Title: Tests on Trial, Science and Technology Section
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00052849/00001
 Material Information
Title: Tests on Trial, Science and Technology Section
Alternate Title: Tests on Trial, Science and Technology Section. The Economist
Physical Description: 2p.
Language: English
Publication Date: September 21, 1991
Spatial Coverage: North America -- United States of America -- Florida
General Note: Box 5, Folder 22 ( SF - SIGNIFICANT NUMBERS ), Item 3
Funding: Digitized by the Legal Technology Institute in the Levin College of Law at the University of Florida.
 Record Information
Bibliographic ID: UF00052849
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: Levin College of Law, University of Florida
Rights Management: All rights reserved by the source institution and holding location.

Full Text

A -3


|bacco, the probability of dying from cancer
Tests o n tr la at a given age is not increasing in the devel-
oped countries.
Dr Ames and Dr Gold have a specific, as
well as a general, critique of animal tests for
cancer. Animal tests look for tumours in rats
BERKELEY, CALIFORNIA and mice fed with high doses of chemicals
over their brief lifespans. If tumours are
IN THE 1970s Bruce Ames, of the Univer- natural ones? The idea that synthetic chemi- found, the product may be banned, ora low
sity of California at Berkeley, was the cals are no more noxious than those occur- limit may be put on its acceptable con-
environmentalists' favourite scientist. He ring in nature is not in itself an argument centration. The concentration which gives a
developed a laboratory test which measured for changing anything-nature can be bad, cancer risk of one in a million per year of
the degree to which various chemicals and there is no reason to add to a natural exposure is a typical limit. That concentra-
might pose a cancer threat; he campaigned, risk. But when tests show that aflatoxin, a tion is calculated from the rate at which the
successfully, for the banning of flame retar- toxin made by moulds, in the average pea- incidence of cancer changes when the dose
dants in clothing because of the cancer nut-butter sandwich poses a hazard 75 to changes.
threat they posed. Yet Dr Ames's one-time 200 times greater than that of ethylene This system reflects a decision taken in
admirers now accuse him of being a turn- dibromide, a fumigant banned in the early the 1970s: that if high levels of a chemical
coat. With his colleague Lois Gold, he has 1980s, then it is fair to ask whether the ban cause cancer, it is safe to assume that low
arrived at new conclusions that lead him to makes sense. doses will too. This idea came from studies
turn his back on a fair bit of what he be- Epidemiology tends to confirm this sort of cancers caused by radiation, which dam-
lieved. He says the chemicals industry may of thinking. Richard Peto of the Imperial ages DNA. The dose of radiation determines
have suffered a grave injustice as a result of Cancer Research Fund's Oxford unit has the extent of DNA damage and thus the
the current system, and that much of the shown that over the past few decades the in- probability of cancer in the irradiated ani-
$100 billion or so spent cutting pollution in cidence of cancer around the world has mal: the higher the dose the more animals
America does little to protect human health. risen, but not because of chemical pollut- get tumours. It was assumed that this meant
Dr Ames's quarrel is with the animal ants in the environment. Smoking and in- that low doses incurred low risks, not no
tests which lead to estimates ofwhether sub- creased life expectancy-the incidence of risks. There is a practical advantage to this
stances cause cancer. He thinks that because cancer rises with age-are mostly to blame. assumption, since one would need to test a
they are based on inaccurate ideas about If one ignores cancer deaths caused by to- chemical on thousands ofanimals to see the
how cancer comes about, they make small effect expected from a low dose.
unwarranted assumptions about the Dr Ames's work in the 1970s was
degree to which data gained from high based on this set of ideas. The "Ames
doses of chemicals can reveal any- test" sorts out potential carcinogens
thing about the effects of low doses. by finding which chemicals damage
He also believes that the tests have led DNA in bacteria. Now he challenges
to misperceptions of the risk of cancer the animal tests on the basis that the
from various sources. cancers measured are not always
From their extensive database on caused directly by damage to DNA. In-
animal tests for carcinogenicity Dr stead, the high doses of chemicals kill
Ames and Dr Gold have found that, of cells, provoking cell division, which
375 synthetic chemicals screened in carries with it its own risks of muta-
both rats and mice, more than half are tion and cancer. So the danger comes
suspected of causing cancer. Looking from the increase in the rate of cell
at 52 "natural pesticides"-chemicals regeneration, rather than the damage
with which plants fight off pests- done to DNA by the chemicals. If this
tested in rats or mice, they found half is the case, it is foolish to extrapolate
of them also failed, ending up listed as the effects of low, non-cell-killing
causing cancer. These account for vir- doses from high doses.
tually all the pesticides that are eaten. Many synthetic chemicals found
Coffee, tomato puree, peanut butter to cause cancer in animals do not
and alcoholic drinks all appear damage DNA at all, the Berkeley scien-
stuffed with naturally occurring tists point out. Saccharin is one such:
carcinogens, and at low doses it appears quite safe.
There is no evidence that eating all On the other hand chemicals cleared
these foods causes cancer, certainly by standard rat tests-such as tobacco
not to any large degree. So why should smoke-have turned out to be carcino-
one think that the synthetic chemicals genic in humans. Michael Cunning-
which fail the tests are anything to ham and his colleagues at America's
worry about, especially as they are National Toxicology Programme,
present in the environment at levels have studied two pairs of chemicals.
thousands of times below those of the What is the point? Each pair is composed of two mole-

cules, each the mirror image among them is Bernard Computer cryptography
of the other. All four chemi- Weinstein, of Columbia Uni-
cals are known to damage versity, New York. He points A cure for the
DNA, but only one of each out that virtually all the spe-
pair causes cancer. In both cific chemicals known from m n
cases, the carcinogenic one epidemiological studies to be
also increases cell division. A" carcinogenic in humans
show up as being so in rodent ANYONE can sign a postcard, but how do
Death and division tests. A paper by David Hoel A you sign a piece of electronic mail?
Dr Ames and Dr Gold are not and colleagues at the Na- Without a "signature" to demonstrate that,
denying that DNA damage, or tional Institute of Environ- say, an electronic transfer of funds really
"mutagenesis", leads to can- mental Health Sciences, pub- comes from someone authorized to make
cer. They are saying that until lished in 1988, provides good the transfer, progress towards all-electronic
cell replication, or "mito- evidence to contradict Dr commerce is stymied. Ways of producing
genesis" takes place, muta- Ames. In a two-year study such signatures are available, thanks to the
genesis does not matter Dr Ames has his doubts with 99 chemicals, they technology of public-key cryptography.
much. So it is often the rate of looked for signs of damage in They will not work to everyone's best advan-
mitogenesis that limits the rate at which tissues taken from rats and mice used in tage, though, until everyone uses the same
cells become cancerous. Part of this view carcinogenicity studies. Cancers occurred public-key system. It is an obvious opportu-
comes from the fact that DNA damage ap- in organs that did not show apparent dam- nity for standards-makers-but in America
pears to be common. In the normal course age; some damaged organs were tumour- they have turned up their noses at all the
ofdaily life, DNA in cells will undergo lots of free. variations on the theme currently in use.
damage due to oxygen radicals, highly reac- According to Dr Ames and Dr Gold, the The alternative standard for digital signa-
tive molecules produced as an incidental epidemiology that links human illness with tures now offered by America's National In-
by-product of normal metabolism. Oxygen chemicals which cause cancer in rats comes stitute of Standards and Technology (NIST)
radicals bind to DNA, forming so-called in the most part from studies in the work- has brought a long-simmering controversy
DNA adducts. When a cell divides, the ad- place. The workers studied have often been back to the boil.
ducts can lead to errors in the DNA copying. exposed to higher concentrations of chemi- Public-key cryptography could become
If these errors accumulate in the wrong cals, nearer to doses used in animal tests one ofthe most common technologies ofthe
genes, they will give rise to a cancerous cell. than those deemed risky for the general information age, underpinning all sorts of
The body has a repair service, a set of en- population. As for Dr Hoel's study, Dr Ames routine transactions. Not only does it prom-
zymes which clip out offending DNA ad- faults it for recording tissue damage, not cell ise to provide the digital equivalent of a sig-
ducts, but it is not totally efficient. Over proliferation-evidence for which has not nature, it could also give users an electronic
time, the errors overlooked by the repair normally been sought. envelope to keep private messages from pry-
mechanisms build up, and the concentra- On a more theoretical level Dr ing eyes. The idea is to create codes that have
tion of damaged DNA in the cell rises. The Weinstein, writing in Science, has argued two related keys. In conventional cryptogra-
higher the level of damage, the greater the that the mitogenesis theory does not incor- phy the sender and receiver share a single
chance that the next time the cell replicates, porate the long-accepted principle that a se- secret key; the sender uses it to encode the
the damage will be converted into a muta- ries of steps are needed to cause a cancer: message, the receiver to decode it. In public-
tion that can cause cancer. So in old people, various genetic changes, then cell prolifera- key techniques, each person has a pair of
and old rats, cancers are more common, be- tion and the expansion of a tumour mass. keys: a disclosed public key and a secret pri-
cause the risks entailed in cell division are Dr Ames disagrees. Among the genes in- vate key. Messages encoded with the private
greater. Dr Ames's experiments show that volved with cancer are the tumour-suppres- key can only be decoded with the cor-
old mice have as many as 2m adducts per sor genes; there are two copies of each of responding public key, and vice versa. The
cell, while young mice have only im. He and them in a healthy cell. Environmental fac- public keys are published like telephone
others have measured the daily excretion of tors such as diet or smoking might damage numbers. The private keys are secret.
DNA adducts clipped out by the repair en- one tumour-suppressor gene and then the With this technology, digital signatures
zymes and released in the urine at a rate of other-but this is made much likelier, Dr are simple. Encode your message, or just the
around 100,000 per cell per day. Against Ames claims, if the cell is reproducing. name you sign it with, using your private
this sort of background, the DNA damage If Dr Ames and Dr Gold are correct, key. If the recipient can decode the message
caused by low doses of industrial chemicals what should replace today's animal tests? with your public key, he can be confident it
begins to look less significant. Dr Ames is not advocating an abolition of came from you. Sending a confidential mes-
Brian Henderson of the University of animal testing, nor ofswitching to low-dose sage-putting electronic mail in a tamper-
Southern California in Los Angeles has testing. In fact, he does not have a solution proof envelope-is equally straightforward.
identified cell division as part of the cause to the problems he raises at all, except to de- To send a secret to Alice encode it with her
in many forms of human cancer. Take lung velop a better understanding of the pro- public key. Only Alice (or someone else who
cancer. Smoking triggers the inflammation cesses that lead to cancer. If it turns out that knows her private key) will be able to de-
ofthe lining of the lung and thus sets off cell he is correct-and although he is in a minor- code the message.
division. The inflammation leads to a high ity, many people in the field respect his The heart of any system of public-key
level of oxygen radicals. Natural repair ideas-then he will have succeeded in re- cryptography is a mathematical function
mechanisms then cause the damaged cells moving a false set of safety requirements, which takes in a message and a key, and puts
to proliferate. Breast cancer in women past and may be able to show that the risks of out a code. This function must be fairly
their menopause is associated with cell divi- cancer from small doses of chemicals in the quick and easy to use, so that putting things
sion brought on by an increase in oestrogen, environment are low. But there is no sign, as into code does not take forever. It must be
Chronic infections, some of which can lead yet, of a system that will tell people exactly very hard to undo, so that getting things out
to cancers, promote cell division and the re- how low-and thus no way of balancing the of code does take forever, unless the decoder
lease of oxygen radicals, risks and benefits, has the decoding key. Obviously, there must
Dr Ames is not without his critics. Chief be no easy way to deduce the private key

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs