FOR decades, scientists have studied
the cellular and genetic damage that follows exposure to high doses
of ionizing radiation such as those resulting from nuclear accidents
or cancer radiotherapy. Much less is known about cellular response
to low doses of ionizing radiationabout 0.1 gray and belowsuch
as that absorbed by our bodies during medical procedures and normal
occupational exposures or while flying in an airplane. (See box
conducted by Lawrence Livermore scientists in the Biology and Biotechnology
Research Program (BBRP) Directorate has revealed that cells exposed
to low-level ionizing radiation respond in a surprisingly robust
manner by turning on or off hundreds of genes, including those specialized
in repairing damaged chromosomes, membranes, and proteins and countering
cellular stress. These genes involved at low dose are different
from the ones responding to high-dose radiation. The discovery that
many different genes are called into action only in response to
low-dose radiation suggests that a cells response at low dose
involves different functions than those occurring at higher doses.
Radiation: A Short Primer
The broad term
radiation includes light and radio waves, but it is
often used to mean ionizing radiation. Ionizing radiation
has sufficient energy to remove electrons from atoms,
thereby creating charged particles (ions or radicals)
in materials it strikes. The different kinds of ionizing
radiation include neutrons and alpha, beta, gamma, and
x radiation. Atoms that emit any of these types of ionizing
radiation are radioactive.
The international standard unit of an absorbed dose
of ionizing radiation is the gray. One gray is equivalent
to the absorption of 1 joule of energy per kilogram
of material. It also equals 100 radiation absorbed doses
(rads) in the old radiation measuring system. A hundredth
of a gray, or one centigray, equals one rad.
Background ionizing radiation levels measure about 0.37
centigray per year, consisting of about 0.3 centigray
from natural sources and about 0.07 centigray from sources
of human activity. Sources of natural ionizing radiation
include radon gas, the human body, rocks and soil, and
cosmic rays. Sources of human-caused ionizing radiation
include medical procedures, consumer products, and,
to a lesser extent, airplane travel, color television,
atmospheric fallout from old nuclear tests, and the
nuclear power industry.
The occupational exposure limits to ionizing radiation
are 5 centigrays per year. Patients undergoing radiation
therapy typically receive a daily dose of about 2 grays,
with a total dose of about 50 grays or more.
Exposure to large amounts of ionizing radiation can
increase the risk of cancer and genetic mutations that
can be passed on to future generations. If the dose
is large enough, massive cell death can occur as part
of acute radiation sickness, which can lead to death.
The extent of cell damage depends on the total amount
of energy absorbed, the time period and dose rate of
exposure, and the particular organs exposed.
Determining exposure limits for workers is an important
task for the Department of Energy and other federal
agencies. The goal of DOEs Low-Dose Radiation
Research Program is to help determine health risks from
exposures to low levels of radiation. This information
is critical to adequately and appropriately protect
people, especially those who are exposed to low levels
of ionizing radiation on the job.
Over the next century, experts predict that radiation
exposures associated with human activity will be primarily
low-dose radiation from medical tests, waste cleanup,
terrorism (dirty bombs), and environmental
isolation of materials associated with nuclear weapons
and nuclear power production.
The chart above shows sources of ionizing radiation
from both natural and human sources.
Livermore research is conducted on tissues of laboratory mice and
human cell cultures. The mouse data show different baselines across
tissues and specialized responses in irradiated brains. The research
in human cells also reveals an intriguing adaptive response, whereby
a very small pretreatment dose of ionizing radiation allows the
cell to better withstand a later, much higher dose. Similar cellular
damage responses may be at work when a cell suffers a low-level
insult (injury) from harmful chemicals or is under attack by bacteria
Livermore research team is led by Andy Wyrobek, head of BBRPs
Health Effects Genetics Division, and is part of the Department
of Energys Low-Dose Radiation Research Program, which aims
to understand the health risks of low-level radiation exposure.
This understanding is critical to setting appropriate exposure standards,
such as those for people receiving medical tests involving radioisotopes
and workers who handle radioactive materials.
BBRPs inception in 1963, Livermore researchers have been studying
the immediate and long-term health effects of radiation on cells,
tissues, and individuals. Livermore-developed techniques, such as
chromosome painting and the Glycophorin A and HPRT assays, have
been used to monitor genetic damage in Japanese survivors of World
War II atomic bomb blasts and in workers cleaning up the Chernobyl
nuclear accident. (See S&TR, September
1999, Researchers Determine Chernobyl Liquidators Exposure.)
Wyrobek says it is well-established that exposure to high doses
of ionizing radiation causes physiological, genetic, and chromosomal
damage. This damage in turn can cause cell death and increase the
risk for later diseases, including cancer and heritable mutations.
simply extrapolating from these effects at higher doses to predict
changes in cells from low-dose exposure is problematic. Numerous
assumptions have traditionally formed the basis for establishing
low-level risk, despite the fact that scientists have been unable
to directly demonstrate irrefutable health risks from low doses
of ionizing radiation.
used high-dose models because, until the past few years, weve
been unable to detect cell changes following low doses of radiation,
says Wyrobek. Thanks to advances in modern molecular biology and
genome instrumentation, much of it developed under the Human Genome
Program, this is changing. We finally have the tools to examine
the damage response patterns in cells from low doses of ionizing
radiation so that we can more scientifically determine health risks
from low-dose exposures to ionizing radiation, he says.
organism’s response to ionizing radiation consists of
a complex set of physical, chemical, and biological events.
Within seconds, radiation produces damage to DNA and oxidizes
proteins and DNA, lipids, and other biomolecules. Within minutes,
the cell responds by changing the activation of certain genes
and modifying some proteins. At high radiation doses, the result
may be acute organ failure leading to death or genomic instability
that causes cancer and birth defects and affects future generations.
Using Mice and Human Cells
low-dose cell responses, the researchers are examining the expression
profiles of thousands of genes in tissues taken from irradiated
adult mice and from irradiated human lymphoblastoid cells (derived
from blood-forming cells). The mammalian brain is a relatively radioresistant
tissue, while the small intestine and blood-forming tissues are
the most sensitive. The team is comparing the findings to control
groups of identical cells that received no radiation.
mouse is an important animal model in radiation biology. Livermore
researchers have studied its genome and found surprising similarities
to the human genome. (See S&TR, May
2001, The Human in the Mouse Mirror.) Mice also provide researchers
an opportunity to study many different organs.
human lymphoblastoid cells were obtained from the National Institutes
of Health, which supplies them to researchers nationwide. The cells,
originally taken from about 450 adults in the U.S. representing
different ethnic backgrounds, are known to be sensitive to ionizing
were performed to study the effects of time and dose on gene expression
in the mouse brain. A group of mice was irradiated with a 0.1-gray
radiation dose from a cesium-137 source, and brain tissue was taken
for analysis at 30 minutes and at 4 hours after irradiation. A second
mouse group was irradiated with a 2-gray dose (20 times the low-dose
radiation and enough to kill some cells), and tissue was sampled
30 minutes and 4 hours later. The same experimental procedure was
used for the human lymphoblastoid tissue cells.
team knew that at higher doses and possibly at low doses of radiation,
some genes would respond by modulating their gene expression; that
is, they would show either an increase or decrease in messenger
RNA (mRNA) or protein levels. (In gene expression, the genes
coded information is converted into mRNA and proteins that are
for cell function and structure.) The researchers examined the
populations of mRNA and proteins present in irradiated cells and
to mRNA and proteins present in nonirradiated cells as a means
to determine whether genes had modulated.
Scientists have been unable to directly
demonstrate irrefutable health risks from low doses of
ionizing radiation. As a result, they have made numerous
assumptions for establishing
Microarrays Are Key
examine the response of tens of thousands of genes, the team turned
to gene-transcript (mRNA) microarray technology, which uses slides
or chips containing arrays of up to 20,000 different genes (specific
sequences of DNA). The team used both Livermore-manufactured DNA
microarrays and commercially available versions.
technology allows us to take a nearly global view of what happens
to a large number of genes in a cell. It replaces the single-gene
approach used in the past, says Wyrobek. He explains that
the technology involves labeling pieces of DNA with fluorescent
molecules and hybridizing (pairing) them to their complementary
DNA target. Much of the fluorescence hybridization technology was
pioneered at Livermore and then transferred to private industry.
the irradiation step, the team extracted the mRNA from the brain
cells, converted it to its complementary DNA (cDNA), labeled that
with a fluorescent dye, and applied the fluid mixture to a microarray.
The different molecules of cDNA in solution paired with their corresponding
genes on the array. The same procedure was done to a control group
BBRP biomedical scientist Francesco Marchetti, We can label
cDNA from an irradiated cell red and label cDNA from a normal cell
green. If we see equal amounts of both red and green for a particular
gene, then we know that radiation causes no modulation of that
By the same logic, if we see all green, then that particular gene
is shut down by radiation. If we see all red, then radiation has
switched on that gene. So the color shifts at each spot on the
give us information on up to 20,000 genes or more.
of the microarray generates large volumes of data that require
biostatistical and bioinformatics methods. Fortunately, Livermore
is the right place to do these kinds of data-intensive experiments, biomedical
scientist Matt Coleman says.
microarray data are beginning to answer several basic questions
the team posed prior to the beginning of the project: Are there
genes with differential expression after radiation exposure? Is
0.1 gray enough to elicit gene expression changes in the adult mouse
brain? What are the cell functions associated with genes affected
by ionizing radiation?
technology allows the simultaneous examination of tens of thousands
of genes through the use of slides or chips. The technology
involves extracting all messenger RNA from the irradiated cells,
converting it to its complementary DNA, labeling it with a
fluorescent dye, and applying it to a microarray. The different
molecules of DNA attach to their corresponding genes. The same
procedure is done to a control group of cells, but with a different
color of fluorescent dye. A laser scans the microarray and
analyzes the intensity of the different colors to give information
on each gene.
Genes Unique to Low Dose
One of the most important findings from the microarray experiments
is that cells exposed to a 0.1-gray radiation dose modulate different
genes than cells exposed to a 2-gray dose. Likewise, there are also
changes over time after exposure; different genes are modulated
at 30 minutes and at 4 hours.
results of one set of experiments involving mice brain cells showed
that at a 0.1-gray dose, 176 genes were modulated at 30 minutes
and 275 genes were modulated at 4 hours. An overlapping set of 48
genes was time-independent. The genes that are switched on are called
REOS genes, or radiation-induced early-onset (within minutes to
hours after exposure) and sensitive genes. At a 2-gray dose, 147
genes were modulated at 30 minutes and 278 genes were modulated
at 4 hours, with 16 genes being time-independent.
surprise was the robust response of cells to ionizing radiation
of only 0.1 gray. Says Wyrobek, When I started this project,
I thought we would see very few changes, if any, from such a low
experiments show that the low-dose response is not simply less than
a high-dose response. Its a lot more complicated than that.
What is happening here is not linear. For the low-dose extrapolation
to be linear, the lower dose would be expected to show less of an
effect on expression than the higher dose. But we found many genes
where something is uniquely happening in response to low dosea
unique set of genes is getting turned on.
the mouse brain cells, the genes that modulated exclusively at
appear to be involved in a broad variety of cell functions, including
cell-cycle control; DNA, RNA, and protein synthesis and repair;
fatty acid metabolism; heat shock; ion regulation; stress response;
membrane repair; and myelin (material surrounding nerve fiber)
repair. The list of these pathways suggests that low-dose ionizing
radiation may activate protective and repair mechanisms, says
biomedical researcher Eric Yin. He notes that low-dose radiation
also depresses genes associated with brain signaling activity,
to divert more resources to repair functions.
general findings for mouse cells were also seen in the human lymphoblastoid
cells, both at the 0.1-gray and 2-gray dose levels. The human tissue
cells showed a different set of genes for low-dose response, as
might be expected because the cells examined were not brain cells.
Wyrobek says it is too early to compare the numbers of genes and
the pathways involved between the two kinds of cells because comparable
microarrays are not yet commercially available.
points out that different tissues are expected to respond differently
to ionizing radiation. In an experiment of unirradiated tissues,
27 percent of 417 genes represented on a microarray were differentially
expressed among five tissues (testis, brain, liver, spleen, and
heart). The expression of the DNA repair genes was the least variable
among the tissues, while genes responsible for coping with general
stress show much greater variability.
(a) The results of one set of experiments
involving mice brain cells showed that at 0.1 gray, 176
genes were modulated (produced more or less messenger RNA)
at 30 minutes and 275 genes were modulated at 4 hours.
An overlapping set of 48 genes was time-independent. (b)
At 2 grays, 147 genes were modulated at 30 minutes, 278
genes were modulated at 4 hours, and 16 genes were time-independent.
Low-Dose Exposure Can Protect
team also discovered that the human lymphoblastoid cells exhibit
what is called an adaptive response to ionizing radiation. An extremely
low dose (also called a priming dose) appears to offer protection
to the cell from a subsequent high dose (2 grays) of ionizing radiation.
The degree of protection was measured by the amount of reduced chromosomal
damage. A priming dose of 0.05 gray, administered about 6 hours
before the high dose, can reduce chromosomal damage by 20 to 50
percent, compared with damage to cells that were not exposed to
the priming dose.
with a low dose of ionizing radiation sets the cell up to better
survive a much higher dose of radiation. A tiny stress apparently
helps a cell get ready for a bigger stress,? says Coleman. About
200 genes were found to be associated with adaptive response in
the human lymphoblastoid cells. Of these, about half were turned
on, and half were turned off. ?We want to know what genes and pathways
are associated with adaptation. Is the adaptive response similar
to the low-dose response? We don?t yet know.? Coleman says that
adaptive responses were first reported in the early 1980s, although
many scientists doubted the accuracy of the reports. ?Now people
are saying this effect happens throughout nature, including in
Regulatory agencies are convinced these effects do appen and that
they may play a role in human health.
The brain seems to respond to low-dose
ionizing radiation by increasing expression (activation)
of genes involved in protective and repair functions while
decreasing brain-signaling activity.
Proteins Provide More Clues
team is also looking for protein changes in irradiated cells. Proteins
give us a more complete picture of cell response to radiation,
says Coleman. However, proteins are more difficult to work with
than mRNA because of their instability and many modified forms.
They can go through many reactions that make them active
researchers are using a number of techniques to identify radiation-induced
proteins. They are collaborating with colleagues at Livermore and
Pacific Northwest Laboratory in using specialized mass spectrometers
to gain a better understanding of the proteins. So far, the spectrometers
have shown that two proteins, as yet unidentified, seem to be produced
in large quantities only in response to high-dose ionizing radiation
and are produced in much lower quantities in response to low-dose
team is also using two-dimensional gel electrophoresis, an old and
more established technique, to separate and identify proteins. This
technology works by separating proteins by their size and electrical
researchers have also begun using protein microarrays, which work
in a similar manner to DNA microarrays. The value of this technique
is limited, however, because users must know in advance what proteins
they are trying to find.
Exposure of the mouse brain to ionizing
radiation induces time-dependent changes in gene-transcript
(mRNA) expression. Genes associated with specific biological
functions show several distinct patterns of radiation response:
early-onset and transient, late-onset, and persistent over
time. Genes associated with ion regulation and control
of gene expression showed early-onset and transient changes.
Genes associated with radiation protection (for example,
heat shock, oxidative stress) and synaptic signaling showed
early onset with both transient and persistent patterns.
Genes associated with cellular repair (for example, myelin,
protein synthesis) showed late-onset changes in expression.
More Work Ahead
work lies ahead. We still have to show exactly what cell
mechanisms and pathways come into play. We need to identify the
genes that are expressed in association with a low-dose ionizing
radiation exposure and those that are expressed for adaptive response, says
Wyrobek. The team also needs to better understand the differences
among tissues and how these relate to the risk of acute radiation
sickness and long-term health effects.
DOE low-dose program is also preparing to study low-dose response
from other chemical and microbial toxins. These radiation
effects studies are setting the stage for modern molecular toxicology
of cells and tissues, says Wyrobek.
Radiation is just one kind of toxic material that damages
chromosomes and kills cells. Does a cell respond in like manner
to toxic chemicals, bacterial toxins, or even to cell toxicity caused
by bacteria or viral infections? We dont know. We do know
that some of the genes involved in cell response to low-dose ionizing
radiation are the same ones that respond to chemical stress and
to viral and bacterial infection. The answers to all the
radiation-response questions may have a huge effect on understanding
whether high doses
of a suspected toxic chemical on laboratory animals are relevant
to humans ingesting the same material but in much smaller doses.
worthwhile avenue of research is determining if individual genetic
differences exist that render some people more or less sensitive
to ionizing radiation. Wyrobek notes also that the Livermore experimental
findings are based on the aggregates of millions of cells. It
is possible, for example, that just one or only a few types of
within a tissue can respond differently to ionizing radiation.
We already know that cells in tissues differ dramatically in their
sensitivity to cell killing, but we know little about the underlying
molecular mechanisms. Determining the differential response of
in tissues to insult is an important next step of research.
Chitra Manohar shows colleague Hitesh Kapur the analytical
results of changes in gene expression of tens of thousands
of genes examined by microarray technology. The microarrays
are on the spotting robot to the right of the computer screen.
Livermore has developed customized microarrays for analyzing
specific sequences of DNA and proteins (see inset).
important new research tool available to Livermore researchers
is a nanoscale
dynamic secondary-ion mass spectrometer (NanoSIMS). This instrument
is only the second such machine in the nation dedicated to biological
research. It can scan a tissue and identify the regions where a
selected gene is expressed. (See An
Inside Attack on Cancer for a discussion of NanoSIMS used for
quantitative imaging of biological materials.)
says the experimental findings are relevant to homeland security
and for assessing biological dose after incidents of chemical
and biological warfare and so-called dirty radiological bombs.
is a suspected exposure incident, investigators will always have
to try to determine exposure dose and assess health effects.
notes that it is too early to tell if exposure standards will be
changed as a result of the work funded by the DOE low-dose program. We know a lot of things are going on at the low-dose level.
What they all mean in terms of health is uncertain, says Wyrobek.
But the new knowledge will certainly help ensure that the existing
standards are appropriate. At the very least, he says, We
should no longer assume that cells respond in a linear fashion
exposure to ionizing radiation.
Key Words: DNA, human lymphoblastoid cells, ionizing radiation,
Low-Dose Radiation Research Program, messenger RNA (mRNA), nanoscale
dynamic secondary ion mass spectrometer (NanoSIMS), radiation-induced
early-onset and sensitive genes (REOS).
For further information contact Andrew Wyrobek (952) 422-6296
a printer-friendly version of this article.