Mouse Genetic Portrait

har06584_refE_109-132
11/06/2006
06:11 PM
Page 109
Mus musculus: Genetic

Portrait of the House Mouse
The common house mouse, Mus musculus, has played a prominent
role in the study of genetics ever since Carl Correns, Hugo De Vries,
and Erich Von Tschermak independently rediscovered Mendels laws
at the beginning of the twentieth century. Because these three scientists, as well as Mendel himself, performed their research entirely on
plants, many in the scientific community questioned whether
Mendels laws could explain the basis for inheritance in animals, especially humans. The reason for this skepticism is easy to see. People, for
example, differ in the expression of many commonly inherited traits
such as skin color, eye color, curliness of hair, and heightthat show
no evidence of transmission according to Mendels laws. We now
know that these traits result from the interaction of many genes with
multiple alleles that each segregate according to Mendels first law
even though the traits themselves do not. At the beginning of the
twentieth century, however, a demonstration of the applicability of
Mendels laws to animal inheritance required the analysis of simple
traits controlled by single genes.
M. musculus has many features that enhance its value as a model organism for
genetic analysis, and foremost among these is the availability of hundreds of singlegene mutations. These mutations arose during the mouses long history of domestication as a pet. Over the centuries, dealers in what became known as the fancy
mouse trade selected and bred mice with numerous coat colors and other visible
mutations, first in China and Japan, later in Europe (Fig. E.1a). In contrast to the
variation that occurs naturally in wild populations, new traits that appear suddenly
in captive-bred mice are almost always the result of single-gene mutations. Early
animal geneticists made note of this fact and used fancy mice to demonstrate that
Mendels laws apply to mammals and, by extrapolation, to humans.
In addition to providing a ready source of single-gene mutations, the house
mouse has several other features that make it the mammal of choice for genetic
analysis. Mice have a very short generation time of just eight to nine weeks. They
are small enough so that thousands can live in relatively small rooms. They have
large litters of eight or more pups. They breed readily in captivity. Fathers do not
harm their young. And after centuries of artificial selection, domesticated mice are
docile and easy to handle (Fig. E.1b).
But why study a mammal at all when animals like fruit flies and nematodes are
even smaller and more amenable to genetic analysis? The answer is that a major
goal of current biological research is the understanding of human beings. And
although many features of human biology, especially at the cellular and molecular
levels, are common to a broad spectrum of life-forms, the most advanced organismlevel human characteristics appear in a limited subset of animals. In fact, many
aspects of human development and disease are common only to placenta-bearing
mammals such as the mouse. Thus, the mouse provides a powerful model system
for investigating the genetic basis of simple and complex human traits, especially
those related to development and disease (Fig. E.2).
Two general themes emerge from our presentation of M. musculus. First, because
of the many similarities between mouse and human genomes, researchers can use
Reference
A member of the 129 strain of

inbred mice commonly used in
targeted mutagenesis studies.
109
har06584_refE_109-132
110
11/06/2006
06:11 PM
Page 110
Reference E Mus musculus: Genetic Portrait of the House Mouse
(a)
(a)
(b)
(b)
Figure E.1
The mouse is a model system for human

biology. (a) Examples of visible phenotypes caused by single-gene
mutations. (b) Mother mouse with her pups.
Figure E.2
Hirschsprung disease: A human developmental

disease that causes deformities of the colon (a) and (b) a mouse
model of the disease with a classical mutation (piebald coat).
homology analysis to identify and locate the same genes in both species. In this context, homologs are genes or regulatory DNA sequences that are similar in different
species because of descent from a common ancestral sequence. Second, genetic
analysis in mice exemplifies the combined use of molecular (that is, recombinant
DNA) and classical breeding techniques to identify and understand the function of
complex genetic systems.
Our genetic portrait of the house mouse describes:
An overview of M. musculus in the laboratory, including a look at the
mouse genome, the mouse life cycle, and two powerful transgenic protocols:
the addition of specific genes to the mouse genome by nuclear injection
and the removal of specific genes from the mouse genome by targeted
mutagenesis.
The uses of transgenic technology in determining the function of gene products, characterizing regulatory regions, establishing links between mutant
phenotypes and particular transcription units, and creating a mouse model
for a human disease.
The Hox genes: a comprehensive example.
har06584_refE_109-132
11/06/2006
06:11 PM
Page 111
E.1 An Overview of Mus musculus in the Laboratory
TABLE E.1
111
Comparison of Mice and Humans
Trait
Mice
Humans
Average weight
30 g
77,000 g (170 lb)
Average length
10 cm (without tail)
175 cm
Genome size
~3,000,000,000 bp
~3,000,000,000 bp
Haploid gene number
~25,000
~25,000
Number of chromosomes
19 autosomes X and Y
22 autosomes X and Y
Gestation period
3 weeks
Average, 38 weeks (8.9 months)
Age at puberty
56 weeks
Average, 624728 weeks (1214 years)
Estrus cycle
4 days
Average, 28 days
Life span
2 years
Average, 78 years
E.1
An Overview of Mus
musculus in the Laboratory
The Mouse Genome

The most important feature of the mouse genome for contemporary geneticists is its close resemblance to the human
genome (Table E.1). The haploid genomes of humans and
mice (and other placental mammals as well) contain approximately 3 billion base pairs of DNA. Nearly every human gene has a homolog in the mouse genome. This does
not mean that the two genomes are equivalent in content.
Nearly all differences, however, appear to result from
species-specific additions to gene families that already
existed in the common ancestor of mice and humans.
The human genome is distributed among 22 autosomes
and 2 sex chromosomes, while the mouse genome is contained within 19 autosomes and 2 sex chromosomes. As we
saw at the beginning of Chapter 14 of the main textbook,
examinations of mouse and human karyotypes under the
microscope reveal no evidence of chromosome banding
similarities between the two species. With the mapping of
thousands of homologous genes in both species, however, a
remarkable pattern has emerged. Genes that are closely
linked in one species are usually closely linked in the other.
When two or more loci are found to be linked in one
species, they are said to be syntenic (meaning on the same
thread, or chromosome). When the same set of loci are
also found to be linked in a second species, they are said to
exist in a state of conserved synteny. A comparison of genetic maps of the whole mouse genome with genetic maps
of the whole human genome shows that regions of conserved synteny extend across nearly the complete length of
both. The average size of each conserved syntenic region is
roughly 17.6 Mb. The implication of this finding is that
during the 75 million years that mice and humans have

been evolving apart from a common ancestor, their
genomes have broken apart and rearranged some 170 times
(17.6 Mb 170 about 3000 Mb the size of the mammalian genome). Conversely, if the proper genome-scale
scissors and glue were available, one could break the
mouse genome into about 170 pieces and reassemble those
pieceslike a puzzlein the form of the human genome.
In addition to its powerful evolutionary implications,
conserved synteny is a useful tool for practicing geneticists.
Once a researcher has mapped a locus in one species, he or
she can look at a homology map and immediately identify
its likely map position in the other species. Of course, for
genes that have already been cloned, it is possible to use
DNA-DNA hybridization, or computer analysis of a wholegenome sequence, to pick out gene homologs from the
other species. But for loci characterized only by their phenotypic expression, conserved synteny enables geneticists
to move back and forth between the analysis of a trait in humans and the analysis of a model for that trait in mice.
The discovery of a locus that predisposes female mice
to excessive consumption of 10% ethanol (the concentration of alcohol found in many wines) provides an example
of the use of conserved synteny for locating human homologs of mouse genes. Mouse geneticists used DNA
markers (as described in Chapter 11 of the main textbook)
to map the Alcohol-preference-2 (Alcp2) locus to the middle of mouse chromosome 11. Now that the whole mouse
genome has been cloned and sequenced, the genes as transcription units in Alcpz region have been identified, but
which one is actually Alcpz is not yet known. However,
even though researchers have not yet identified the specific
gene, scrutiny of a conserved synteny homology map
shows that the most likely location for the human homolog
of Alcp2 is on the short arm of human chromosome 17,
close to the centromere (Fig. E.3). With this information
about the likely location of an alcohol-preference locus,
har06584_refE_109-132
112
11/06/2006
06:11 PM
Page 112
Boxes show
regions of
conserved
synteny in
the human
genome
Male
Female
Oogonia
Birth
Testes
Spermatogonia
Ovary
Fetus
Implantation
Blastocyst
Zygote
Fertilization
Sperm
Egg
Primary
oocytes
Birth
Ovulation
Mature
secondary
oocyte
Spermatozoa
Diploid phase
Haploid phase
Figure E.4
The mammalian life cycle. The life cycles of males

and females are shown separately in the traditional format. Notice
that primary germ cells are formed in the female before birth
but in the male after birth. Notice also that male germ cells are
self-renewing, while in females, no new germ cells are formed
after birth.
Alcp2
Mouse
chromosome
Mouse
loci
Human
Human
homologs map positions
Figure E.3
Conserved synteny: An example with mouse

chromosome 11. The mouse chromosome, on the left, shows
positions of loci with mapped homologs in the human genome.

The map locations of human homologs are on the right.
human geneticists can now look for linkage between DNA

markers on human chromosome 17 and a phenotypic
expression of a predisposition to alcohol abuse.
The Mammalian Life Cycle

The mouses life cycle is similar to that of humans and all
other placental mammals, although the timing of events is
unique to each species (see Table E.1). For example, for
mice, the average life span is just 2 years; for humans it averages 78 years. Gestationthe period from fertilization to
birthlasts only 21 days in mice but 8.9 months in humans.
After birth, mice reach puberty in just 56 weeks, 100 times
sooner than humans, who mature for roughly 12 years before
they attain puberty and the ability to conceive children of
their own (although in most societies, they wait 510 years
longer). It is thus possible to go from the birth of one mouse
to the birth of its offspring in just 8 weeks, whereas with humans, completion of the same cycle would take 1325 years.
Even with significant differences in timing, however,
the details of each stage of development, both before and
after birth, are remarkably similar in all mammalian
species. As a result, the overview presented here applies
equally to mice and humans.
The mammalian life cycle, like that of all sexually reproducing species, can be visualized as continuous circles
(Fig. E.4), with any point along the circumferences marking the start. For ease of presentation, we begin with the
haploid phase in both males and females.
Male Germ Cell Development

As we saw in Chapter 4 of the main textbook, once a male
mammal has reached puberty, he continuously produces a
large number of haploid germ cells for the rest of his life.
The mature haploid cell is a sperm cell, or spermatozoa,
and the process by which it arises is spermatogenesis (review Fig. 4.19 in the main textbook). Mature spermatozoa
released into the lumen at the center of the seminiferous
tubule join millions of other sperm cells, which together
pass through countless passageways to reach the epididymis;
there they mature further and then continue on to the vas
deferens, where they await ejaculation during copulation.
Female Germ Cell Development

Unlike males, females are born with all the haploid cells
they will ever have (~50,000 in the mouse; 1 million in
women). When mature, these haploid cells are known as
eggs, or oocytes, and the process by which they arise is
oogenesis (review Fig. 4.18 in the main textbook). As we
saw in Chapter 4, oogenesis begins inside the newly
formed ovaries of the developing fetus. Long before birth,
primordial germ cells differentiate into oogonia and enter
meiosis, but they stop at the diplotene stage of the first
meiotic prophase. These primary oocytes remain arrested
in suspended animationfor weeks in mice and many
years in humansuntil after puberty.
har06584_refE_109-132
11/06/2006
06:11 PM
Page 113
From this time on, the female progresses through an

estrus cycle that lasts about 4 days in mice and about 28
days in humans. During each cycle, primary oocytes (810
in mice, usually only 1 in women) are stimulated to complete the first meiotic division and extrude the first polar
body at the end of this division; the resulting secondary
oocyte begins the second meiotic division but stops at
metaphase and is released from the ovary in a process
known as ovulation. Following ovulation, the secondary
oocyte passes into an oviduct (called a fallopian tube in humans), where for a brief time, known as estrus, it remains
alive and receptive to fertilization.
In nature, most mammals die while they still have the
ability to reproduce. Many human females, however, live
long enough to pass through menopause, during which
they stop cycling through estrus, no longer ovulate, and
thus lose the ability to reproduce.
Fertilization
Just before and during the estrus phase of the estrus cycle,
female mammals of nonhuman species release speciesspecific chemical signals, or pheromones. In a behavioral
response to these pheromones, a male will copulate with a
female and ejaculate semen containing millions of sperm
into her reproductive tract. The sperm swim from the
vagina into the uterus and thence up the oviducts. Only 100
or fewer sperm survive this journey to the waiting eggs.
Fertilization is a multistep process illustrated in Fig. E.5.
First, surviving sperm bind to the zona pellucidathe thick
solid shell composed of glycoproteins that surrounds the egg
proper. The act of binding induces each sperm to release special proteases that enable it to burn its way through the zona
113
pellucida into the space that surrounds the egg membrane.

Although multiple sperm can make it into this space, usually
only one fuses with the egg. This fusion causes rapid electrochemical changes in the egg membrane that prevent the entry
of additional sperm and activate the newly fertilized egg to
enter the pathway of animal development.
After fusion, the fertilized egg, or zygote, contains two
haploid pronuclei. The two pronuclei never merge; instead,
replication occurs within both of the pronuclei. The onecell embryo carries two replicated pronuclei right up to the
moment of the first mitosis, at which time the membranes
of the two pronuclei break down, and the two sets of chromosomes, one from the paternal pronucleus, the other from
the maternal pronucleus, align along the midplane of the
fertilized egg and thence segregate chromatids into the two
daughter cells.
For the purposes of analysis, scientists divide mouse
development into two distinct stages of unequal length,
separated by the process of embryonic implantation into
the uterus: a preimplantation stage that lasts 45 days in
mice, and a postimplantation stage that lasts about 16.5
days in mice. During the preimplantation phase, the embryo is a free-floating object within the females body. It is
easy to remove this naturally free-floating preimplantation
embryo from the animal, culture it in a petri plate, and
expose it to genetic manipulation before placing it back
in the reproductive tract of an adult female for development
to a newborn animal. After implantation, however, such
manipulation is no longer possible because the embryo, if
removed from the adults body, cannot be returned. The
accessibility of the preimplantation embryo provides the
basis for many of the genetic manipulations researchers
use to study mammalian development.
Fertilized egg = zygote = 1-cell embryo
2-cell
embryo
Fertilization
4-cell
embryo
Zona pellucida
Maternal pronucleus
sperm head
expands
Ovulation
Second
cleavage
First
cleavage
Sperm
One sperm penetrates
8-cell
embryo
Paternal pronucleus
Blastocyst
16-cell embryo:
Blastocoele
cavity
Third
cleavage
Fourth
cleavage
Blastocyst
formation
outside cells differentiate
into trophectoderm
Figure E.5
Trophectoderm
Early development of mammals from fertilization to implantation.
hatching, and
implantation
Inner cell mass
(ICM)
Uterine wall
har06584_refE_109-132
114
11/06/2006
06:11 PM
Page 114
For Most of the Preimplantation Stage, the

Embryonic Cells Remain Undifferentiated
The preimplantation stage starts with the zygote (Fig. E.5).
Development proceeds slowly in the beginning, with the
first 22 hours devoted to the expansion of the highly compacted sperm head into a paternal pronucleus that matches
the size of the eggs maternal pronucleus. After the paternal
pronucleus has completed its expansion and replicated its
chromosomes and the maternal pronucleus has replicated
its chromosomes, the embryo undergoes the first of four
equal divisions, or cleavages, that increase the number of
cells from 1 to 16 over 60 hours.
The period of these four equal divisions is called the
cleavage stage. During this stage, all the cells in the developing embryo are equivalent and totipotent, that is, they have
not yet differentiated and each one retains the ability, or
potency, to produce every type of cell found in the developing
embryo and adult animal. This is very different from the developmental patterns found in most nonmammalian animal
species, including Caenorhabditis elegans, where totipotency disappears as early as the two-cell stage. Because of the
mouse cells totipotency, cleavage-stage embryos can be
divided into smaller groups of cells that each have the
potential to develop into a normal individual. Identical human twins, or more rarely, identical triplets or quadruplets,
are examples of the outcome of this process. (Twinning is impossible in C. elegans or Drosophila melanogaster.) In the
laboratory, scientists have obtained completely normal mice
from individual cells that they dissected out of the fourcell-stage mouse embryo and placed back into the female
reproductive tract (Fig. E.6a). This experimental feat demonstrates the theoretical possibility of obtaining four identical
clones from a single embryo of any mammalian species.
Another more bizarre consequence of the equivalency
of cleavage-stage cells is the formation of chimeras, which
are the opposite of clones (Fig. E.6b). The term chimera
comes from the Greek word for a mythological beast that is
part lion, part goat, and part serpent. Geneticists use the term
to designate an embryo or animal composed of cells from
two or more different origins. The Polish embryologist Andrezej Tarkowski reported the first mouse chimeras in 1961.
To construct them, he removed the zona pellucida from two
cleavage-stage mouse embryos, obtaining denuded cell
masses that are naturally sticky; he then pushed the sticky
denuded embryos up against each other. Denuded embryos
pressed together in this way form a single chimeric cell mass
that is capable of undergoing normal development within the
female reproductive tract. If the two embryos of a chimera
come from different females mated to different males, the resulting individual is tetraparental, that is, has four parents. It
is also possible to produce hexaparental animals derived
from a combination of three embryos. Every organ and tissue in the adultincluding the germ linecan contain cells
derived from all three original embryos. As we see later, the
production of chimeric mice has been an essential compo-
Identical quadruplets
(a)
Four-cell embryo
(b)
Two four-cell embryos
Chimeric embryo
Chimeric mouse
Embryo 1
Embryo 2
Figure E.6
Early mammalian embryos are highly malleable

in their development. There is no requirement for a one-to-one
correspondence between embryo and adult. (a) Creating identical

quadruplets from a single fertilized egg. (b) Creating a single
chimeric animal from the fusion of two embryos.
nent of the targeted mutagenesis technology that has revolutionized the use of the mouse as a model organism for studying human diseases.
A comparison of the early developmental program of
placental mammals with that of other animals, including
C. elegans and D. melanogaster, shows how different these
programs can be. In nematodes, embryonic cells are highly
restricted in their developmental potential, or fate, beginning at the two-cell stage; and in fruit flies, polarization of
the egg before fertilization generates distinct cytoplasmic
regions dedicated to supporting different developmental
programs within the nuclei that end up in these locations.
Consequently, half a nematode embryo or half a fly embryo
can never give rise to a whole animal.
Events Restricting the Developmental

Potency of Individual Cells Occur Near
the End of the Preimplantation Stage
The first differentiation events of mouse embryogenesis
occur in the 16-cell embryo (see Fig. E.5). The cells on the
outside of the embryo turn into a trophectoderm layer that
har06584_refE_109-132
11/06/2006
06:11 PM
Page 115
will eventually take part in the formation of the placenta.

Soon thereafter, the cells on the inside of the embryo compact into a small clump called the inner cell mass, or ICM,
that remains attached to one spot along the inside of the
hollow trophectoderm sphere. The entire fetus and animal
are derived entirely from the cells of the ICM. As Figure E.5
shows, compaction of the ICM causes the appearance of a
fluid-filled space that is devoid of cellular material and surrounded by trophectoderm. This space is called the blastocoel cavity. The embryo is now called a blastocyst. Two
more rounds of cell division occur during the blastocyst
stage, producing the 64-cell embryo that implants.
Throughout normal preimplantation development, the
embryo remains protected within the inert zona pellucida.
As a result, there is no difference in size between the 1-cell
zygote and the 64-cell embryo. To accomplish implantation, the embryo must first hatch from the zona pellucida
so that it can make direct membrane-to-membrane contact
with cells in the uterine wall. Embryonic and fetal development within a uterus inside the body of a female is a characteristic unique to all mammals except the primitive
egg-laying platypus.
After Implantation, the Placenta Develops,

the Embryo Grows, and the Tissues and
Organs Emerge
Implantation initiates development of the placenta, a mix
of embryonic and maternal tissues that mediates the flow of
nutrients entering the embryo from the maternal blood supply and the flow of waste products exiting the embryo to
the maternal circulation. The placenta maintains this intimate connection between mother and embryo, and later between mother and fetus, until the time of birth.
Development of the placenta enables a period of rapid embryonic growth. Cells from the ICM differentiate into the
three germ layers of endoderm, ectoderm, and mesoderm
during a stage known as gastrulation. The foundation of the
spinal cord is put into place, and the development of the
various adult tissues and organs begins. With the appearance of organs, the embryo becomes a fetus, which continues to grow rapidly. Birth occurs at about 21 days after
conception. Newborn animals remain dependent on their
mothers during a suckling period that lasts 1825 days. By
five to eight weeks after birth, mice reach adulthood and
are ready to begin the next reproductive cycle.
Two Powerful Transgenic Techniques

for Analyzing the Mouse Genome
Geneticists have capitalized on certain features of the
mouse genome and the mouse life cycle to develop protocols that make it possible to add and remove specific genes
from embryonic or germ cells.
115
The Addition of Genes to the Mouse

Genome by Nuclear Injection
The 1981 development of a method for inserting foreign
DNA into the germ line of mice thrust a primarily observational discipline into the realm of genetic engineering
with all its implications. Yet the incredibly powerful transgenic technology is based on a very simple process. Recall
that a transgene is any piece of foreign DNA that
researchers have inserted into the genome of a complex
organism, such as a mouse or a pea plant, through experimental manipulation of early stage embryos or germ cells;
any individual carrying a transgene is known as a transgenic
animal or plant.
To create a transgenic mouse carrying a foreign DNA
sequence integrated into one of its chromosomes, a researcher simply injects foreign DNA into a pronucleus of a
fertilized egg and then places the injected one-cell embryo
back into a female oviduct, where it can continue its development. Roughly 25% to 50% of the time, for a skilled
investigator, the injected DNA will integrate at random into
a chromosomal location. Integration can occur while the
embryo is still in the one-cell stage, in which case the transgene will appear in every cell of the adult body. Or integration may occur somewhat later, after the embryo has
completed one or two cell divisions; in this case, the mouse
will be a mosaic of cells, some with the transgene and some
without it. The relatively high rate of integration appears to
be a consequence of naturally occurring DNA repair enzymes present in all eukaryotic cells. During evolution,
these enzymes acquired the ability to seek out and ligate together open-ended DNA molecules, which can result naturally from mutagenesis. Figure E.7 illustrates the details of
the transgenic procedure.
Up to 50% of the mice born from injected embryos
have the foreign DNA stably integrated into their genomes.
They will thus transmit this DNA to their offspring. There
are no limits to the type of DNA that can be incorporated.
It can come from any natural sourceanimal, plant, or
microbialor directly from a DNA synthesizer. It is very
common for investigators to construct DNA molecules
(called DNA constructs) composed of genetic elements
from different sources. For example, a DNA construct
might have a coding region that is a composite of human
and Escherichia coli sequences flanked by an upstream
regulatory region that is a composite of mouse and synthetic
sequences.
Although embryonic nuclear injection is a powerful
transgenic tool, it has two significant limitations. First, it
can only addnot subtractgenetic material. Second,
experimenters cannot target the insertion of foreign DNA
to specific genomic locations. Consequently, transgenic
mice produced by embryonic nuclear injection are useful
only for the analysis of dominant phenotypes. By 1989,
geneticists had developed a way to circumvent these
limitations.
har06584_refE_109-132
116
11/06/2006
06:11 PM
Page 116
Several embryos recovered from sacrificed female
Embryos transferred to a depression slide containing culture medium

Culture medium
Oil
As embryo is held in place, DNA

is injected into pronucleus.
Holding pipette
Pronucleus
DNA to be injected
Injection pipette
Several injected embryos are placed into oviduct of receptive female.
Figure E.7 How transgenic mice are created. About 12

hours after conception, the female mouse is sacrificed and the onecell embryos recovered. They are transferred to a depression on a
specialized microscope slide containing a drop of culture medium
under oil to prevent evaporation. The slide is placed on the stage of
an inverted microscope (as the name implies, the objective lens is
beneath the stage rather than above it, as in a typical microscope).
This arrangement gives the researcher space to manipulate the embryos from above. (In the photo used here, only one of the two
pronuclei is visible.) Suction holds the embryo in place on the blunt
end of a special holding pipette. A second type of pipette with a
very narrow bore (the injection pipette) is used to inject transgenic
DNA through the plasma membrane and into the pronucleus,
where the foreign DNA is released. The altered embryos are then
placed into the oviduct of a physiologically receptive female.
Targeted Mutagenesis, a Second

Transgenic Technology, Makes It
Possible to Remove, or Knock out, Specific
Sequences from the Mouse Genome
In addition to knocking out the function of a sequence altogether, targeted mutagenesis can produce an allele with an
altered function. The emergence of targeted mutagenesis,
which is technically more demanding and more complex
than the nuclear injection technology just described, depended on two advances in cell culture techniques that
occurred during the 1980s.
The first advance was the establishment of in vitro conditions that enable researchers to place mouse embryos at
the blastocyst stage into culture such that the embryonic

cells from the ICM continue to divide without differentiating. Cultured cells that behave in this way are called
embryonic stem cells, or ES cells for short. ES cells appear to be similar in their state of differentiation to cells
from the ICM. It is possible to grow cultures containing
many millions of ES cells from a single embryo and then
recover a handful of cells from this culture for injection
back into the blastocoel cavity of a normal embryo. Once
inside the cavity, the ES cells can become incorporated
into the ICM, and they can contribute to all of the tissues
in the mouse that develops from the embryo. Most importantly for geneticists, the ES cells even contribute to the
germ lines of these chimeric mice so that reproducing
adults can transmit mutated genes present in the ES cells
to future generations.
The second critical advance that provided a foundation
for targeted mutagenesis was development of a protocol for
homologous recombination in ES cells. The transformation
of mammalian cells is known as transfection. During the
transfection of mouse cells with mouse-derived DNA, the
foreign mouse DNA almost always integrates at random
into a chromosome at a site other than its point of origin.
Occasionally, however, the added DNA will find and replace its homolog by homologous recombination. The frequency of homologous recombination events as a fraction
of the total number of integrations is on the order of
103105.
If researchers transfect mouse ES cells with unaltered,
cloned fragments of mouse DNA, homologous recombination events do not cause genomic changes. But with the
recombinant DNA technology described in Chapter 9 of
the main textbook, investigators can modify cloned genes
so that they no longer function; cloned genes modified in
this way are known as knockout constructs. When
homologous recombination occurs with a knockout
construct, the nonfunctional knockout allele replaces the
endogenous wild-type allele (Fig. E.8). To construct mice
in which homologous recombination has knocked out
specific genes, researchers developed protocols for identifying and recovering the very rare ES cells in which
homologous recombination occurs.
E.2
How Biologists Use

Transgenic Tools to Study
Mice and Create a Mouse
Model for Human Disease
Both add-on and knockout transgenic technologies have

tremendous value in genetic research. The protocols
enable researchers to determine the function of gene
har06584_refE_109-132
11/06/2006
06:11 PM
Page 117
E.2 How Biologists Use Transgenic Tools to Study Mice and Create a Mouse Model for Human Disease
(a) Construction of a knockout allele in ES cells

Early blastocyst
( 10,000)
Clone containing gene of interest

5'
117
Finding the cell with the knockout allele.

Subject culture to drug that kills all cells that do not contain
selectable marker.
3'
Build knockout construct by
adding in selectable marker.
5'
Culture into millions
of embryonic-like
(ES) cells.
3'
Marker disrupts transcription
unit.
Survivor cells have knockout allele (1% or less).

Begin new culture with survivor cells.
Add cloned DNA to

culture of cells.
(b)
Homologous recombination inside ES cell nucleus

Knockout construct
5'
3'
ES cell chromosome with wild-type allele
ES cell chromosome with knockout allele
Figure E.8
Knocking out a mouse gene in ES cells. (a) An early mouse blastocyst can be grown in culture under conditions that allow the cells of the ICM (inner cell mass) to remain undifferentiated as ES cells. A DNA clone, containing the gene of interest, can be
modified in the laboratory into a disrupted allele with a selectable marker. This knockout construct is added to the ES cell culture,
where homologous recombination will occur. (b) A chimeric mouse composed of cells derived from a normal embryo (albino) and ones
derived from the mutated ES cell (dark agouti).
products, characterize genetic regulatory regions, establish links between mutant phenotypes and particular
transcriptional units as an aid to verifying the identification of a cloned gene, and create mouse models of human
genetic diseases.
Two one-cell female mouse embryos (with two X chromosomes)
Pronuclei
Inject SRY DNA
No injection
Using Transgenic Technology

to Determine Gene Function
In Chapter 11 of the main textbook we saw how geneticists
used transgenic technology to demonstrate that the cloned
SRY (for sex-determining region on the Y chromosome)
gene could confer maleness on an animal without a
Y chromosome. Incorporation of the SRY genes coding
region and regulatory sequences into a mouse embryo
with two X chromosomes produced an animal with male
genitalia and testes. This result demonstrated that the
product of a single gene on the Y chromosome is all that is
needed to switch the developmental pathway of the fetus
from female to male (Fig. E. 9).
Biologists have used this same transgenic technology to
examine the functions of many other genes. By combining
19 pairs autosomes,
two X chromosomes
FEMALE
19 pairs autosomes,
two X chromosomes,
and SRY transgene
MALE
Figure E.9
The transgenic mouse protocol proves that SRY

is the testis-determining locus responsible for the production of maleness during embryogenesis. DNA fragments con-
taining the mouse SRY gene and its regulatory sequence were
injected into a series of embryos that were allowed to develop
into live animals. Normal XX animals without a transgene develop as females. But the presence of the SRY transgene in an XX
embryo induces development as a male.
har06584_refE_109-132
118
11/06/2006
06:11 PM
Page 118
a mouse gene of interest with regulatory regions from other

mouse genes, they can cause transgenic mice to express the
natural transgene product in an unnatural manner: at a
higher than normal level, in an alternative tissue, or at an alternative developmental stage. They can then use the aberrant level, time, or place of expression to elucidate the
normal function of the wild-type gene.
Experiments analyzing the myc gene demonstrate the
power of transgenic technology for uncovering the function of genes in ectopic expression (that is, expression at
an abnormal place, time, or level). Investigators originally
discovered the myc gene in the genome of the myelocytomatosis chicken retrovirus; exposure to this virus
caused cultured chicken cells to become tumorigenic. Hybridization studies demonstrated the existence of a myc
gene homolog in the genomes of mice and other vertebrates, including humans (where it resides on the long
arm of chromosome 8) but not in the genomes of nonvertebrate model organisms, such as Drosophila. Moreover,
in human cancer cells obtained from patients with
Burkitts lymphoma (a cancer of the immune systems B
cells), the myc gene often appears close to one of the
breakpoints of a reciprocal translocation characteristic of
these cancer cells between the long arms of chromosomes
8 and 14; in this translocated position, the gene is usually
expressed at a higher than normal level. Noncancerous
animal cells have a very low level of myc gene expression.
These findings constitute circumstantial evidence that abnormally high levels of myc expression might help transform a cell to a cancerous state. (The biochemical
mechanism by which the myc gene product functions is
described in Chapter 18 of the main textbook.)
The results of experiments using cells grown in culture support this hypothesis. In almost every case studied, however, cultured mammalian cells display
programs of gene expression that do not correspond with
those of the cells they are supposed to model. Thus, the
results of cell culture studies do not necessarily reflect
how cells in vivo (that is, within the body) behave in response to a change in gene activity. In addition to differences in gene activity, there are differences in chromatin
structure and patterns of DNA methylation between cells
growing in vitro and in vivo. These discrepancies are not
surprising since cells in the body, unlike those in culture,
exist in a complex environment that includes constant exposure to molecular signals released by other body cells.
The living organism also has a pervasive immune system
that is impossible to imitate in vitro. It is therefore possible that a phenotype observed in response to the abnormal expression of a gene in cultured cells might be a
consequence not of one genes abnormal expression but
of interactions with other genes that are expressed differently in cultured cells than in cells in vivo. For these reasons, it is not possible to rely on cell culture results for
an explanation of the true function of a gene; rather it is
(a)
(1) The myc locus found in the mouse genome.
Promoters
Exons
3'
5'
1 kb
(2) Hybrid DNA construct containing the myc coding region

regulated by an inducible promoter
3'
5'
Inducible
promoter
(b)
Figure E.10 Transgenic expression of the myc gene provides information on the genes role in tumor formation.
(a) Construction of a transgene containing the myc gene under
the control of an inducible promoter: (1) structure of the endogenous myc gene and (2) transgene construct with the MTV
(dexamethasone-inducible) promoter attached to a portion of
the myc gene that contains the coding region. (b) Northern
blot showing induction of transgene expression in a range of
adult tissues.
necessary to examine the effects of aberrant gene expression in vivo.

To learn whether increased expression of the myc
gene affects tumor formation in various tissues of the
mouse, researchers used transgenic technology (Fig. E.10).
In one experiment, they attached the immunoglobulin gene
promoter to the myc coding sequence to produce a transgenic mouse line that expressed myc at high levels in the
precursors to immunoglobulin-producing B cells. In another experiment, they attached tissue-specific promoters, including one for mammary gland expression, to the
myc coding region. And in yet another experiment, they
attached to the myc coding region a promoter that is
recognized in all tissues but only after the embryos or
animals exposure to dexamethasone, a glucocorticoid
hormone.
The results of all these experiments showed that overexpression of the myc gene does not have any effect on normal developmental processes. Even when the gene was
expressed at high levels in many developing tissues, normal animals were born. Moreover, even in the adult, most
cells that overexpress the myc gene never display an aberrant phenotype. However, the rate of tumor formation increased significantly in most, but not all, types of tissue.
The conclusion was that aberrant expression of the myc
gene alone does not cause cells to become cancerous; but
it can operate with other somatic mutational events to
har06584_refE_109-132
11/06/2006
06:11 PM
Page 119
transform a cell to the cancerous state. These observations

support the hypothesis that the cancer phenotype results
from the accumulation of multiple mutations in the clonal
progeny of one cell (see Chapter 19 of the main textbook).
Biochemical and genetic studies indicate that the protein product of the wild-type myc gene is a component of a
transcription factor that helps control the growth of cell
populations in almost every tissue of the body (as described
in Chapter 18 of the main textbook).
E. coli -galactosidase gene
Mouse genomic clone

Coding region
5'
119
3'
5'
3'
Regulatory
region
E. coli
Mouse
Inject transgene construct into one pronucleus of one-cell mouse
embryos. Place embryos into oviduct of receptive female.
Using Transgenic Technology to

Characterize Regulatory Regions
To understand embryonic development, it is essential to
learn not only how the proteins critical to this complex
process function but also how the genes that encode these
proteins are regulated. The pattern of gene expression in a
normally differentiating cell is finely tuned, with each gene
expressed at exactly the right level and each gene turned on
and off at exactly the right times. Geneticists know that ciscontrol elements (that is, regulatory regions on the same
chromosome) that are closely linked to each gene regulate
the timing and intensity of each genes expression. Transgenic technology can help determine the location and
sequence of such regions.
Researchers can characterize the cis-control elements
of single-cell organisms, such as bacteria and yeast, by
random mutagenesis experiments. In these experiments,
they select from cultures of millions of mutagenized cells
those cells with single base changes or small deletions in
genetic regions that influence the cis-regulation of a gene
of interest. They can then use the regulatory mutations to
identify the base pairs necessary for proper gene function.
Because of the large number of organisms required, it is
not possible to use a similar approach to study gene regulation in mice. Instead, mouse geneticists must turn to
transgenics.
Once they have obtained a genomic clone of a mouse
gene of interest, researchers can subclone flanking sequences that are likely to contain its regulatory region. In
most cases studied to date, the regulatory regions have been
confined to the 510 kb of DNA just upstream (to the 5
side) of the coding sequence; thus, this is the first region an
investigator examines at the start of a project. Although it
might seem counterintuitive at first, the researcher has no
use for the actual coding region of a gene when the aim is
to study the genes regulatory region. In fact, it is best to
replace the true coding region with that of a reporter gene
whose protein product is easy to detect. The most commonly used reporter gene is the -galactosidase (lacZ)
gene from E. coli bacteria (Fig. E.11). As described in
Chapter 9 of the main textbook, the -galactosidase
enzyme encoded by this gene will convert a special
Mate transgenic animal to establish pregnancy.

Recover and stain fetuses to detect -gal activity.
Figure E.11 Transgenic technology can be used to analyze cis-acting regulatory regions. A DNA construct containing the mouse regulatory region of interest is attached to the
E. coli reporter gene. The function of the regulatory region can
be ascertained by observing -gal expression in transgene fetuses. In the example shown here, expression is observed in the
developing forelimb (blue) of three independently derived
transgenic mice containing the complete regulatory region of
the mouse Tbx5 gene, which is known to play a specific role in
forelimb development of all vertebrates. John Schiementi,
The Jackson Laboratory. Reproduced from Promoter Mapping
of the Mouse TcP-10bt Gene I Transgenic Mice Identifies Essential Male Germ Cell Regulatory Sequences, Ewulonu et al.,
Molecular Reproduction and Development 43:290297, 1996.
Reproduced by permission of Wiley-Liss, Inc., a subsidiary of
John Wiley & Sons, Inc.
substrate known as X-Gal into a colored product that can

be visualized under the microscope.
To characterize the regulatory region associated with
a gene, you first create a series of different transgene constructs, each formed by splicing together different fragments or mutated forms of the putative regulatory region
with the -galactosidase reporter gene. You next establish
a series of transgenic lines by injecting the transgene constructs into different mouse embryos. Then, after setting
up timed matings within each of these lines, you recover
har06584_refE_109-132
06:11 PM
Page 120
(a)
1 1.6 kb
2
3
4
5
6
-galactosidase
1.3 kb
1.16 kb
0.97 kb
0.75 kb
0.6 kb
1.6
H Bsp Bam Sfa X Bam

1.3
(b)
Bg
1.16
R
0.97
embryos at the developmental stage or stages during

which the mouse gene normally undergoes expression. By
examining the distribution of the colored -galactosidase
product at each embryonic stage and comparing it to the
distribution of the natural gene product, it is possible to
map the extent of the cis-acting regulatory region. The distribution of the natural gene product can be tracked with
specific antibodies labeled for examination by immunohistochemistry. If the antibody label is a different color
than the -galactosidase label, the natural gene product
and the -galactosidase enzyme can be simultaneously observed and distinguished under the microscope. Further
studies of the effects of individual base-pair substitutions
or small deletions on details of the expression pattern can
lead to a highly sophisticated understanding of the various
cis-control elements. The collective behavior of these cisacting regulatory elements helps determine the complex
patterns of spatial and temporal expression of genes that
play a role in development.
An example of the use of transgenic mice to study gene
regulation is the analysis of a gene called T complex protein10bt (Tcp10bt). Only differentiating male germ cells in the
process of maturing into spermatozoa express Tcp10bt.
Cells in the seminiferous tubules, called sertoli cells, mediate sperm differentiation. When investigators disrupt the
testes and place sperm cells in culture, they cannot simulate
natural conditions completely; as a result, germ cell differentiation continues in vitro for only a brief time. But differentiation from stem cell to mature spermatozoa takes six
weeks in mice.
Sperm differentiation is of interest to cell biologists
because the mechanisms that regulate it may differ significantly from those that control the differentiation of other
cells. This is, in part, because the transformations of differentiating sperm cells are much more dramatic than those
experienced by other types of cells. Not only do the spermatogenic cells change in shape and size from large round
stem cells to tiny, sleek spermatozoa almost without cytoplasm, they also drastically change their genetic program:
Stem cells have a normal program of gene expression as
well as chromosomes with a normal chromatin structure; in
contrast, the chromosomes of differentiated sperm cells
have a unique chromatin structure with no histones attached, and they exhibit no gene activity. To understand
these differences, mouse geneticists have tried to characterize the regulatory regions associated with gene activity in
spermatogenic cells.
To analyze the regulatory region associated with the
Tcp10bt gene, geneticists first made a series of transgene
constructs carrying different lengths of DNA from the 5
flanking region of the Tcp10bt gene, ligated to the lacZ
coding sequence. The result was six DNA constructs
containing from 0.61.6 kb of DNA from the putative
Tcp10bt regulatory region. The researchers injected
copies of each DNA construct into multiple mouse embryos, obtaining at least four independent transgenic
0.75
120
11/06/2006
Nhel Bam
Figure E.12 An example of the use of transgenic technology to map the cis-acting regulatory region associated with
the Tcp10bt gene. (a) Different hybrid DNA constructs were
made with varying lengths of the 5 flanking sequence adjacent to
the Tcp10bt gene fused to the E. coli -galactosidase gene (which
acts as a reporter). (b) Testicular RNA was obtained from transgenic
mice containing the various lengths of 5 flanking region (shown in
kilobases). With 0.75 kb of flanking region, no transcription of the
reporter gene was observed, but with all larger flanking regions,
transcription did occur. Source E.12b: John Schiementi, The
Jackson Laboratory. Reproduced from Promoter Mapping of the
Mouse TcP-10bt Gene I Transgenic Mice Identifies Essential Male
Germ Cell Regulatory Sequences, Ewulonu et al., Molecular
Reproduction and Development 43: 290297, 1996. Reproduced
by permission of Wiley-Liss, Inc., a subsidiary of John Wiley &
Sons, Inc.
lines of mice for each construct. (It is important to use

multiple transgenic lines in an experiment of this type
to verify that sequences in the transgene construct
itself, rather than sequences that coincidentally flank the
transgene insertion site, are responsible for a particular
phenotype.)
Figure E.12 depicts how it was possible to map the
regulatory region associated with Tcp10bt by simply testing for the presence of lacZ transcripts in Northern blots of
testicular RNA obtained from each transgenic line of mice.
As the figure shows, there was no detectable transcription
of the lacZ gene in testes from transgenic mice that carried
0.75 kb or less of flanking sequence 5 to the Tcp10bt gene;
but with 0.97 kb or more of the flanking sequence, high
levels of transcription occurred in the testes, but in no other
tissue, of all mouse lines.
These observations located a critical testes-specific, cisregulatory sequence within a 227 bp region between 746 and
973 bases upstream of the Tcp10bt gene. With this information, it became possible to design additional experiments to
examine the regulatory region in more detail and identify
har06584_refE_109-132
11/06/2006
06:11 PM
Page 121
transacting proteins that bind to this region. Through these

additional experiments, researchers identified specific testicular proteins that activate the Tcp10bt gene.
Using Transgenic Technology

to Link Mutant Phenotypes
to Specific Transcription Units
A third significant use of transgenic tools is in establishing
whether a cloned gene corresponds to a locus previously
defined by a mutant phenotype. Consider, for example, the
mouse Brachyury locus, symbolized by a T. This locus is
defined genetically by a dominant, X-ray-induced mutation
that causes heterozygous T / animals to develop short
tails. Through high-resolution linkage analysis and positional cloning (described in Chapter 11), researchers
showed that the T locus mutation is associated with a 200 kb
deletion on chromosome 17. They also identified a transcription unit called pme75 that is located in this same 200 kb
deletion; normally, pme75 is expressed in the embryonic
tissue that develops into the tail. While highly suggestive,
these data do not prove that the absence of the pme75 gene
causes the short-tail phenotype. It is very likely that the
200 kb deletion also removes other genes, and on the basis
of the genetic data alone, you cannot rule out the possibility that the absence of one of these undiscovered genes produces the mutant phenotype.
You can resolve this impasse with the transgenic protocol illustrated in Fig. E.13. The first step is to make a
transgene construct containing the complete pme75 coding
region together with its regulatory sequences. You next inject
the construct into wild-type mouse embryos to create a
transgenic line. By breeding this transgenic line to animals
with the T locus mutation, you can obtain offspring that
have the 200 kb deletion over the T locus as well as a functional pme75 gene on a different chromosome (at the locus
where the transgene construct integrated at random). These
animals sport a tail of normal length. By demonstrating
that the transgene can correct the mutant phenotype, this
result proves that the T locus is the equivalent of the pme75
gene.
Using Targeted Mutagenesis to Create

a Mouse Model for Human Disease
Researchers can use targeted mutagenesis, in combination
with other protocols, to create a mouse model for human
diseases that result from a loss of gene function. As an example, Figure E.14 illustrates the step-by-step creation of
a mouse model for cystic fibrosis. With the identification
and cloning of the human cystic fibrosis gene (designated
CFTR), the first step toward a mouse model was the
121
cloning of the mouse homolog. Development of the mouse

model also required fabrication of a CFTR knockout construct, derivation of an ES cell culture from a mouse blastocyst, transfection of the ES cells with the CFTR knockout
construct, selection of cells in which homologous recombination had replaced the wild-type CFTR gene with the
mutant knockout allele, and finally, the production and
analysis of chimeric mice and their offspring. Animals
homozygous for the CFTR knockout allele display a mutant
phenotype that is very similar to that expressed by humans
suffering from cystic fibrosis. Thus, in developing drugs
to alleviate CF symptoms in humans, pharmaceutical
researchers can first test new products in mice to determine
their efficacy.
Transgene construct
contains pme75 gene with
its regulatory sequences
Inject into pronucleus of wildtype one-cell mouse

embryos
Create transgenic line
Transgenic animal with

Tg (pme75)
Normal phenotype
T/+ genotype
Short-tail phenotype
Normal tail
Offspring: T /+ genotype
Tg (pme75)
Normal phenotype
Figure E.13 Transgenic technology can be used to identify the locus responsible for a mutant phenotype. A dominant deletion mutation at the T locus causes a short tail. A
transgenic animal containing the pme75 transgene is mated
with a mutant animal to create animals containing both the
deletion and the transgene. A normal phenotype demonstrates
that the deletion of the pme75 gene is responsible for the
short-tail phenotype.
har06584_refE_109-132
122
11/06/2006
06:11 PM
Page 122
(c) Early blastocyst recovered from

mating between two agouti parents
of the 129/ SvJ strain
(a) Plasmid clone containing portion of

mouse CFTR locus with first exon
3'
exon 1
(b) Develop DNA construct
by adding selectable marker (neo)
and TK gene to CFTR restriction
fragments
5'
neo
TK
exon 1
neo
Develop ES cell culture by

placing blastocysts in petri
dish to undergo many cell
divisions without differentiation
TK
(d) Add cloned DNA knockout

construct to culture
of ES cells
ES culture
HOMOLOGOUS RECOMBINATION
neo
DNA construct can

integrate through
two different mechanisms
INTEGRATION INTO RANDOM LOCUS
TK
neo
TK
neo
TK
exon 1
endogeneous
random locus
disrupted CFTR exon 1

neo
Expose ES culture to neomycin.
Remaining cells contain CFTR construct
integrated either randomly or homologously
Expose colonies to
ganciclover. TK-containing
cells eliminated
Transfer remaining
colony to plate.
Begin new culture.
(e) Cell contains disrupted copy of CFTR exon 1
Figure E.14
Creating a mouse model for cystic fibrosis.
har06584_refE_109-132
11/06/2006
06:11 PM
Page 123
(f ) Mate B6 black mice. Embryos recovered

from pregnant B6 female.
(+/+)
(+/+)
10 ES cells are placed in embryos which

are returned to uterus of B6 foster mother.
Colony of ES cells heterozygous

for a knockout of the CFTR locus (+/)
Embryos develop into live-born mice
(+/+)
Chimera
[agouti (+/)] and [black (+/+)]
Mate chimera with B6 black mouse
[agouti (+/)] and [black (+/+)]
black (+/+)
(g)
Three types of offspring

agouti (+/)
(h)
agouti (+/+)
black (+/+)
Use DNA analysis to

identify male and female agouti animals that are heterozygous
for the knockout allele of CFTR (+/) and breed them toghether
(+/)
(+/)
Offspring homozygous
for mutant allele serve
as models for CF disease
state.
(+/+)
(+/)
(+/)
(/)
Use DNA analysis to identify

offspring homozygous for
knockout allele to serve as
models for cystic fibrosis
disease state
123
har06584_refE_109-132
124
11/06/2006
06:11 PM
Page 124
Embryonic axis
Anterior
Antennapedia locus
Drosophila
lab
pb
(Zan)
Scr
Dfd
Posterior
Bithorax locus
Antp
Ubx
Abd-A
Abd-B
Drosophila homeotic selector gene
3'
5'
Mouse
Hox A, chromosome 6
A1
A2
A3
A4
A5
A6
A7
B4
B5
B6
B7
C4
C5
C6
A9
A10
A11
A13
Hox B, chromosome 11
B1
B2
B3
B8
B9
C8
C9
C10
C11
C12
C13
D8
D9
D10
D11
D12
D13
Hox C, chromosome 15
Hox D, chromosome 2
D1
D3
D4
Direction of transcription of mouse genes
Figure E.15 The mouse Hox gene superfamily contains multiple homologs of each member of the Drosophila homeotic
selector gene family. Genes within the four mouse Hox clusters are lined up according to their homology with each other and specific
Drosophila genes. Not all clusters have homologs of each Drosophila gene. From gene 9 and higher, there are multiple homologs of the
Drosophila Abd-B gene in Hox clusters A, C, and D.
E.3
The Hox Genes: A

Comprehensive Example
A particularly striking example of how mouse biologists

used both nuclear injection and targeted mutagenesis technologies to decipher gene function comes from an analysis
of the mouse Hox gene family. The Hox genes are
homeotic selector genes, that is, genes that control the development of body segment characteristics. Members of
the Hox family are distributed among four unlinked gene
clusters that each contain 911 genes (Fig. E.15).
Researchers discovered the mouse Hox gene family
through cross-hybridization studies using the homeotic
selector genes of D. melanogaster as probes. In fact,
homeotic selector genes were first identified on the basis
of mutations that produced flies with four wings instead
of two, flies whose mouthparts developed incorrectly as
legs, and flies with legs instead of antenna growing out of
their heads. The proteins encoded by these genes turned
out to be transcription factors that act as on/off switches,
instructing segments of the fly to develop into one type of
tissue or another. William Bateson, the same man who
coined the term genetics, chose the designation of
homeotic selector from the Greek word homoios, which
describes a type of variation in which something has
been changed into something else. Homeotic genes appear to control the development of each Drosophila body
segment (as discussed in the Drosophila portrait, Reference D on our website). The bizarre phenotypes just
described result when expression of a particular homeotic
gene does not occur at the proper time and place. Lack of
appropriate expression flicks the binary switch, transforming the recipient body segment into a different type
of tissue.
Drosophila homeotic genes were first cloned in the
early 1980s. By the end of that decade, it had become
clear from cross-hybridization and cloning studies that
homologs of these genes are likely to exist in every
species of multicellular animal, from C. elegans to Homo
sapiens.
In segmented animals such as flies, homeotic genes are
active in the discrete segments that define the body plan,
where they determine the proper differentiation of tissues.
But what do they do in mice and humans, organisms that do
not have obvious body segments? To answer this question,
researchers had to overcome a serious drawback: the lack
of known mutations at any of the Hox loci. This problem
was not unique to understanding the functions of mammalian Hox genes. Since the discovery of homeotic gene
homologs in the mouse genome, it has become routine for
developmental geneticists to use cross-hybridization protocols to look for mouse homologs of every Drosophila
gene found to have a role in development. This strategy
has led to the discovery of dozens of new mouse genes,
most of which were not associated with any known mutant
phenotypes.
har06584_refE_109-132
11/06/2006
06:11 PM
Page 125
125
E.3 The Hox Genes: A Comprehensive Example
Posterior
Anterior
Caudal
Sacral
Lumbar
Thoracic
Cervical
Occipital
Vertebrae
atlas
axis
10
11
12
13
Extent of expression
3' Genes
HoxA1
HoxB1
HoxA3
HoxD4
HoxA4
HoxB4
HoxA5
HoxB5
HoxA6
HoxA7
HoxB9
HoxB7
HoxC9
HoxD8
HoxD9
HoxD10
HoxD11
HoxD12
5'
Figure E.16
HoxD13
Spatial extent of expression of some representative Hox genes along the developing spine. The top of the
mature spine is shown to the right, bottom to the left, with vertebrae numbered and grouped by name.
How Scientists Determine the

Function of a Gene in the Absence of
Previously Characterized Mutations
Analyses of Expression Patterns in Developing
Embryos Can Provide a Clue to the Time and
Location of Gene Action
A clone of a gene can be a tool for analyzing the genes expression. One way to convert a clone into this type of tool
is to label it, denature it, and use the resulting DNA strands
as probes in in situ hybridization. To study development,
investigators can perform in situ hybridization on the RNA
present in fixed tissue sections obtained from embryos at
different stages of development. When developmental geneticists examined the expression patterns of the Hox
genes, they discovered that each one is transcribed along a
portion of the developing embryonic axis that extends from
the same most posterior point to a specific anterior boundary (Fig. E.16). Analysis of the pooled data on Hox gene
expression showed that the anterior boundary of expression
corresponds with the position of each gene in its cluster.
Genes at the 5 end of a cluster (for example, D13) have the
least extensive expression which is restricted to the posterior region of the embryonic axis. In contrast, genes at the
3 end of the cluster (for example, A1 and B1) have a
broader range of expression that extends further to the anterior region of the embryonic axis. These data suggest that
different Hox genes might be involved in controlling the
development of different sections of the embryonic axis.
Expression data alone, however, cannot provide conclusive
evidence of function.
Ultimately, Only Genetic Tests

Can Determine Gene Function
To understand what role a particular Hox gene plays in
development, a scientist must be able to examine embryos that do not express that gene at all, or express it
outside its normal time or place. Examining the changes
in development that arise as a result of these genetic
changes makes it possible to decipher the normal role of
the gene and, in the case of the Hox family, test general
hypotheses concerning the functional interactions of different Hox genes. We now present two examples of this
approach.
har06584_refE_109-132
126
11/06/2006
06:11 PM
Page 126
HoxA1 regulatory region
HoxD4 coding sequence
(a)
Validating the Hypothesis That Expression

of the 5 Gene in a Hox Cluster Is Epistatic
to Expression of the More 3 Genes
The expression data show that some Hox genes are expressed
across many segments of the embryo, even as those segments
develop differently from each other. To account for this difference in the simplest way, researchers proposed that the
only gene that counts in any particular embryonic segment is
the most 5 gene in a Hox cluster. In other words, expression
of the most 5 gene is epistatic to expression of the other,
more 3 Hox genes. By examining Figures E.15 and E.16,
you can see that this hypothesis could explain how each
spinal segment develops in a different manner.
A Transgene Test Confirmed the Prediction

of a Homeotic Transformation
(b)
As one test of this hypothesis, investigators made a transgene construct with a HoxA1 regulatory region attached to
a HoxD4 coding sequence (Fig. E.17a). According to the
hypothesis, HoxD4 normally controls the development of
the C1 and C2 vertebrae in the cervical region of the spinal
column, while HoxA1 normally controls the development
of the occipital bone at the base of the skull. In a transgenic
fetus, however, the presence of the transgene construct
causes expression of the HoxD4 gene in the occipital region along with HoxA1; and as predicted by the hypothesis,
a homeotic transformation converts the occipital bone into
cervical vertebrae (Fig. E.17b and c).
Knockout Studies Confirmed Predictions

of Aberrant Phenotypes
(c)
Figure E.17 Transgenic technology provides support for

the mode of action of the Hox gene family. (a) A transgenic
construct is produced to missexpress HoxD4 in a more anterior
region where HoxA1 is normally expressed (refer to Fig. E.16 for
the normal extents of expression of each of these genes). In panel
(b), the complete skeleton of a wild-type animal is shown on the
left, and that of an animal expressing the transgene construct is
shown on the right. A blowup of the cervical regions from both
skeletons is shown in panel (c), again the wild type is on the left
and the transgenic construct is on the right. In transgenic newborns, what would have been occipital bone (region E in wildtype animal) has been transformed into ectopic arches (E1) that
look like cervical vertebrae.
The hypothesis of 5 epistasis also leads to the prediction

that aberrant phenotypes will arise when different Hox genes
are knocked out by homologous recombination. With each
knockout, one would expect a particular embryonic segment
to become transformed to a more anterior-like structure. The
data obtained from knocking out various Hox genes support
the hypothesis of 5 epistasis. For example, a knockout of
HoxB4 produces a partial homeotic transformation of the
second cervical vertebra from axis to atlas (see Fig. E.16).
Transgenic Studies Lead to an

Understanding of the Developmental
Role of Hox and Other Homeotic Genes
With the accumulation of transgene and knockout data for
many of the mouse Hox genes, a general understanding of
the developmental role played by this gene family, not only
in mice but in other animals as well, has emerged. What the
Hox genes and their homologs in other species apparently
do is establish signals identifying the position of each region along the embryos anterior-posterior axis. The genes
har06584_refE_109-132
11/06/2006
06:11 PM
Page 127
Solved Problems
do not determine the differentiation of any particular cell

type or tissue. Rather, they provide positional information
that other genes act on to promote the differentiation of
particular tissues. The positional information as well as the
genes that act on it vary from species to species.
It is likely that the emergence of the Hox-like gene family in our most recent single-cell ancestor set the stage for the
evolution of developmental complexity, with the consequent
127
appearance of metazoan organisms sometime between 1 billion and 600 million years before the present time.
Thus, the analysis of the Hox gene family is an example
of how detailed genetic studies in one model species can provide a general understanding of gene function across large
segments of the animal kingdom as well as clues to the conserved mechanisms by which complex developmental
processes are carried out.
Connections
Geneticists can now produce transgenic animals by combining the classic tools of mutagenesis with an understanding of molecular biology and embryology. Transgenic
technology has become so sophisticated that, in theory, it is
possible to make any genetic change imaginable to the
mouse genome and determine its effect on the individual
that emerges.
With the ability to produce mice carrying add-ons and
knockouts of Hox and other genes that play a role in devel-
opment, mouse geneticists have begun to dissect the

process by which the mouse embryo develops from the
one-cell zygote, through gastrulation, and into the organbuilding stage. Remarkably, the development of all mammals is similar, especially in these early stages. For
example, all the genes important to mouse development are
conserved in the human genome. Thus, much of what we
learn about the genetic basis for mouse development will
apply to normal and abnormal human development.
Essential Concepts
1. The availability of hundreds of single-gene mutations
and a short life cycle contribute to M. musculuss
value as a model organism.
tion to determine gene function, characterize regulatory regions, and correlate mutant phenotypes with
specific transcription units.
2. The mouse genome closely resembles the human

genome in size, gene content, and syntenic loci. Researchers can thus use homology and conserved synteny analysis to identify, locate, and determine the
function of genes in both species.
5. Targeted mutagenesis is the basis for creating a

mouse model of human diseases caused by a loss of
gene function.
3. The mouse life cycle is representative of the mammalian life cycle, although the timing of events is
unique to each species. The totipotency of preimplantation cells in mammals makes it possible to create
chimeras.
4. Researchers can use the transgenic technology of
adding genes to the mouse genome by nuclear injec-
6. Studies using transgenic technology have revealed

that the Hox genes generate signals that identify the
position of each region along a mouse embryos
anterior-posterior axis. Normal development of the
embryo depends on this information.
7. Knowledge of how the Hox genes function in mice
has elucidated the general role played by homeotic
selector genes in the evolution of all metazoan organisms, including humans.
Solved Problems
I. Gain-of-function mutations can produce a novel phe-
notype and act in a dominant fashion. In loss-offunction mutations no functional gene product is
made; most, but not all, loss-of-function mutations
are recessive.
a.
b.
Which of these types of mutations would you

study using add-on transgenic technology?
Which type of mutation would you have to study
with the implementation of homologous recombination in ES cells? Why?
har06584_refE_109-132
128
11/06/2006
06:11 PM
Page 128
BamHI
Answer
This problem requires an understanding of add-on
transgenic and homologous recombination techniques
and their applications.
a. Dominant alleles caused by gain-of-function mutations are expressed phenotypically when a single
copy of the dominant allele is present. The effect of a
dominant gain-of-function mutation can be characterized in transgenic mice where a transgene containing the dominant allele has been inserted into the
genome and the original gene copy is still present.
b. Loss-of-function mutations would have to be studied using homologous recombination, where a
normal copy of the gene is replaced with a defective version. To observe a phenotype of a recessive
loss-of-function mutation, there should not be any
normal version of the gene present. After replacing one gene copy with a mutant version, heterozygous mice would be mated to obtain a mouse
homozygous for the recessive allele and to observe
the resulting phenotype.
II. You have isolated muscle cell RNA and made a muscle
cDNA library. You want to study the regulatory region

of one of the genes you identified as being expressed in
muscle cells.
a. You need to isolate a genomic clone of this gene.
Why is the cDNA clone not sufficient for your
studies?
b. Outline the steps to isolate the genomic clone of
this gene.
c. You isolated a genomic clone containing the entire
coding region and 1.2 kb upstream of the coding
region. A map of the genomic clone is shown here.
To make a fusion construct of the presumed regulatory region attached to the reporter gene lacZ,
you cloned the 1.3 kb BamHI fragment into a vector containing the lacZ gene. The fusion construct,
when injected into cultured muscle cells, expressed -galactosidase protein. What experiment
would you do next to determine if the fusion construct is developmentally regulated?
d. Using an antibody against the -galactosidase
portion of the fusion protein you analyzed the tissues of a transgenic mouse and found that the protein encoded by this gene is also present in bone
cells. How could you determine if the same transcript was expressed in both cell types?
e. You decided to analyze the promoter region of
the gene to identify sequences responsible for the
expression in the two different cell types. The results obtained with constructs containing different fragments from the 5 region flanking the
gene to the lacZ coding region are shown here.
What region(s) are important for expression in
muscle? in bone?
bp
BamHI
300
400
S
DNA in subclone
1.
2.
3.
4.
5.
6.
7.
50
S
350
Coding region
100
S
Expression
in bone
+
+
+
S = Sau 3A
Expression
in muscle
+
+
+
Answer
This problem requires familiarity with several molecular techniques including analysis of cDNAs and regulatory sequences in the promoter.
a. The cDNA clone was derived from the mRNA and
therefore it will probably not contain the regulatory region usually found upstream of 15 to) the
transcription start site.
b. The cDNA clone is a good source of a probe for
finding the genomic clone in a genomic library.
1. Isolate a DNA fragment from the cDNA clone
and label for use as a probe.
2. Grow clones (bacteriophages or cosmids)
from a genomic library on plates.
3. Transfer clones from plates to filter paper.
4. Incubate the labeled probe with the filters to
find, through hybridization, a genomic clone
containing the gene.
c. To study developmental expression of this gene,
the fusion construct would be transferred into the
mouse genome, using transgenic techniques. At
different stages in development the tissues of the
mouse can be tested for the presence of the
mRNA or protein using the lacZ DNA or antibody that recognizes the -galactosidase protein
respectively.
d. To examine the transcripts in these two cell types,
you would prepare mRNA from isolated muscle
and bone cells and use the cDNA clone to probe
separated RNAs that have been transferred to a filter (Northern hybridization analysis). Transcripts may be different sizes because of different
transcription start sites or different processing
(splicing) events.
e. The presence or absence of bone or muscle transcripts in mice that carry different transgene constructs provides a way to identify DNA
fragments containing tissue-specific regulatory
sequences. There are three constructs (4, 5, and
6) that express the transgene in bone tissue. The
har06584_refE_109-132
11/06/2006
06:11 PM
Page 129
Problems
only portion of the regulatory region that they all

have in common is a 50 bp Sau3A fragment. The
bone-specific regulatory element must be within
this fragment. For muscle expression, there are
three clones (5, 6, and 7) that express the transgene in muscle tissue and these share the adja-
129
cent 350 bp Sau3A fragment. This analysis shows

that expression of the same gene is subject to different mechanisms of regulation in two tissue
types that use different regulatory sites in the 5
region flanking the gene.
Problems
E-1 Choose the matching phrase in the right column for
each of the terms in the left column.
E-4 A mouse gene was localized to a region of chromo-
some 4. From the synteny map, it appears this gene

would localize to chromosome 8 in humans.
a. How could you determine if the gene is in this region of human chromosome 8? (Do not use the approach of cloning the gene here.)
b. Briefly outline how you could obtain the human
clone containing the homolog of this mouse
gene.
a. stem cells
1. cells destined to become the fetus
b. conserved synteny
2. animals derived from two or more

genetically different embryonic cells
c. inner cell mass
3. undifferentiated cells that serve as a source

of two or more types of differentiated cells
d. trophectoderm
4. experimental elimination of gene function
e. embryonic stem cells
5. the same genes are genetically linked in

two different species
f. chimeras
6. cells destined to become extraembryonic

tissue such as the placenta
g. knockout
7. undifferentiated cells isolated from the

blastocyst that are able to be reconstituted
with normal cells in an embryo and differentiate normally into every tissue type
E-6 How are different coat color alleles used in the proto-
E-2 The mouse genome (3000 Mb) is 30 times larger than
pressed for several days during development of the

embryonic heart and pituitary in mice. Expression
continues in the pituitary in the adult and also is seen
in the germ cells of the adult testis.
a. You bred a male mouse carrying one normal CTF
allele () and one disrupted CTF allele () with a
female heterozygous for the same alleles and examined 30 of their offspring. (The disrupted allele
contains an insertion within the gene.) Twentyone of the offspring were / and 9 were / .
What would you conclude about the CTF gene in
this case?
b. Suppose instead that you obtained 30 offspring
and 7 of the mice that had the ( / ) genotype
were half the normal size. Of these, all 4 males
were sterile, whereas the 3 females were fertile.
What would you conclude about the CTF gene
from these results?
c. In the preceding experiments, what technique(s)
were used to determine if the offspring mice were
( / ), ( / ), or ( / )?
the C. elegans genome (~100 Mb).

a. What challenges are there for a geneticist studying
an organism that has a large genome size?
b. Despite the large genome size, many geneticists
choose to use the mouse as a model system?
Why?
E-3 The CFTR gene, which is defective in humans with
cystic fibrosis, encodes a membrane protein that acts

as a channel for the passage of Cl. Although one particular mutation (a deletion called D508) is predominant in Caucasians, more than 200 different mutations
in the gene have been identified that cause the disease.
Many Cl channel genes, including CFTR homologs,
have been identified in other organisms such as C. elegans and yeast. For each of the following research
questions, indicate whether you would be more likely
to pursue an answer using yeast or mouse and why you
would choose that organism.
a. Do mutations in different regions of the gene have
the same effect on Cl channel function?
b. Do mutations in different regions of the gene affect channel function in all organs normally involved in cystic fibrosis?
c. What portion of the protein receives a signal to
open the channel?
d. Are there drugs that can cause an opening of the
channel?
E-5 Why do identical twins occur in mice (and other mam-
mals) but not in Drosophila?

col for making a gene knockout in mice?
E-7 You have discovered a gene, called CTF, that is ex-
E-8 Retinoic acid, which acts on cells via a protein recep-
tor, is thought to be important for limb development in

mammals.
a. Several different strains of mice that have recessive
defects in limb development are available. You have
tested each of these strains for the presence of the
retinoic acid receptor (RAR protein), RAR mRNA,
har06584_refE_109-132
130
11/06/2006
06:11 PM
Page 130
and the RAR gene using Western (protein), Northern (RNA), and Southern (DNA) blots, respectively.
Based on the following data, give a reasonable hypothesis regarding the defect in each mutant.
Strain
RAR protein
RAR mRNA
RAR gene
A
Normal*
Normal
Normal
B
Absent
Normal
Normal
C
Absent
Short
Normal
D
Absent
Short
Altered
E
Absent
Absent
Altered
F
Absent
Absent
Absent
*Normal means that the protein, RNA, and DNA bands were
normal in size and abundance; normal does not refer to function of the protein.
One of the bands hybridizing with the RAR cDNA clone migrated to a different position in strains D and E.
b.
c.
The limb deformity in strain A could be due to

a defect in some gene other than RAR. Suggest a
simple genetic experiment that would enable you
to test this possibility.
If you wanted to test the developmental consequences of RAR gene expression in neurons,
which do not normally make RAR, how would
you do this?
E-10 Several different mouse mutants have been identi-
fied that have an obese phenotype. The Ob gene, defective in one class of these mutants, was the first
gene involved in obesity to be cloned. The Ob gene
product is made in fat cells and is transported to the
brain, where it informs the brain that the animal is
satiated (full).
a. You made an antibody to the Ob protein and
used this to test predictions of the hypothesis
that Ob is a satiety factor. You isolated protein
from mice that had been eating normally, from
mice that had been starved, and from mice that
had been force-fed a high-calorie diet. You did a
Western analysis using your antibody as a probe
against proteins from these animals. Results are
shown here. Are these results consistent with the
hypothesis for the role of the Ob protein? Why
or why not?
Force - fed
Starved
Normal
E-9 When DNA is injected into the fertilized mouse egg,
the DNA can insert at random in any of the chromosomes. Subsequent matings produce animals homozygous for the transgene insertion. Sometimes an
interesting mutant phenotype is generated by the insertion event. In one case, after injection of DNA containing the mouse mammary tumor virus (MMTV)
promoter fused to the c-myc gene, investigators identified a recessive mutation that causes limb deformity. In
this mouse, the distal bones were reduced and fused together; the mutation also caused kidney malfunction.
a. The mutant phenotype could be due to insertion of
the transgene in a particular region of the chromosome or a chance point mutation that arose in the
mouse. How could you distinguish between these
two possibilities?
b. The mutation in this example was in fact caused
by insertion of the transgene. How could you use
this transgene insertion as a tag for cloning?
c. The insertion mutation was mapped to chromosome 2 of mice in a region where a mutation
called limb deformity (ld) had previously been
identified. Mice carrying this mutation are available from a major mouse research laboratory.
How could you tell if the ld mutation was in the
same gene as the transgenic insertion mutation?
d. Analysis of transcripts from the ld gene showed
that many different transcripts (formed by alternate
splicing) were present in both the embryo and
adult. Is this consistent with a role of the ld gene
product in limb development in the embryo?
How could you determine if the Ob gene is transcriptionally regulated?

c. If the amount of mRNA was the same for all three
types of mice, what type of regulation is involved?
d. You isolated RNA from several different Ob
mutants and analyzed the RNA using Northern
analysis. What conclusions could you reach
about each mutation based on the results shown
here?
b.
Wild type
har06584_refE_109-132
11/06/2006
06:11 PM
Page 131
Problems
e.
You have now decided to clone the receptor for

the Ob protein and express the gene in mammalian cells. Which is the correct alternative for
each of the following steps?
1. (a) Isolate brain mRNA.
(b) Isolate fat mRNA.
2. (a) Clone cDNA into a plasmid vector next
to a mouse metallothionein promoter.
(b) Clone cDNA into a plasmid vector next
to the yeast LEU2 promoter.
131
3. (a) Treat cells with zinc.

(b) Deprive cells of leucine.
f. You identified the Ob receptor clone using the
screen outlined in part e and decided to make a
knockout of the gene in mice. What phenotype
would you predict for mice that lack the Ob receptor? (obese, slim, normal)
g. If you inject Ob protein into Ob mutant mice, what
effect do you predict there will be on the phenotype?
har06584_refE_109-132
11/06/2006
06:11 PM
Page 132

Mouse Genetic Portrait

Cargado por

Información del documento

Derechos de autor

Formatos disponibles

Compartir este documento

Compartir o incrustar documentos

Opciones para compartir

¿Le pareció útil este documento?

¿Este contenido es inapropiado?

Copyright:

Formatos disponibles

Mouse Genetic Portrait

Cargado por

Copyright:

Formatos disponibles

har06584_refE_109-132

Mus musculus: Genetic

A member of the 129 strain of

Reference E Mus musculus: Genetic Portrait of the House Mouse

The mouse is a model system for human

Hirschsprung disease: A human developmental

model of the disease with a classical mutation (piebald coat).

E.1 An Overview of Mus musculus in the Laboratory

Comparison of Mice and Humans

77,000 g (170 lb)

Haploid gene number

Average, 38 weeks (8.9 months)

Average, 624728 weeks (1214 years)

The Mouse Genome

during the 75 million years that mice and humans have

Reference E Mus musculus: Genetic Portrait of the House Mouse

The mammalian life cycle. The life cycles of males

Conserved synteny: An example with mouse

positions of loci with mapped homologs in the human genome.

human geneticists can now look for linkage between DNA

The Mammalian Life Cycle

Male Germ Cell Development

Female Germ Cell Development

E.1 An Overview of Mus musculus in the Laboratory

From this time on, the female progresses through an

pellucida into the space that surrounds the egg membrane.

Fertilized egg = zygote = 1-cell embryo

Early development of mammals from fertilization to implantation.

Reference E Mus musculus: Genetic Portrait of the House Mouse

For Most of the Preimplantation Stage, the

Early mammalian embryos are highly malleable

correspondence between embryo and adult. (a) Creating identical

Events Restricting the Developmental

E.1 An Overview of Mus musculus in the Laboratory

will eventually take part in the formation of the placenta.

After Implantation, the Placenta Develops,

Two Powerful Transgenic Techniques

The Addition of Genes to the Mouse

Reference E Mus musculus: Genetic Portrait of the House Mouse

Several embryos recovered from sacrificed female

Embryos transferred to a depression slide containing culture medium

As embryo is held in place, DNA

Figure E.7 How transgenic mice are created. About 12

Targeted Mutagenesis, a Second

the blastocyst stage into culture such that the embryonic

How Biologists Use

Both add-on and knockout transgenic technologies have

(a) Construction of a knockout allele in ES cells

Clone containing gene of interest

Finding the cell with the knockout allele.

Survivor cells have knockout allele (1% or less).

Add cloned DNA to

Homologous recombination inside ES cell nucleus

ES cell chromosome with wild-type allele

ES cell chromosome with knockout allele

Two one-cell female mouse embryos (with two X chromosomes)

Using Transgenic Technology

The transgenic mouse protocol proves that SRY

Reference E Mus musculus: Genetic Portrait of the House Mouse

a mouse gene of interest with regulatory regions from other

(2) Hybrid DNA construct containing the myc coding region

necessary to examine the effects of aberrant gene expression in vivo.

transform a cell to the cancerous state. These observations

E. coli -galactosidase gene