Content uploaded by Jennie Dusheck
Author content
All content in this area was uploaded by Jennie Dusheck on Aug 17, 2014
Content may be subject to copyright.
52
I
NATURAL
HISTORY 10/02
The Interpretation
The
^^expression"
of
a genome
is
best
understood as a dialogue
with
an
organism's environment.
That
dialogue, not
the genes alone,
determines
which ant becomes a
queen, which
fish
becomes a male.
of Genes By
Jennie
Dusheck
~W"
^
r "'e sometimes think
of the environ-
% W /
ment
as "out there," a place sepa-
\/\/
rats
from
us, a place
we can enter
and
W W leave
at
wiU. But the environment is,
quite smiply, the context for all of life; it is what
makes us what we are. Plants in dry soil grow
deeper roots than those in wet soil. Turtle eggs be-
come male or female depending on temperature. A
fish may become female in one social environment,
male in another. Genes not only direct, they also
take orders. In a sense, our genes are the means
by
which the environment regulates our development.
Everything about us—from the shape of a toe
to the shape of a protein, from the year we enter
puberty to the amount of stress hormone we re-
lease when another car gets our parking space— is
a
manifestation of an ongoing conversation between
genome and environment. This conversation
started biUions of years ago, when life began, and
goes on every minute of our lives. Yet, strangely,
it's a conversation to which most biologists turned
a deaf ear for decades, starting in the
1940s,
when
the focus of biological research became over-
whelmingly genetic. We've all read or been told
repeatedly
that genes provide a "blueprint" for the
body, that genes "program" development, that we
are "products
of our genes." A 1996 introductory
biology
text used by more than half of all college
biology majors in
the United States asserts:
"An
organism's development
is largely determined
by
the genome of [the fertihzed
egg] and the
organi-
zation of the cytoplasm of the
egg
cell."
No
men-
tion is made of any influences outside the egg.
How did biologists come to snub so
thoroughly
one partner in the developmental
conversation?
The answers lie deep in the political and
scientific
history of biolog)'. Decades before the advent of
genetics in
1900,
biologists sought to
understand
heredity by studying development,
the process by
Leopard
frogs
{Rana
pipiens)
in duckweed.
The common
herbicide
atrazine causes
eggs to develop
in
the testes of
male
tadpoles
of
the species,
making
them
incapable
of
reproducing.
54
NATURAL
HISTORY
10/02
Soviet
biologist
(and
Stalinist
ideologue)
Trofim
Lysenko,
above,
thought
the environment
alone shaped an
organism.
Western
scientists held
a
view just as
extreme: their
experiments with
phenotypically
inflexible
organisms
—
fruit
flies,
below,
among others
—
led them
to
virtually ignore
developmental
plasticity in
all
species.
which
organisms
take shape from seem-
ingly
formless
fertihzed eggs. Indeed,
for
the first
experimental embryolo-
gists,
the most
obvious place to look for
answers to the
mysteries of heredity was
not
deep in the
genome
—
an entity
whose
existence they
barely
sus-
pected—
but within the
environment ot
the embryo.
During the latter
halt
of
the nineteenth
century, biologists
showed, for example,
how different
color morphs of the same
butterfly spe-
cies resulted
from changes in temperature.
Others
examined the effects ot ion
or nutrient levels on de-
velopment or looked at how
environmental factors
such as
temperature could determine sex.
Then, in the early
twentieth century, a conflu-
ence
of discoveries and new
technologies turned the
attention of most biologists to genetics and
physiol-
ogy.
Increasingly, in
the West, biologists saw every
individual as a
self-contained unit whose study
could answer virtually every
biological question.
Developmental
biologists focused their attention on
laboratory experiments
in which the 'role of the
environment was deliberately eliminated.
In the Soviet
Union, however, biologist Trofim
Lysenko believed that
environment determined
phenotype—that is, all of an organism's
observable
attributes,
both structural and functional. As a stu-
dent, Lysenko had been laughed
at
by
geneticists;
once he rose to power, he denounced old acquain-
tances and even
mentors. Under Stalin and
Lysenko, an entire generation of
Soviet geneticists
was exiled or murdered. Those who
survived fled
to
Europe or to North or South America.
Biologists in the West
recoiled violently from
Lysenkoism. Many had lost
personal friends in the
purge or were themselves expatriate Soviets
who
had
fled.
The very idea that the en-
vironment influences phenotype be-
came associated with the worst
aspects
of Stalin's bloodthirsty reign,
with
Communism, and with left-wing
poli-
tics in
general—but not with science.
In
the 1940s and
1950s, a
handful of
Europeans
and Americans attempted
to reintroduce environmental consid-
erations
into developmental biology
but met with little
success. The
molec-
ular
genetics revolution of the
1960s
swept
up many of the brightest young
minds.
Throughout the 1960s and
1970s,
biologists interested in the effects
of envi-
ronment on development, survival, and repro-
duction
worked primarily
in
ecology, agriculture,
conservation
biology,
and related fields.
As
developmental biologists increasingly
focused
on how genes
"determine" phenotype,
they
turned
to just a handful of "model" organisms that would
reproduce rapidly and easily, primarily in the labo-
ratory.
Studies
of the development of sLx animals
—
nematode worms,
Drosophila firuit flies, zebra fish,
African clawed frogs,
domestic chickens, and house
mice—formed the
basis
for
nearly all we know
about the genetics of development in animals. All
six share
certain traits, such as rapid development
and early
sexual maturation, that tend to minimize
the effects of environment.
The biological
focus
on mice
and
fruit flies
tended to
obscure
environmental
effects.
Jessica
Bolker, an evolutionary
developmental
biologist at
the University of New Hampshire,
has
argued that
biologists, in choosing organisms little
affected by the environment,
have unwittingly rein-
forced
assumptions about the primacy of genes.
All
six of these lab
organisms give molecular genetics
the answer it expects,
namely, that genes rigidly
program
development, independent of the
environ-
ment of the
embryo. As Bolker says, "Most
of our
models are small and fast and
hardwired. . . . And so
we
think
of
development as being
hardwired."
But in the past
decade biologists
have
come to
realize that
development
is
far from hardwired;
instead,
organisms show enormous
developmental
plasticity. Very
recently, a new field of study
—called
ecological
developmental biology, or
eco-devo
—
has
emerged. Eco-devo
examines how developing
individuals
integrate environmental
and genetic in-
formation, as
well
as
how this process
of integration
influences the
direction of evolution.
A basic tenet of
eco-devo is that
individuals with
the same genes can turn out
difierendy,
depending
on
the environment in
which the embryos
find
themselves. This plasticity,
however, is not a general
trait
covering everything an
organism does and is.
Instead,
plasticity itself varies across
traits and species.
Sonia Sultan, of
Wesleyan University
in Middle-
town,
Connecticut, has studied
plasticity in four
closely related
species of buckwheat
in the genus
Polygonum. One of
the four produced
different-size
leaves in
response
to changes
in light intensit)-.
whereas
another
species did not
adapt to Hght at aU.
Sultan
has
also
shown that a species that is plastic
with respect
to one
trait, such as leaf size, may
show
Httle
plasticity
with respect to
another.
Her four
species of
buckwheat differed in the
magnitude,
direction, and
timing of plasticity in
traits as
varied as leaf size,
root length and form, and
rate of
photosynthesis.
These
differences
corre-
sponded
roughly to
the
ecological distribution ot
the
plants. For
instance, the
generalist species Poly-
goiniiii persicaria was
quite
plastic. It reproduced well
in
poor conditions
(doubling
its leat tissue in low
light,
for instance)
but did
better than the other
three
species in
environments
rich in Hght, water, or
nutrients. By
contrast, P
hydropipcr, a more special-
ized
species, showed
far less
plasticity. In poor, shady
conditions it
increased
leaf rissue very litde, and it
only slightly
increased its
reproductive output—as
measured by the
number and
size ot its fruits—
even
in the most
resource-rich
emdronments. The species
apparently could
not take
advantage ot
a
bonanza.
Clearly, the
environment
somehow influences
the
genetic
pathways that guide the
development of
the phenot^'pe.
British biologist C.
H. Waddington
considered
how that might happen.
He found that
two distinct
triggers—one
environmental and one
genetic—can
activate the same
molecular
pathway
during development.
In the 1940s he
was struck by
the fact
that ostriches
hatch with calluses
on their
chests
and abdomens,
in just those
places where
contact
with the ground
later abrades the
skin. Skin
that
is rubbed regularly
becomes
thicker and
tougher as skin cells
prohferate, as a
glance at
our
heels and toes will
confirm. But
our own calluses
are
triggered
by
the
abrasion itself
Waddington
suggested
that in ostriches,
the
trigger for making
calluses had been
transformed from
an
environ-
mental switch to
a
genetic
one—
a
process
he called
"genetic assimilation."
Such developmental
switches
can be found any-
where in the network
of genes
involved in
the
formation of a trait. Ehab
Abouheif
now
at
the
University of Chicago, has
demonstrated
this idea
beautifully in
wingless ants.
Most ant
species have
several castes. Plicidolc
inorrisi
has four:
two with
wings (queens and
males) and
two wingless
(soldiers
and
workers). The network
of six genes
that regu-
lates
wing formation in
these ants
does the same
thing in fruit flies
(Drosophila
melanogaster). And
the
winged ant castes
express the
six genes
controUing
wing formation
in almost
exactly the same
way as
fi-uit flies express
them. These
genes constitute a
sort of gene "cascade,"
with
one gene coding for a
protein that in
turn regulates the
next gene. In sol-
diers, the first five
genes are expressed
normally, just
as
in the winged queens,
but the most
downstream
gene in the cascade is
not. So at
the last moment,
genetically
speaking, the
soldier ants shut off wing
formation. In the
workers,
wing tormation is inter-
rupted farther
upstream in the
gene
cascade.
Sisters in
an anthill are 75
percent genetically
identical. But
whether they become
soldiers, work-
ers, or queens
depends not on any
differences in
their genes, but
on
a
set of
environmental switches.
At the first
switch, the right light and
temperature
cause
the ant embryos to
release a burst
ofjuvenile
hormone, setting
them on the
path to becoming
queens.
Otherwise they become
soldiers or workers.
The desert
"plague" locust
has two
environmentally
influenced forms.
At low
population
densities the
insect is
green,
with small
wings
and legs; at
higher
densities
its
colors
become
mottled
and
brighter, and its
appendages
larger.
When a
species
of
social wasp
attacks
red-eyed
tree
frog
eggs,
some
of the eggs
save
themselves
by
hatching
prematurely.
561
NATURAL HISTORY 10/02
Another Silent
Spring?
That
frogs are disappearing
all over the
world is hardly news. But
that minute amounts of a
common
herbicide can demasculinize
frogs was front-page
news
for
days
last spring,
thanks
to
the work
of developmental
biologist Tyrone
Hayes, of the
University of
Cal-
ifornia, Berkeley.
Atrazine,
considered harmless because it breaks
down in a few days,
is the most
commonly used herbicide in the
world. Farmers make up
for its
rapid breakdown by applying it in
huge
quantities. In
the United States
alone, farmers spray more
than 60
million pounds
of it each year, on corn, soybeans, and
other crops.
Rivers and
streams in agricultural areas may contain
100 to
2,300
parts per billion (ppb).
The ubiquitous herbicide
even falls in
rainwater—1 ppb in
non-agricultural
areas
and
up to
40 ppb
in
agricultural areas.
Early laboratory studies
showed gross malformations in amphib-
ians
exposed
to
atrazine, but this
effect occurred only
at
concentra-
tions that
animals in the wild would rarely encounter. Investigators
did not look for
subtler effects,
so
atrazine was declared safe for the
environment, and the safe
level for drinking water was set at
3 ppb.
But
Hayes's lab has shown that doses as low as
0.
1 ppb
turned male
Xenopiis laevis
tadpoles into hermaphrodites, with three ovaries and
three testes; doses of 1 ppb reduced
the size of muscles of the larynx,
which frogs depend on to call and attract their mates.
Hayes and his coworkers backed up their lab study with a drama-
tic field study of
leopard frogs
(Rana
pipiens). Starting in California,
the biologists drove east across the United States,
collecting
tadpoles
and water samples from Utah to Illinois. The wild frogs were being
hit even harder
than their lab cousins:
nearly fully formed eggs con-
taining
large amounts of yolk were found in the testes
of male
tad-
poles. (Normally, eggs do not develop in tadpoles of either sex.)
And the more atrazine in the water,
says
Hayes, the worse the mal-
formation.
How can
a chemical manufactured to kill plants by
interfering
with photosynthesis
have such profound effects on animals? Atrazine
is
a
potent
endocrine disrupter
because
it
boosts
levels of an enzyme
that normally transforms
testosterone into
estrogen. The result in
male tadpoles is dramatically
reduced testosterone levels and elevated
estrogen levels, an
effect Hayes and his colleagues measured in both
R. pipiens and X. laevis.
Biologists have
known for years
that
amphibian numbers are
dropping precipitously,
with
populations winking out one by
one.
Yet no single cause
seemed to
explain more than a few regional de-
cHnes. Hayes's work
suggests
a major contributing factor in the eighty
countries
that
use atrazine.
Beyond that, his work
suggests
the
im-
portance,
when evaluating
potentially
harmful molecules,
of
looking
at the internal morphology of
developing embryos. Endocrine
dis-
rupters may be capable of destroying
entire populations
and species,
but such compounds
wiU
not necessarily
reveal their effects through
extra legs or other malformations obvious
to a layperson.
At a second switch, a protein-rich diet can trigger
another pulse ofjuvenile hormone, turning the em-
bryos into soldiers; on a poorer diet, they become
workers. Both switches operate
by
means
of a hor-
mone, but the triggers that throw the switch—
food,
temperature, and Ught—are purely environmental.
In
three other ant species,
Abouheif found,
wing
formation
was
interrupted
at a different point
in every caste.
He concludes
that although the net-
work of genes for wing formation is evolutionarUy
stable—conserved in various insects over some
300
million years—ants can turn off wing formation
anywhere in the network. Making wings
is a con-
servative process, but not making them is a flexible
one. Abouheif hypothesizes that such evolutionary
flexibihty may be a general characteristic of organ-
isms that
have
more than one form.
The idea that traits can be controlled by multi-
ple triggers, both endogenous (originating within
the
organism) and exogenous (originating outside
it), is generaHzable and
useful. The
more
medical
investigators understand the triggers that instruct
juvenile brain cells to
multiply and form healthy
new brain tissue, for example,
the more success
they may have turning on this activity
in
adults
Without symbiotic bacteria,
neither mice nor squid
nor humans
develop normally.
whose brains have been damaged (ultimately,
with
a
drug that mimics the
endogenous trigger). And un-
derstanding exogenous
triggers in development can
help identify which synthetic compounds
are hkely
to wreak havoc
on humans and other
organisms
when released into
the environment.
The mechanisms for
these triggers wlU almost
certainly
Ue among the signaling
molecules (hor-
mones and
neurotransmitters, for instance)
that cells
use to talk among
themselves. Signaling
molecules
appear to be the means by
which an organism
con-
verses with its environnrent,
both during early de-
velopment and throughout
life. Other molecules
called
"heat-shock proteins" also
seem to act as
switches that can
decrease or increase
plasticity, es-
pecially when an
organism is under stress.
As we have seen,
signals from the
environment
can
be
physical:
temperature, Ught, pressure,
abra-
sion. They can also
be molecular
(when, for ex-
ample, a
compound that mimics a
hormone alters
gene expression) or
social. Social milieus
can in-
duce many fish to change
from male to female
and
back again. Take the
Japanese
goby
Triinina oki-
lunme. If the resident male in a group
leaves
or dies,
one of the group's females can become a male. But
if a larger male then shows up, the recently remod-
eled "he" reverts to
a
"she." Such transformations
can take place in as Httle as four days.
The
development of some animals is influenced
by
predators. A substance released by predatory
dragonfly
larvae causes wood frog (Rana syhetica)
tadpoles
to
grow smaller than usual and to develop a
deeper tail musculature (which seems to enable
faster swimming and sharper turns). To efl^ect these
changes,
the dragonfly larvae need only
be
in the
water; they needn't actually
be
attacking the tad-
poles. SiiTiilarly, the tiny water flea (Dapluiia aiadlata)
develops a
large protective
"helmet"
when predaceous larvae
of the Chaoboms
fly are present in the water nearby. And
Daphnia is a predator in its own right,
capable of inducing changes in its prey:
green algae. Chemicals released by graz-
ing
water
fleas cause the algae to
give
up
the single-celled life and form colonies.
Relations
between symbioric
bacteria
and their
hosts are another major strand
of the eco-devo
tapestry. The most de-
tailed work in
this area comes from
the
laboratory of Margaret McFall-Ngai at
the University of Hawaii, where biolo-
gists study codevelopment in the squid
Euprytnna
scolopes and the luminescent
bacterium
Vibrio
fischeri.
The bac-
terium guides normal development of
the squid's Hght organs, which illumi-
nate the squid's
body so that
it
does not
appear
to
predators as a conspicuous dark silhouette against
the brightly lit surface of the ocean.
The immature Hght organs of a young
squid
de-
velop
a field of
ciliated cells, which help draw Vib-
rio in from
ocean
water,
as
well
as a
series of deep
pockets,
or
crypts,
in which these bacteria will live.
Within just a few hours, the new arrivals induce the
cells of the light organs to swell and to grow
tiny,
hairlike microvilli. These changes help the
bacteria
flourish within the Hght organs.
Young squid raised
experimentally
in water without
Vibrio don't re-
ceive
the right molecular
signals and thus fail to go
through
normal development.
Invertebrates aren't the only organisms to have
coevolved with bacteria. Mammals and other verte-
brates are walking ecosystems. We
humans nor-
mally carry hundreds of kinds of bacteria in our
mouths alone. And these
symbiotic organisms are
not merely the
inevitable
result
of living in a mi-
crobe-ridden
world. Colonizing the body soon
after birth, they are
in fact essential for normal de-
velopment,
as has been
shown
in
laboratory studies
of mice raised in sterile
environments. Develop-
ment of nearly
aU
the
major organ systems is aber-
rant in these "germ-free" mice, says
McFaU-Ngai.
The Hning of the intestines, for instance,
appears
to
have evolved to interact with bacteria. A few days
before mice are
weaned,
when
bacteria normally
first appear in the gut, the intestinal cells
cover
themselves with a sugar called fiicose,
on which
some symbiotic bacteria can
Hve.
If
none of the
right bacteria
show
up, the fucose
disappears. But if
Left: Predatory
midges
alter the
development
of
the water
flea
Daphnia
cucuUata. In the
top
sequence, a
water flea
exposed
to
these
predators
develops
a long,
pointed "helmet"
and a spiky tail;
in the bottom
sequence, a
water flea grows
in the absence of
predators.
As the
caterpillar
Nemoria
darwiniata feeds,
it adopts
the
color of its host
plant. The white
insect,
far left, is
feeding on the
pale
flowers
of
Ceanothus
velutinus, an
evergreen shrub
in the
buckthorn
family;
the same
caterpillar
turns
purple-red,
left,
when it enjoys
Amelanchier
aim'
folia, a
relative of apples
and roses.
58 NATURAL
HISTORY 10/02
When a school of
scalefin anthias
(Pseudanthias
squamipinnis),
above, lacks
a
male, the
dominant female
turns into
one
(mauve fish
with
yellow
flanks).
Right: The
queen
and
worker ants
shown
here
are sisters
and 75 percent
genetically
identical.
Their obvious
differences arise
out
of differing
diets, which
lead
to a cascade
of
genetic effects.
the right ones do show
up, they induce the gut cells
to make more fucose.
"The host tissues" writes
McFall-Ngai,
"are poised for interaction
with
the
symbiont." Germ-free mice, which
never en-
counter their coevolved
symbionts, need
30
per-
cent more
calories to live than do mice with
a
full
complement
of gut bacteria,
because vertebrates
generally depend
on such bacteria to help digest
food and
even to synthesize vitamins.
One
of the best-known examples
of how
environment
can influence development
comes
from
research on endocrine
dis-
rupters
—molecules
in the envi-
ronment
that bind to receptors
that
normally
link to the body's
own hormones.
Some of these
molecules
are natural
substances,
such
as
the plant
estrogens in
soy-based
baby formula
and
other
soy
products. Many
others
are human
made,
including
the plastic
stabilizers
in
baby bottles,
pacifiers,
dental sealants,
plumb-
ing pipes,
and
gallon milk
jugs, not
to mention
dispersants
used
to spread
pesticides or
to
keep
the
spots off dishes
in dishwashers.
Some of the most disturbing
news on endocrine
disrupters recently emerged
from the laboratory
of
Tyrone Hayes,
a
developmental
endocrinologist
at
the
University of CaHfornia, Berkeley.
Hayes's lab
showed
that minute amounts of atrazine,
a nearly
ubiquitous
herbicide, can derail reproduction
in nat-
ural populations of leopard frogs
by causing males
to
make
eggs (see
"Another Silent
Spring?,"
page
56).
Why is the study of
eco-devo blossoming
now?
One reason is concern
over the increasingly
obvi-
ous effects of endocrine disrupters.
Another, say
several investigators, is the infectious
zeal of devel-
opmental biologist
Scott Gilbert, who in the
past
two years added a chapter on
eco-devo to his best-
selling developmental biology textbook.
Gilbert
recently published an influential
review article in
the journal
Developmental Biology describing
and
naming
the new field, and organized,
with
Jessica
Bolker,
a
symposium
on the subject. Converts to
eco-devo,
who come from every area of biology,
are extraordinarily enthusiastic.
All the enthusiasm in the world
wouldn't have
sufficed, however,
without the major advances in
genetics of
the past decade. Abouheif's research
on
developmental switches in wingless
ants is a good ex-
ample of how developmental
genetics (including the
nrany studies already done on fruit flies) provided the
basis and tools to
do eco-devo. The new tools
—
which include
polymerase chain reaction, a tech-
nique for multiplying traces ofDNA, and microarray
analysis, a method for simultaneously studying the
expression of tens of
thousands of genes— are en-
abhng
scientists to ask and answer whole new sets of
questions. Many bacteria that Hve in animals, for
instance, cannot
be
cultured in
the lab, and they
occur
in numbers too low to detect by conventional
methods.
Only recently, with the advent of micro-
array analysis, have biologists been able to sample and
characterize whole conrmunities of
microorganisms, whether in the
mouth, the gut, or the
soil.
Just
as important as the new
technologies is an increasing em-
phasis on cross-discipUnary work.
Hayes
recalls that when he was in
graduate school,
biologists knew
everything
about
the genetics of
Xenoptis laet'is, the African clawed
frog,
but
little
about the animal itself. "Everything was so
special-
ized," he says. Now he sees entire
fields
as
tools to
ask larger questions: "I used to
think of endocri-
nology as a field, and now I think of it as a
tool to
understand something else, to understand biology."
Ecological developmental biology may lead to
fundamental changes in the
way biologists
think.
For example, an
assumption
of standard
evolu-
tionary
theory
has
been that genetic differences
rigidly determine the relative success or failure of
organisms. But phenotypic plasticity implies
a
degree of play, or looseness, in selection processes.
Biologist Philip Yund, ot the University of
Both
environmental and genetic
triggers can activate the same
molecular pathway.
Maine, says that when
biologists better under-
stand how environmental
information is incorpo-
rated into developmental
processes, they
will
have
a
much more sophisticated
understanding of how
selective
pressures form
the phenotype
over
evolutionary
time.
Hayes points out that
leopard
frog
populations
have now
been exposed
to atrazine
for some
forty
years, long enough
for selective
forces
to have
changed
their biology.
"Effectively,
we've
done
a
pretty' awful
experiment,"
says Hayes,
who
specu-
lates that
a population of frogs living
in
a pond with
high levels ot
the herbicide might
evolve
toward
early metamorphosis
and delayed
sexual maturit\'.
If they can
get out of the pond
soon enough, he
reasons, their gonads
could complete
development
away from the
influence of the
atrazine. In
the
fixture, Hayes will
be looking
for signs that leopard
frogs are evolving in
response to this
herbicide.
Just
as our emaronment
is the context
tor how we
become who we
are, we are also the
context for
the
development of
other organisms
—a conversation
of
which we are
only now becoming fuUy
aware. D
Calluses
usually
form
only when
external
abrasion
turns
on the
right
genes,
but
ostriches
hatch
with calluses
on
their abdomens
and chests.
Evolution can
"genetically
assimilate" an
environmentally
induced
trait.
