that many complex ratios
is supported by the results of cross 7, where turquoise in genetics, and trial and error brings us to the 12:3:1
shows up in the progeny of a blue ! green cross. quite quickly. In the 128 progeny total, the numbers of

07_GriffitITGA12e_11478_Ch05_153_192.indd 176 03/10/19 12:02 PM

 CHAPTER 5 REVIEW 177

96: 24:8 are expected, but the actual numbers fit these Cross 2 can be partly written out without further
expectations remarkably well. thought by using our arbitrary gene symbols:
One of the principles of this chapter is that modi-
A/A ; –/– ! a/a ; B/B
fied Mendelian ratios reveal gene interactions. Cross 3 3
A/– ; –/–
gives F2 numbers appropriate for a modified dihybrid 4
A/a ; B/–
Mendelian ratio, and so it looks as if we are dealing 1
with a two-gene interaction. It seems the most promis- 4 a/a ; B/–
ing place to start; we can return to crosses 1 and 2 and We know that the F2 of cross 2 shows single-gene seg-
try to fit them in later. regation, and it seems certain now that the A /a allelic
Any dihybrid ratio is based on the phenotypic pro- pair has a role. But the B allele is needed to produce
portions 9:3:3:1. Our observed modification groups the spiny-tip phenotype, and so all plants must be
them as follows: homozygous B /B:
9 A /– ; B /–
3 A/– ; b /b
} 12 piping A/A ; B/B ! a/a ; B/B
3
4 A/– ; B/B
3 a /a ; B /– 3 spiny tip A/a ; B/B
1
1 a /a ; b /b 1 spiny 4 a/a ; B/B

So, without worrying about the name of the type of Notice that the two single-gene segregations in
gene interaction (we are not asked to supply this any- crosses 1 and 2 do not show that the genes are not
way), we can already define our three pineapple-leaf interacting. What is shown is that the two-gene inter-
phenotypes in relation to the proposed allelic pairs A /a action is not revealed by these crosses—only by cross
and B /b: 3, in which the F1 is heterozygous for both genes.
piping = A /– (B /b irrelevant) b. Now it is simply a matter of using Mendel’s laws to
predict cross outcomes:
spiny tip = a /a ; B /–
spiny = a /a ; b /b
}
(1) A /a ; B /b × a /a ; b /b ⎯⎯⎯
→ 1 A /a ; B /b
4
piping
What about the parents of cross 3? The spiny parent (independent 1
4
A /a ; b /b
must be a /a ; b /b, and, because the B gene is needed assortment in
to produce F2 spiny-tip leaves, the piping parent must
1 a /a ; B /b spiny tip
a standard 4
be A /A ; B /B. (Note that we are told that all parents testcross) 1 a /a ; b /b spiny
4
are pure, or homozygous.) The F1 must therefore be
A /a ; B /b. (2) A /a ; B /b × a /a ; B /b ⎯⎯⎯
→
Without further thought, we can write out cross 1

}
3 3
as follows: 4
B /− ⎯ ⎯→ 8
1 A /a 1
2 2
piping
1 b /b ⎯ ⎯→ 1
a/a ; B/B ! a/a ; b/b 4 8
3
a/a ; B/– 3 3
4
B /− ⎯ ⎯→ spiny tip
a/a ; B/b 1 a /a
4 8
1 2 1 1
4 a/a ; b/b 4
b /b ⎯ ⎯→ 8
spiny

PROBLEMS

Visit SaplingPlus for supplemental content. Problems with the icon are available for review/grading. Problems with the
icon have a Problem Solving Video.

WORKING WITH THE FIGURES c. In the system defined in this figure, if we assume
(The first 19 questions require inspection of text figures.) the gene codes for an enzyme catalyzing the
synthesis of a black pigment, what would be the
1. a. In Figure 5-1, what do the yellow stars represent?
phenotype of the heterozygote?
b. Explain in your own words why the heterozygote
is functionally wild type.

07_GriffitITGA12e_11478_Ch05_153_192.indd 177 03/10/19 1:58 PM

 178 C H A P T E R 5 ! ! Gene Interaction

d. At the structural level, what might be the difference e. Do you think it would be possible to treat albinism
between the proteins represented by the colors by ingesting melanin? (Research this possibility
orange and yellow? yourself.)
2. a. In Figure 5-2, explain how the mutant polypeptide 12. a. In Figure 5-15, what do the dollar, pound, and yen
acts as a spoiler and what its net effect on pheno- symbols represent?
type is. b. Why can’t the left-hand F1 heterozygote synthesize
b. What might cause a bend in the mutant protein? blue pigment?
c. In Model 1, what can you say about the pos- c. Draw out the results of crossing the $ and ¥ lines.
sibility of up-regulation of the protein in the d. Write out all the genotypes and phenotypes of
heterozygote? progeny from a self of the blue F1.
3. In Figure 5-4, does the photo show the blood of one e. Write out all the genotypes and phenotypes pro-
individual (if so, which one), or a mixture of bloods (if duced by crossing the blue F1 to the white F1.
so, which ones)?
13. a. In Figure 5-16, explain at the protein level why this
4. a. In Figure 5-5, what is the object represented by the heterokaryon can grow on minimal medium.
color blue?
b. A heterokaryon produces spores by pinching off
b. Is it true to say that the sickle-cell hemoglobin cells that contain a few nuclei. Will any of these
migrates faster than normal hemoglobin? spores be arginine-dependent? Explain.
c. What might cause the different migration rates? 14. a. In Figure 5-17, write possible genotypes for each of
5. a. In Figure 5-6, assess the allele V f with respect to the four snakes illustrated.
the V by allele. Is it dominant? Recessive? Codomi- b. Explain in short sentences the meaning of the
nant? Incompletely dominant? header for this figure.
b. In this figure, is there any case in which the hetero- 15. a. In Figure 5-18, which panel represents the double
zygote has a truly new phenotype? mutant?
c. Predict the phenotype of the heterozygotes of v b. State the function of the regulatory gene.
combined with the other alleles.
c. In the situation in panel b, would protein from the
6. In Figure 5-7, if you assume that all the progeny are active protein gene be made?
visible, is the observed color ratio the one expected?
d. What is the function of the pale green region?
7. In Figure 5-9, propose a specific genetic explanation
e. What is the element represented in yellow?
for individual Q (give a possible genotype, defining the
alleles). f. Panels b and d have the same outcome: state the
two different mechanisms that produce this.
8. In Figure 5-10, point to the individuals that show full
expressivity. 16. a. In Figure 5-19, if you selfed 10 different F2 pink
plants, would you expect to find any white-
9. Speculate logically on the minimum number of mod-
flowered plants among the offspring? Any blue-
ifying genes and alleles that could produce the varia-
flowered plants?
tion shown by the allele S P in Figure 5-11.
b. Some white F2 plants have a functional enzyme 2:
10. From a knowledge of the structures shown in Figure
Since enzyme 2 produces the blue pigment, why
5-13, do you think Beadle and Tatum might have had
are these plants not blue?
a clue about the sequential steps in the synthetic path-
way before doing their genetic tests? 17. In Figure 5-21, write down possible genotypes for each
of the three petals.
11. a. In Figure 5-14, in view of the position of HPA oxi-
dase earlier in the pathway compared to that of 18. a. In Figure 5-22, what do the square/triangular pegs
HA oxidase, would you expect people with tyrosi- and holes represent?
nosis to show symptoms of alkaptonuria? b. Is the suppressor mutation alone wild type in
b. If a double mutant could be found, would you phenotype?
expect tyrosinosis to be epistatic to alkaptonuria? c. Would it be reasonable to call the s gene a suppres-
c. Do you think it might be possible to cure the symp- sor gene in a wild-type cell?
toms of PKU by ingesting tyrosine? (Research this 19. a. In Figure 5-23, explain why the interacting alleles
possibility yourself.) are called synthetic lethals.
d. How might you treat cretinism? (Research this b. For the model to work, is it essential that the red
possibility yourself.) and blue proteins bind to each other?

07_GriffitITGA12e_11478_Ch05_153_192.indd 178 03/10/19 12:02 PM

 CHAPTER 5 REVIEW 179

BASIC PROBLEMS a. What is the order of compounds A to E in the

pathway?
20. In humans, the disease galactosemia causes intellectual
disabilities at an early age. Lactose (milk sugar) is b. At which point in the pathway is each mutant
broken down to galactose plus glucose. Normally, blocked?
galactose is broken down further by the enzyme galac- c. Would a heterokaryon composed of double
tose-1-phosphate uridyltransferase (GALT). However, mutants 1,3 and 2,4 grow on a minimal medium?
in patients with galactosemia, GALT is inactive, lead- Would 1,3 and 3,4? Would 1,2 and 2,4 and 1,4?
ing to a buildup of high levels of galactose, which, in
24. In a certain plant, the flower petals are normally pur-
the brain, causes intellectual disabilities. How would
ple. Two recessive mutations arise in separate plants
you provide a secondary cure for galactosemia? Would
and are found to be on different chromosomes. Muta-
you expect this disease phenotype to be dominant or
tion 1 (m1) gives blue petals when homozygous (m1 /m1).
recessive?
Mutation 2 (m2 ) gives red petals when homozygous
21. In humans, PKU (phenylketonuria) is a recessive dis- (m2 /m2 ) . Biochemists working on the synthesis of
ease caused by an enzyme inefficiency at step A in flower pigments in this species have already described
the following simplified reaction sequence, and AKU the following pathway:
(alkaptonuria) is another recessive disease due to an
enzyme inefficiency in one of the steps summarized as enzym
eA blue pigment
step B here: colorless (white)
A
phenylalanine ! ! ! B
� tyrosine ! ! � CO2 + H 2O compound enzym
eB red pigment
A person with PKU marries a person with AKU.
What phenotypes do you expect for their children? All a. Which mutant would you expect to be deficient in
normal, all having PKU only, all having AKU only, all enzyme A activity?
having both PKU and AKU, or some having AKU and b. A plant has the genotype M1 /m1; M2 /m2 . What
some having PKU? would you expect its phenotype to be?
22. In Drosophila, the autosomal recessive bw causes a c. If the plant in part b is selfed, what colors of prog-
dark brown eye, and the unlinked autosomal recessive eny would you expect, and in what proportions?
st causes a bright scarlet eye. A homozygote for both
d. Why are these mutants recessive?
genes has a white eye. Thus, we have the following
correspondences between genotypes and phenotypes: 25. In sweet peas, the synthesis of purple anthocyanin pig-
ment in the petals is controlled by two genes, B and D.
st + /st + ; bw+ /bw+ = red eye (wild type) The pathway is
st + /st + ; bw /bw !!= brown eye gene B gene D
white enzyme blue enzyme anthocyanin
st /st ; bw+ /bw+ = scarlet eye ! !!! � ! !!! �
intermediate intermediate (purple)
st /st ; bw /bw !!= white eye
a. What color petals would you expect in a pure-
Construct a hypothetical biosynthetic pathway
breeding plant unable to catalyze the first reaction?
showing how the gene products interact and why the dif-
ferent mutant combinations have different phenotypes. b. What color petals would you expect in a pure-
breeding plant unable to catalyze the second reaction?
23. Several mutants are isolated, all of which require com-
pound G for growth. The compounds (A to E) in the c. If the plants in parts a and b are crossed, what
biosynthetic pathway to G are known, but their order color petals will the F1 plants have?
in the pathway is not known. Each compound is tested d. What ratio of purple : blue : white plants would you
for its ability to support the growth of each mutant expect in the F2?
(1 to 5). In the following table, a plus sign indicates
26. If a man of blood-group AB marries a woman of
growth and a minus sign indicates no growth.
blood-group A whose father was of blood-group O, to
Compound tested what different blood groups can this man and woman
expect their children to belong?
A B C D E G
27. Most of the feathers of erminette fowl are light col-
Mutant 1 # # # + # +
ored, with an occasional black one, giving a flecked
2 # + # + # + appearance. A cross of two erminettes produced a
3 # # # # # + total of 48 progeny, consisting of 22 erminettes, 14
4 # + + + # + blacks, and 12 pure whites. What genetic basis of the
erminette pattern is suggested? How would you test
5 + + + + # +
your hypotheses?

07_GriffitITGA12e_11478_Ch05_153_192.indd 179 03/10/19 12:02 PM

 180 C H A P T E R 5 ! ! Gene Interaction

28. Radishes may be long, round, or oval, and they may be 32. Consider two blood polymorphisms that humans
red, white, or purple. You cross a long, white variety have in addition to the ABO system. Two alleles LM
with a round, red one and obtain an oval, purple F1. and LN determine the M, N, and MN blood groups.
The F2 shows nine phenotypic classes as follows: 9 The dominant allele R of a different gene causes a
long, red; 15 long, purple; 19 oval, red; 32 oval, pur- person to have the Rh+ (rhesus positive) phenotype,
ple; 8 long, white; 16 round, purple; 8 round, white; whereas the homozygote for r is Rh# (rhesus nega-
16 oval, white; and 9 round, red. tive). Two men took a paternity dispute to court,
a. Provide a genetic explanation of these results. Be each claiming three children to be his own. The blood
sure to define the genotypes and show the constitu- groups of the men, the children, and their mother
tion of the parents, the F1, and the F2. were as follows:

b. Predict the genotypic and phenotypic proportions Person Blood group

in the progeny of a cross between a long, purple husband O M Rh+
radish and an oval, purple one.
wife’s lover AB MN Rh#
29. In the multiple-allele series that determines coat color
in rabbits, c+ encodes agouti, c ch encodes chinchilla (a wife A N Rh+
beige coat color), and c h encodes Himalayan. Dom- child 1 O MN Rh+
inance is in the order c+ > c ch > c h . In a cross of
c+ /c ch � c ch /c h , what proportion of progeny will be child 2 A N Rh+
chinchilla? child 3 A MN Rh#
30. Black, sepia, cream, and albino are coat colors of
guinea pigs. Individual animals (not necessarily from From this evidence, can the paternity of the children be
pure lines) showing these colors were intercrossed; the established?
results are tabulated as follows, where the abbrevia- 33. On a fox ranch in Wisconsin, a mutation arose that
tions A (albino), B (black), C (cream), and S (sepia) gave a “platinum” coat color. The platinum color
represent the phenotypes: proved very popular with buyers of fox coats, but the
breeders could not develop a pure-breeding platinum
Phenotypes of progeny
strain. Every time two platinums were crossed, some
Parental normal foxes appeared in the progeny. For example,
Cross phenotypes B S C A the repeated matings of the same pair of platinums
1 B�B 22 0 0 7 produced 82 platinum and 38 normal progeny. All
2 B�A 10 9 0 0 other such matings gave similar progeny ratios. State
a concise genetic hypothesis that accounts for these
3 C �C 0 0 34 11
results.
4 S �C 0 24 11 12
34. For several years, Hans Nachtsheim investigated an
5 B�A 13 0 12 0 inherited anomaly of the white blood cells of rabbits.
6 B �C 19 20 0 0 This anomaly, termed the Pelger anomaly, is the arrest
7 B�S 18 20 0 0 of the segmentation of the nuclei of certain white cells.
This anomaly does not appear to seriously burden the
8 B�S 14 8 6 0
rabbits.
9 S�S 0 26 9 0
a. When rabbits showing the Pelger anomaly were
10 C�A 0 0 15 17 mated with rabbits from a true-breeding normal
stock, Nachtsheim counted 217 offspring showing
a. Deduce the inheritance of these coat colors, and the Pelger anomaly and 237 normal progeny. What
use gene symbols of your own choosing. Show all is the genetic basis of the Pelger anomaly?
parent and progeny genotypes.
b. When rabbits with the Pelger anomaly were mated
b. If the black animals in crosses 7 and 8 are crossed, with each other, Nachtsheim found 223 normal
what progeny proportions can you predict by using progeny, 439 with the Pelger anomaly, and 39
your model? extremely abnormal progeny. These very abnor-
31. In a maternity ward, four babies become acciden- mal progeny not only had defective white blood
tally mixed up. The ABO types of the four babies are cells, but also showed severe deformities of the
known to be O, A, B, and AB. The ABO types of the skeletal system; almost all of them died soon after
four sets of parents are determined. Indicate which birth. In genetic terms, what do you suppose these
baby belongs to each set of parents: (a) AB � O , extremely defective rabbits represented? Why were
(b) A � O, (c) A � AB , (d) O � O. there only 39 of them?

07_GriffitITGA12e_11478_Ch05_153_192.indd 180 03/10/19 12:03 PM

 CHAPTER 5 REVIEW 181

c. What additional experimental evidence might you a. Were the mutations in the two auxotrophs in the
collect to test your hypothesis in part b? same gene in the pathway for synthesizing leu-
d. In Berlin, about 1 human in 1000 shows a Pelger cine, or in two different genes in that pathway?
anomaly of white blood cells very similar to that Explain.
described for rabbits. The anomaly is inherited as a b. Write the genotype of the two strains according to
simple dominant, but the homozygous type has not your model.
been observed in humans. Based on the condition c. What progeny, and in what proportions, would
in rabbits, why do you suppose the human homo- you predict from crossing the two auxotrophic
zygous has not been observed? mutants? (Assume independent assortment.)
e. Again by analogy with rabbits, what phenotypes 39. A yeast geneticist irradiates haploid cells of a strain
and genotypes would you expect among the that is an adenine-requiring auxotrophic mutant,
children of a man and woman who both show the caused by mutation of the gene ade1. Millions of the
Pelger anomaly? irradiated cells are plated on minimal medium, and a
(Data from A. M. Srb, R. D. Owen, and R. S. Edgar, small number of cells divide and produce prototro-
General Genetics, 2nd ed. W. H. Freeman and Com- phic colonies. These colonies are crossed individu-
pany, 1965.) ally with a wild-type strain. Two types of results are
35. Two normal-looking fruit flies were crossed, and, in obtained:
the progeny, there were 202 females and 98 males. 1. prototroph � wild type: progeny all prototrophic
a. What is unusual about this result? 2. prototroph � wild type: progeny 75% prototrophic,
b. Provide a genetic explanation for this anomaly. 25% adenine-requiring auxotrophs
a. Explain the difference between these two types
c. Provide a test of your hypothesis.
of results.
36. You have been given a virgin Drosophila female. You
b. Write the genotypes of the prototrophs in each
notice that the bristles on her thorax are much shorter
case.
than normal. You mate her with a normal male (with
long bristles) and obtain the following F1 progeny: c. What progeny phenotypes and ratios do you
1 short-bristled females, 1 long-bristled females, and predict from crossing a prototroph of type 2 by
3 3
1 long-bristled males. A cross of the F long-bristled the original ade1 auxotroph?
3 1
females with their brothers gives only long-bristled F2 . 40. In roses, the synthesis of red pigment is by two steps in
A cross of short-bristled females with their brothers a pathway, as follows:
gives 13 short-bristled females, 13 long-bristled females, gene P
and 13 long-bristled males. Provide a genetic hypothesis colorless intermediate ! !!! �
to account for all these results, showing genotypes in gene Q
magenta intermediate ! !!! � red pigment
every cross.
a. What would the phenotype be of a plant homozy-
37. A dominant allele H reduces the number of body bris- gous for a null mutation of gene P?
tles that Drosophila flies have, giving rise to a “hair-
less” phenotype. In the homozygous condition, H is b. What would the phenotype be of a plant homozy-
lethal. An independently assorting dominant allele S gous for a null mutation of gene Q?
has no effect on bristle number except in the presence c. What would the phenotype be of a plant homozy-
of H, in which case a single dose of S suppresses the gous for null mutations of genes P and Q?
hairless phenotype, thus restoring the hairy pheno- d. Write the genotypes of the three strains in parts a,
type. However, S also is lethal in the homozygous (S /S) b, and c.
condition.
e. What F2 ratio is expected from crossing plants from
a. What ratio of hairy to hairless flies would you find parts a and b? (Assume independent assortment.)
in the live progeny of a cross between two!hairy
flies both carrying H in the suppressed condition? 41. Because snapdragons (Antirrhinum) possess the pig-
ment anthocyanin, they have reddish purple petals.
b. When the hairless progeny are backcrossed with Two pure anthocyaninless lines of Antirrhinum were
a parental hairy fly, what phenotypic ratio would developed, one in California and one in Holland.
you expect to find among their live progeny? They looked identical in having no red pigment
38. After irradiating wild-type cells of Neurospora (a hap- at all, manifested as white (albino) flowers. How-
loid fungus), a geneticist finds two leucine-requiring ever,!when petals from the two lines were ground
auxotrophic mutants. He combines the two mutants in up!together in buffer in the same test tube, the solu-
a heterokaryon and discovers that the heterokaryon is tion, which appeared colorless at first, gradually
prototrophic. turned red.

07_GriffitITGA12e_11478_Ch05_153_192.indd 181 03/10/19 12:03 PM

 182 C H A P T E R 5 ! ! Gene Interaction

a. What control experiments should an investigator 3. What is a variant?

conduct before proceeding with further analysis? 4. What are the variants in this problem?
b. What could account for the production of the red 5. What does “in nature” mean?
color in the test tube?
6. In what way would the variants have been found in
c. According to your explanation for part b, what nature? (Describe the scene.)
would be the genotypes of the two lines? 7. At which stages in the experiments would seeds be
d. If the two white lines were crossed, what used?
would!you!predict the phenotypes of the F1 and F2 8. Would the way of writing a cross “blue � white,”
to be? for example, mean the same as “white � blue” ?
42. The frizzle fowl is much admired by poultry fanciers. Would you expect similar results? Why or why not?
It gets its name from the unusual way that its feathers 9. In what way do the first two rows in the table differ
curl up, giving the impression that it has been (in the from the third row?
memorable words of animal geneticist F. B. Hutt)
10. Which phenotypes are dominant?
“pulled backwards through a knothole.” Unfortu-
nately, frizzle fowl do not breed true: when two frizzles 11. What is complementation?
are intercrossed, they always produce 50 percent friz- 12. Where does the blueness come from in the progeny
zles, 25 percent normal, and 25 percent with peculiar of the pink � white cross?
woolly feathers that soon fall out, leaving the birds 13. What genetic phenomenon does the production of a
naked. blue F1 from pink and white parents represent?
a. Give a genetic explanation for these results, 14. List any ratios that you can see.
showing genotypes of all phenotypes, and pro-
15. Are there any monohybrid ratios?
vide a statement of how your explanation works.
16. Are there any dihybrid ratios?
b. If you wanted to mass-produce frizzle fowl for
sale, which types would be best to use as a breed- 17. What does observing monohybrid and dihybrid
ing pair? ratios tell you?
18. List four modified Mendelian ratios that you can
43. The petals of the plant Collinsia parviflora are
think of.
normally blue, giving the species its common name,
blue-eyed Mary. Two pure-breeding lines were 19. Are there any modified Mendelian ratios in the
obtained from color variants found in nature; the first problem?
line had pink petals, and the second line had white 20. What do modified Mendelian ratios indicate
petals. The following crosses were made between pure generally?
lines, with the results shown: 21. What is indicated by the specific modified ratio or
Parents F1 F2 ratios in this problem?

blue � white blue 101 blue, 33 white 22. Draw chromosomes representing the meioses in the
parents in the cross blue � white and representing
blue � pink blue 192 blue, 63 pink meiosis in the F1.
pink � white blue 272 blue, 121 white, 89 pink 23. Repeat step 22 for the cross blue � pink.

a. Explain these results genetically. Define the allele Now try to solve the problem. If you are unable to do so,
symbols that you use, and show the genetic con- try to identify the obstacle and write a sentence or two
stitution of the parents, the F1, and the F2 in each describing your difficulty. Then go back to the expansion
cross. questions and see if any of them relate to your difficulty. If
this approach does not work, inspect the Learning Objec-
b. A cross between a certain blue F2 plant and a certain
tives and Key Concepts of this chapter and ask yourself
white F2 plant gave progeny of which 83 were!blue, 81
which might be relevant to your difficulty.
were pink, and 12 were white. What must the geno-
types of these two F2 plants have been? 44. A woman who owned a purebred albino poodle (an
autosomal recessive phenotype) wanted white pup-
www
UNPACKING PROBLEM 43 pies. She took the dog to a breeder, who said he would
www
mate the female with an albino stud male, also from a
Before attempting a solution to this problem, try answer- pure stock. When six puppies were born, all of them
ing the following questions: were black; so the woman sued the breeder, claiming
1. What is the character being studied? that he replaced the stud male with a black dog, giv-
ing her six unwanted puppies. You are called in as an
2. What is the wild-type phenotype?
expert witness, and the defense asks you if it is possible

07_GriffitITGA12e_11478_Ch05_153_192.indd 182 03/10/19 12:03 PM

 CHAPTER 5 REVIEW 183

to produce black offspring from two pure-breeding Cross Parents Progeny

recessive albino parents. What testimony do you give?
1 blue � scarlet 1 !blue: 1 purple: 1 scarlet
4 2 4
45. A snapdragon plant that bred true for white petals
purple � purple 1 blue: 1 purple: 1 scarlet
was crossed with a plant that bred true for purple pet- 2 4 2 4
3 1 purple
als, and all the F1 had white petals. The F1 was selfed. 3 blue � blue 4
blue: 4
Among the F2 , three phenotypes were observed in the 4 blue � purple 3
blue: 4 purple: 1 scarlet
8 8 8
following numbers: 1 purple: 1 scarlet
5 purple � scarlet 2 2
white 240
solid purple 61 49. Corn breeders obtained pure lines whose kernels turn
spotted purple 19 sun red, pink, scarlet, or orange when exposed to sun-
light (normal kernels remain yellow in sunlight). Some
Total 320 crosses between these lines produced the following
a. Propose an explanation for these results, showing results. The phenotypes are abbreviated O, orange; P,
genotypes of all generations (make up and explain pink; Sc, scarlet; and SR, sun red.
your symbols).
Phenotypes
b. A white F2 plant was crossed with a solid purple F2
plant, and the progeny were Cross Parents F1 F2
white 50% 1 SR � P all SR 66 SR : 20 P
solid purple 25% 2 O � SR all SR 998 SR :314 O
spotted purple 25% 3 O� P all O 1300 O: 429 P
4 O � Sc all Y 182 Y:80 O: 58 Sc
What were the genotypes of the F2 plants crossed?
46. Most flour beetles are black, but several color variants Analyze the results of each cross, and provide a unify-
are known. Crosses of pure-breeding parents produced ing hypothesis to account for all the results. (Explain
the following results (see table) in the F1 generation, all symbols that you use.)
and intercrossing the F1 from each cross gave the
50. Many kinds of wild animals have the agouti coloring
ratios shown for the F2 generation. The phenotypes
pattern, in which each hair has a yellow band around!it.
are abbreviated Bl, black; Br, brown; Y, yellow; and
W,!white. a. Black mice and other black animals do not have
the yellow band; each of their hairs is all black.
Cross Parents F1 F2 This absence of wild agouti pattern is called nona-
1 Br � Y Br 3 Br :1 Y gouti. When mice of a true-breeding agouti line
2 are crossed with nonagoutis, the F1 is all agouti
Bl � Br Bl 3 Bl :1 Br
and the F2 has a 3:1 ratio of agoutis to nonagou-
3 Bl � Y Bl 3 Bl :1 Y tis. Diagram this cross, letting A represent the allele
4 W �Y Bl 9 Bl :3 Y: 4 W responsible for the agouti phenotype and a, nona-
5 W � Br gouti. Show the phenotypes and genotypes of the
Bl 9 Bl :3 Br : 4 W
parents, their gametes, the F1, their gametes, and
6 Bl � W Bl 9 Bl :3 Y : 4 W the F2.

a. From these results, deduce and explain the inheri- b. Another inherited color deviation in mice substi-
tance of these colors. tutes brown for the black color in the wild-type
hair. Such brown-agouti mice are called cinna-
b. Write the genotypes of each of the parents, the F1, mons. When wild-type mice are crossed with cin-
and the F2 in all crosses. namons, all of the F1 are wild type and the F2 has
47. Two albinos marry and have four normal children. a 3:1 ratio of wild type to cinnamon. Diagram this
How is this possible? cross as in part a, letting B stand for the wild-type
48. Consider the production of flower color in the Japa- black allele and b stand for the cinnamon brown
nese morning glory (Pharbitis nil). Dominant alleles allele.
of either of two separate genes (A /# ! b /b or a /a ! B /#) c. When mice of a true-breeding cinnamon line are
produce purple petals. A /# ! B /# produces blue petals, crossed with mice of a true-breeding nonagouti
and a /a ! b /b produces scarlet petals. Deduce the geno- (black) line, all of the F1 are wild type. Use a genetic
types of parents and progeny in the following crosses: diagram to explain this result.

07_GriffitITGA12e_11478_Ch05_153_192.indd 183 03/10/19 12:03 PM

 184 C H A P T E R 5 ! ! Gene Interaction

d. In the F2 of the cross in part c, a fourth color called

chocolate appears in addition to the parental cin-
namon and nonagouti and the wild type of the F1.
Chocolate mice have a solid, rich brown color.
What is the genetic constitution of the choco-
lates? Single Walnut Pea Rose
e. Assuming that the A /a and B /b allelic pairs assort
a. What comb types will appear in the F1 and in the
independently of each other, what do you expect to
F2 and in what proportions if single-combed birds
be the relative frequencies of the four color types
are crossed with birds of a true-breeding walnut
in the F2 described in part d? Diagram the cross of
strain?
parts c and d, showing phenotypes and genotypes
(including gametes). b. What are the genotypes of the parents in a
walnut � rose mating from which the progeny are
f. What phenotypes would be observed in what 3
8
rose, 83 walnut, 81 pea, and 81 single?
proportions in the progeny of a backcross of F1
mice from part c with the cinnamon parental c. What are the genotypes of the parents in a
stock? With the nonagouti (black) parental stock? walnut � rose mating from which all the progeny
Diagram these backcrosses. are walnut?
g. Diagram a testcross for the F1 of part c. What col- d. How many genotypes produce a walnut pheno-
ors would result, and in what proportions? type? Write them out.
h. Albino (pink-eyed white) mice are homozygous 53. The production of eye-color pigment in Drosoph-
for the recessive member of an allelic pair C /c , ila requires the dominant allele A. The dominant
which assorts independently of the A /a and B /b allele! P! of a second independent gene turns the
pairs. Suppose that you have four different highly pigment to purple, but its recessive allele leaves
inbred (and therefore presumably homozygous) it!red. A fly producing no pigment has white eyes.
albino lines. You cross each of these lines with a Two pure lines were crossed with the following
true-breeding wild-type line, and you raise a large results:
F2 progeny from each cross. What genotypes for P red-eyed female � white-eyed male
the albino lines can you deduce from the following
F2 phenotypes? %
F1 purple-eyed females
Phenotypes of progeny
red-eyed males
F2 of Wild Cinna- Choco- F1 � F1
line type Black mon late Albino
%
1 87 0 32 0 39 3
F2 both males and females: 8
purple eyed
2 62 0 0 0 18 3
8
red eyed
3 96 30 0 0 41 2
8
white eyed
4 287 86 92 29 164
Explain this mode of inheritance, and show the geno-
(Adapted from A. M. Srb, R. D. Owen, and R. S. Edgar, types of the parents, the F1, and the F2.
General Genetics, 2nd ed. W. H. Freeman and Com- 54. When true-breeding brown dogs are mated with cer-
pany, 1965.) tain true-breeding white dogs, all the F1 pups are
51. An allele A that is not lethal when homozygous causes white. The F2 progeny from some F1 � F1 crosses were
rats to have yellow coats. The allele R of a separate 118 white, 32 black, and 10 brown pups. What is the
gene that assorts independently produces a black coat. genetic basis for these results?
Together, A and R produce a grayish coat, whereas 55. Wild-type strains of the haploid fungus Neurospora
a and r produce a white coat. A gray male is crossed can make their own tryptophan. An abnormal allele
with a yellow female, and the F1 is 83 yellow, 83 gray, td renders the fungus incapable of making its own
1 black, and 1 white. Determine the genotypes of the
8 8 tryptophan. An individual of genotype td grows only
parents. when its medium supplies tryptophan. The allele su
52. The genotype r /r ; p /p gives fowl a single comb, assorts independently of td; its only known effect is
R / # ; P / # gives a walnut comb, r /r ; P / # gives a pea to suppress the td phenotype. Therefore, strains car-
comb, and R / # ; p /p gives a rose comb (see the illus- rying both td and su do not require tryptophan for
trations). Assume independent assortment. growth.

07_GriffitITGA12e_11478_Ch05_153_192.indd 184 03/10/19 12:03 PM

 CHAPTER 5 REVIEW 185

a. If a td ; su strain is crossed with a genotypically Cross Parents Progeny

wild-type strain, what genotypes are expected in
the progeny, and in what proportions? 1 dark purple � white with 1
dark purple:
2
yellow spots 1
light purple
b. What will be the ratio of tryptophan-dependent 2
to tryptophan-independent progeny in the cross 2 white with � light purple 1
white with purple
2
of part a? yellow spots spots: 41 dark
56. Mice of the genotypes A /A ; B /B ; C /C ; D /D ; S /S and purple: 41 light purple
a /a ; b /b ; c /c ; d /d ; s /s are crossed. The progeny are in-
In each case, give the genotypes of parents and prog-
tercrossed. What phenotypes will be produced in the F2,
eny with respect to the three genes.
and in what proportions? [The allele symbols stand for
the following: A = agouti, a = solid (nonagouti) ; B = 60. In one species of Drosophila, the wings are normally
black pigment, b = brown ; C = pigmented, c = albino; round in shape, but you have obtained two pure lines,
D = nondilution, d = dilution (milky color) ; S = unspot- one of which has oval wings and the other sickle-
ted, s = pigmented spots on white background.] shaped wings. Crosses between pure lines reveal the
following results:
57. Consider the genotypes of two lines of chickens: the pure-
line mottled Honduran is i /i ; D /D ; M /M ;W /W, and the Parents F1
pure-line leghorn is I /I ; d /d ; m /m ; w /w, where Female Male Female Male
I != white feathers, i = colored feathers sickle round sickle sickle
D = duplex comb, d = simplex comb round sickle sickle round
M = bearded, m = beardless sickle oval oval sickle
W != white skin, w = yellow skin
a. Provide a genetic explanation of these results,
These four genes assort independently. Starting with defining all allele symbols.
these two pure lines, what is the fastest and most con-
b. If the F1 oval females from cross 3 are crossed with
venient way of generating a pure line that has colored
the F1 round males from cross 2, what phenotypic
feathers, has a simplex comb, is beardless, and has
proportions are expected for each sex in the progeny?
yellow skin? Make sure that you show
61. Mice normally have one yellow band on each hair, but
a. the breeding pedigree.
variants with two or three bands are known. A female
b. the genotype of each animal represented. mouse having one band was crossed with a male hav-
c. how many eggs to hatch in each cross, and why ing three bands. (Neither animal was from a pure line.)
this number. The progeny were

d. why your scheme is the fastest and the most conve- Females 1
2
one band Males 1
2
one band
nient. 1 1
2
three bands 2
two bands
58. The following pedigree is for a dominant phenotype
governed by an autosomal allele. What does this ped- a. Provide a clear explanation of the inheritance of
igree suggest about the phenotype, and what can you these phenotypes.
deduce about the genotype of individual A? b. In accord with your model, what would be the
outcome of a cross between a three-banded daugh-
ter and a one-banded son?
62. In minks, wild types have an almost black coat. Breeders
A have developed many pure lines of color variants for
the mink-coat industry. Two such pure lines are plat-
inum (blue gray) and aleutian (steel gray). These lines
59. Petal coloration in foxgloves is determined by were used in crosses, with the following results:
three genes. M encodes an enzyme that synthesizes
anthocyanin, the purple pigment seen in these petals; Cross Parents F1 F2
m /m produces no pigment, resulting in the pheno- 1 wild � platinum wild 18 wild, 5 platinum
type albino with yellowish spots. D is an enhancer
2 wild � aleutian wild 27 wild, 10 aleutian
of anthocyanin, resulting in a darker pigment; d /d
does not enhance. At the third locus, w /w allows pig- 3 platinum � aleutian wild 133 wild
ment deposition in petals, but W prevents pigment 41 platinum
deposition except in the spots and so results in the 46 aleutian
white, spotted phenotype. Consider the following
two crosses: 17 sapphire (new)

07_GriffitITGA12e_11478_Ch05_153_192.indd 185 03/10/19 12:03 PM

 186 C H A P T E R 5 ! ! Gene Interaction

a. Devise a genetic explanation of these three crosses. Do these results support the hypothesis that the
Show genotypes for the parents, the F1, and the F2 original plant was P /p ; Q /q?
in the three crosses, and make sure that you show 67. A plant of phenotype 1 was selfed, and, in the progeny,
the alleles of each gene that you hypothesize for there were 100 plants of phenotype 1 and 60 plants of
every mink. an alternative phenotype 2. Are these numbers com-
b. Predict the F1 and F2 phenotypic ratios from cross- patible with expected ratios of 9: 7, 13:3, and 3:1?
ing sapphire with platinum and with aleutian pure Formulate a genetic hypothesis on the basis of your
lines. calculations.
63. In Drosophila, an autosomal gene determines the 68. Four homozygous recessive mutant lines of Drosoph-
shape of the hair, with B giving straight and b giving ila melanogaster (labeled 1 through 4) showed abnor-
bent hairs. On another autosome, there is a gene of mal leg coordination, which made their walking highly
which a dominant allele I inhibits hair formation so erratic. These lines were intercrossed; the phenotypes
that the fly is hairless ( i has no known phenotypic of the F1 flies are shown in the following grid, in which
effect). “+ ” represents wild-type walking and “#” represents
a. If a straight-haired fly from a pure line is crossed abnormal walking:
with a fly from a pure-breeding hairless line 1 2 3 4
known to be an inhibited bent genotype, what
1 # + + +
will the genotypes and phenotypes of the F1 and
the F2 be? 2 + # # +
b. What cross would give the ratio 4 hairless: 3 + # # +
3!straight:1 bent? 4 + + + #
64. The following pedigree concerns eye phenotypes in
a. What type of test does this analysis represent?
Tribolium beetles. The solid symbols represent black
eyes, the open symbols represent brown eyes, and the b. How many different genes were mutated in creat-
cross symbols (X) represent the “eyeless” phenotype, ing these four lines?
in which eyes are totally absent. c. Invent wild-type and mutant symbols, and write
out full genotypes for all four lines and for the F1
I
1 2 3 flies.
d. Do these data tell us which genes are linked? If not,
II
1 2 3 4 5 how could linkage be tested?
e. Do these data tell us the total number of genes
III taking part in leg coordination in this animal?
1 2 3 4
IV 69. Three independently isolated tryptophan-requiring
1 mutants of haploid yeast are called trpB, trpD, and
trpE. Cell suspensions of each are streaked on a plate
a. From these data, deduce the mode of inheritance of
of nutritional medium supplemented with just enough
these three phenotypes.
tryptophan to permit weak growth for a trp strain.
b. Using defined gene symbols, show the genotype of The streaks are arranged in a triangular pattern so that
beetle II-3. they do not touch one another. Luxuriant growth is
65. A plant believed to be heterozygous for a pair of alleles noted at both ends of the trpE streak and at one end of
B /b (where B encodes yellow and b encodes bronze) the trpD streak (see the figure below).
was selfed, and, in the progeny, there were 280 yellow
and 120 bronze plants. Do these results support the
hypothesis that the plant is B /b?
66. A plant thought to be heterozygous for two inde-
pendently assorting genes (P /p ; Q /q) was selfed, and
the progeny were
88 P /– ; Q /–! 25 p /p ; Q /–
32 P /– ; q /q 14 p /p ; q /q

07_GriffitITGA12e_11478_Ch05_153_192.indd 186 03/10/19 12:03 PM

 CHAPTER 5 REVIEW 187

a. Do you think complementation has a role?

b. Briefly explain the pattern of luxuriant growth.
c. Draw the enzymatic steps that are defective
in!mutants trpB, trpD, and trpE in order in the
tryptophan-synthesizing pathway.
d. Why was it necessary to add a small amount of
tryptophan to the medium to demonstrate such a
growth pattern?
Long Sphere Disk
CHALLENGING PROBLEMS

70. A pure-breeding strain of squash that produced disk- long 32 sphere 178 disk 270
shaped fruits (see the accompanying illustration) was
crossed with a pure-breeding strain having long fruits. Propose an explanation for these results, and show the
The F1 had disk fruits, but the F2 showed a new pheno- genotypes of the P, F1, and F2 generations.
type, sphere, and was composed of the following pro- 71. Marfan’s syndrome is a disorder of the fibrous connec-
portions, shown in the next column: tive tissue, characterized by many symptoms, including
long, thin digits; eye defects; heart disease; and long
limbs. (Flo Hyman, the American volleyball star, suf-
fered from Marfan’s syndrome. She died from a rup-
tured aorta.)

II

III

Symptoms
Unknown, presumed normal Eye lens displacement Long fingers and toes
Examined, normal Congenital heart disease Very long, thin fingers and toes
Questionably affected

a. Use the pedigree above to propose a mode of inher- dominant allele, as specified in Problem 72. The domi-
itance for Marfan’s syndrome. nant allele Pr of a fourth independently assorting gene
b. What genetic phenomenon is shown by this is required to convert the biochemical precursor into
pedigree? a purple pigment, and its recessive allele pr makes the
pigment red. Plants that do not produce pigment have
c. Speculate on a reason for such a phenomenon. yellow seeds. Consider a cross of a strain of genotype
(Data from J. V. Neel and W. J. Schull, Human Heredity. A /A ; C /C ; R /R ; pr /pr with a strain of genotype a /a ;
University of Chicago Press, 1954.) c /c ; r /r ; Pr /Pr.
72. In corn, three dominant alleles, called A, C, and R, a. What are the phenotypes of the parents?
must be present to produce colored seeds. Genotype b. What will be the phenotype of the F1?
A /# ; C/# ; R /# is colored; all others are colorless. A col-
ored plant is crossed with three tester plants of known c. What phenotypes, and in what proportions, will
genotype. With tester a /a ; c /c ; R /R, the colored plant appear in the progeny of a selfed F1?
produces 50 percent colored seeds; with a /a ; C /C ; r /r, d. What progeny proportions do you predict from the
it produces 25 percent colored; and with A /A ; c /c ; r /r, testcross of an F1?
it produces 50 percent colored. What is the genotype 74. The allele B gives mice a black coat, and b gives a
of the colored plant? brown one. The genotype e /e of another, independently
73. The production of pigment in the outer layer of seeds of assorting gene prevents the expression of B and b,
corn requires each of the three independently assorting making the coat color beige, whereas E/– permits
genes A, C, and R to be represented by at least one the expression of B and b. Both genes are autosomal.

07_GriffitITGA12e_11478_Ch05_153_192.indd 187 03/10/19 12:03 PM

 188 C H A P T E R 5 ! ! Gene Interaction

In! the following pedigree, black symbols indicate A third pathway, whose compounds do not
a black coat, pink symbols indicate brown, and contribute pigment to petals, normally does not affect
unshaded symbols indicate beige. the blue and red pathways, but, if one of its interme-
diates (white3) should build up in concentration, it can
I
1 2 be converted into the yellow intermediate of the red
pathway.
In the diagram, the letters A through E represent
enzymes; their corresponding genes, all of which are
II
1 2 3 4 5 6 unlinked, may be symbolized by the same letters.
Assume that wild-type alleles are dominant and
encode enzyme function and that recessive alleles result
in a lack of enzyme function. Deduce which combi-
III
1 2 3 4 5 6 7 nations of true-breeding parental genotypes could be
crossed to produce F2 progeny in the following ratios:
a. What is the name given to the type of gene interac- a. 9 purple : 3 green: 4 blue
tion in this example?
b. 9 purple : 3 red : 3 blue : 1 white
b. What are the genotypes of the individual mice in
c. 13 purple: 3 blue
the pedigree? (If there are alternative possibilities,
state them.) d. 9 purple : 3 red : 3 green : 1 yellow
75. A researcher crosses two white-flowered lines of Antir- ( Note: Blue mixed with yellow makes green; assume
rhinum plants as follows and obtains the following that no mutations are lethal.)
results: 77. The flowers of nasturtiums ( Tropaeolum majus )
pure line 1� pure line 2 may be single (S), double (D), or superdouble (Sd).
Superdoubles are female sterile; they originated from
%
a double-flowered variety. Crosses between varieties
F1 all white gave the progeny listed in the following table, in which
F1 � F1 pure means “pure breeding.”
% Cross Parents Progeny
F2 131 white 1 pure S � pure D All S
29 red 2 cross 1 F1 � cross 1 F1 78 S : 27 D
a. Deduce the inheritance of these phenotypes; use 3 pure D � Sd 112 Sd : 108 D
clearly defined gene symbols. Give the genotypes of
the parents, F1, and F2. 4 pure S � Sd 8 Sd : 7 S

b. Predict the outcome of crosses of the F1 with each 5 pure D � cross 4 Sd progeny 18 Sd : 19 S
parental line. 6 pure D � cross 4 S progeny 14 D : 16 S
76. Assume that two pigments, red and blue, mix to give the
Using your own genetic symbols, propose an explana-
normal purple color of petunia petals. Separate biochem-
tion for these results, showing
ical pathways synthesize the two pigments, as shown in
the top two rows of the accompanying diagram. “White” a. all the genotypes in each of the six rows.
refers to compounds that are not pigments. (Total lack b. the proposed origin of the superdouble.
of pigment results in a white petal.) Red pigment forms
78. In a certain species of fly, the normal eye color is
from a yellow intermediate that is normally at a concen-
red (R). Four abnormal phenotypes for eye color
tration too low to color petals.
were found: two were yellow (Y1 and Y2), one was
pathway I … › white1 !!!
E
� blue brown (B), and one was orange (O). A pure line
pathway II … › white2 !!!
A
� yellow !!!
� red B was established for each phenotype, and all possible
combinations of the pure lines were crossed. Flies of
c each F1 were intercrossed to produce an F2 . The F1 and
the F2 flies are shown within the following square; the
pathway III … � white3 !!!
D
� white4 pure lines are given at the top and at the left-hand side.

07_GriffitITGA12e_11478_Ch05_153_192.indd 188 03/10/19 12:03 PM

 CHAPTER 5 REVIEW 189

Y1 Y2 B O 80. The following pedigree shows the inheritance of

deaf-mutism.
F1 all Y all R all R all R
I 1 2 3 4
Y1 F2 all Y 9R 9R 9R
7Y 4Y 4O
3B 3Y II
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
F1 all Y all R all R
Y2 F2 all Y 9R 9R
4Y 4Y III
1 2 3 4 5 6 7
3B 3O
a. Provide an explanation for the inheritance of this
F1 all B all R rare condition in the two families in generations I
B F2 all B 9R and II, showing the genotypes of as many persons
4O as possible; use symbols of your own choosing.

3B b. Provide an explanation for the production of only

normal persons in generation III, making sure that
F1 all O
your explanation is compatible with the answer to
O F2 all O part a.
81. The pedigree below is for blue sclera (bluish thin outer
a. Define your own symbols, and list the genotypes of
wall of the eye) and brittle bones.
all four pure lines.
b. Show how the F1 phenotypes and the F2 ratios are I 3
1 2
produced.
c. Show a biochemical pathway that explains the II
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
genetic results, indicating which gene controls
which enzyme.
III
79. In common wheat, Triticum aestivum, kernel color is 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
determined by multiply duplicated genes, each with
an R and an r allele. Any number of R alleles will IV
1 2 3 4 5 6 7 8 9 10 11 12 13
give red, and a complete lack of R alleles will give
the white phenotype. In one cross between a red pure , ! blue sclera ! brittle bones
63 1
line and a white pure line, the F2 was 64 red and 64
a. Are these two abnormalities caused by the same
white.
gene or by separate genes? State your reasons clearly.
a. How many R genes are segregating in this
b. Is the gene (or genes) autosomal or sex-linked?
system?
c. Does the pedigree show any evidence of incom-
b. Show the genotypes of the parents, the F1, and
plete penetrance or expressivity? If so, make the
the!F2.
best calculations that you can of these measures.
c. Different F2 plants are backcrossed with the white
82. Workers of the honeybee line known as Brown (noth-
parent. Give examples of genotypes that would
ing to do with color) show what is called “hygienic
give the following progeny ratios in such back-
behavior”; that is, they uncap hive compartments
crosses: (1) 1 red : 1 white, (2) 3 red : 1 white, (3) 7
containing dead pupae and then remove the dead
red : 1 white.
pupae. This behavior prevents the spread of infec-
d. What is the formula that generally relates the num- tious bacteria through the colony. Workers of the Van
ber of segregating genes to the proportion of red Scoy line, however, do not perform these actions, and
individuals in the F2 in such systems? therefore this line is said to be “nonhygienic.” When

07_GriffitITGA12e_11478_Ch05_153_192.indd 189 03/10/19 12:03 PM

 190 C H A P T E R 5 ! ! Gene Interaction

a queen from the Brown line was mated with Van a. Explain the inheritance of these colors.
Scoy drones, all the F1 were nonhygienic. When drones b. Write the genotypes of the parents, the F1, and
from this F1 inseminated a queen from the Brown line, the!F2.
the progeny behaviors were as follows:
84. Consider the following F1 individuals in different spe-
1 hygienic
4 cies and the F2 ratios produced by selfing:
1 uncapping but no removing of pupae
4
F1 Phenotypic ratio in the F2
1 nonhygienic
2 12 3 1
1 cream 16
cream 16
black 16
gray
However, when the compartment of dead pupae
9 7
was uncapped by the beekeeper and the nonhygienic 2 orange 16
orange 16
yellow
honeybees were examined further, about half the bees 13 3
3 black 16
black 16
white
were found to remove the dead pupae, but the other
half did not. 4 solid red 9 3 4
16
solid red 16
mottled red 16
small red dots
a. Propose a genetic hypothesis to explain these beha-
vioral patterns. If each F1 were testcrossed, what phenotypic ratios
would result in the progeny of the testcross?
b. Discuss the data in relation to epistasis, domi-
nance, and environmental interaction. 85. To understand the genetic basis of locomotion in
the diploid nematode Caenorhabditis elegans, reces-
(Note: Workers are sterile, and all bees from one line
sive mutations were obtained, all making the worm
carry the same alleles.)
“wiggle” ineffectually instead of moving with its
83. The normal color of snapdragons is red. Some pure usual smooth gliding motion. These mutations pre-
lines showing variations of flower color have been sumably affect the nervous or muscle systems. Twelve
found. When these pure lines were crossed, they gave homozygous mutants were intercrossed, and the F1
the following results (see the table): hybrids were examined to see if they wiggled. The
results were as follows, where a plus sign means that
Cross Parents F1 F2
the F1 hybrid was wild type (gliding) and “w” means
1 orange � yellow orange 3 orange : 1 yellow that the hybrid wiggled.
2 red � orange red 3 red : 1 orange
3 red � yellow red 3 red : 1 yellow
4 red � white red 3 red : 1 white
5 yellow � white red 9 red : 3 yellow : 4 white
6 orange � white red 9 red : 3 orange : 4 white
7 red � white red 9 red : 3 yellow : 4 white

1 2 3 4 5 6 7 8 9 10 11 12
1 w + + + w + + + + + + +
2 w + + + w + w + w + +
3 w w + + + + + + + +
4 w + + + + + + + +
5 w + + + + + + +
6 w + w + w + +
7 w + + + w w
8 w + w + +
9 w + + +
10 w + +
11 w w
12 w

07_GriffitITGA12e_11478_Ch05_153_192.indd 190 03/10/19 12:04 PM

 CHAPTER 5 REVIEW 191

a. Explain what this experiment was designed to test. b. What cross(es) would they do to make these
b. Use this reasoning to assign genotypes to all 12 worms?
mutants. c. What results would they expect in the F2 if
c. Explain why the phenotype of the F1 hybrids (1) it did act as a suppressor of bli-4?
between mutants 1 and 2 differed from that of the (2) it did not act as a suppressor of bli-4?
hybrids between mutants 1 and 5.
88. Six proline-requiring auxotrophic mutants were
86. A geneticist working on a haploid fungus makes a obtained in the haploid fungus Saccharomyces cerevisiae
cross between two slow-growing mutants called mossy (yeast). Each was crossed to wild type in order to
and spider (referring to the abnormal appearance of obtain the mutants in each mating type (“sex”), then
the colonies). Tetrads from the cross are of three types all combinations were crossed and the resultant ran-
(A, B, C), but two of them contain spores that do not dom ascospores were plated onto minimal medium. In
germinate. some cases, proline-independent colonies were obtained
Spore A B C (numbers are not shown); but in other cases, none of
the ascospores grew. The results are summarized in
1 wild type wild type spider
the following table, in which + indicates presence of
2 wild type spider spider colonies, and 0 indicates no colonies. Formulate a
3 no germination mossy mossy hypothesis that explains the results of each cross.
4 no germination no germination mossy 1 2 3 4 5 6
1 0 + + 0 + 0
Devise a model to explain these genetic results, and
propose a molecular basis for your model. 2 + 0 0 + + +
87. In the nematode C. elegans, some worms have blistered 3 + 0 0 + + +
cuticles due to a recessive mutation in one of the bli 4 0 + + 0 + 0
genes. Someone studying a suppressor mutation that 5 + + + + 0 +
suppressed bli-3 mutations wanted to know if it would
6 0 + + 0 + 0
also suppress mutations in bli-4. They had a strain that
was homozygous for this recessive suppressor muta-
GENETICS AND SOCIETY
tion, and its phenotype was wild type.
a. How would they determine whether this recessive 1. How might a recessive lethal allele reveal itself in a
suppressor mutation would suppress mutations in human pedigree?
bli-4? In other words, what is the genotype of the 2. How has Beadle and Tatum’s pioneering work influ-
worms required to answer the question? enced the therapy of human disease?

07_GriffitITGA12e_11478_Ch05_153_192.indd 191 03/10/19 12:04 PM