Environmental Health Perspectives • volume 117 | number 3 | March 2009
Regulatory agencies in the United States
and the European Union (EU) have justi-
fied the decision to declare the estrogenic
chemical bisphenol A (BPA) safe at current
levels of human exposure based on a few
studies conducted using Good Laboratory
Practices (GLP). In contrast, these agencies
have rejected for consideration in their risk
assessment of BPA hundreds of laboratory
animal and mechanistic cell culture studies
conducted by academic and government sci-
entists reporting harm at very low doses of
BPA. These studies were rejected primarily
because they were not conducted using GLP.
We suggest that decisions based on this logic
are misguided and will result in continued
risk to public health from exposure to BPA,
as well as other manmade chemicals.
GLP is a federal rule for conducting
research on the health effects or safety testing
of drugs or chemicals submitted by private
research companies for regulatory purposes.
The GLP outlines basic guidelines for conduct-
ing scientific research, including the care and
feeding of laboratory animals, standards for
facility maintenance, calibration and care of
equipment, personnel requirements, inspec-
tions, study protocols, and collection and
storage of raw data (Goldman 1988). These
regulations were developed in response to wide-
spread misconduct by private research compa-
nies; this misconduct was possible because their
data usually do not go through the rigorous,
multi stage scientific review that is normal for
academic data funded by federal agencies and
published in the peer-reviewed literature. The
lack of these safeguards from academic science
had enabled fraud. The U.S. Food and Drug
Address correspondence to J.P. Myers, Environmental
Health Sciences, 421 Park St., Charlottesville, VA
22902 USA. Telephone: (434) 220-0348. Fax: (434)
220-0347. E-mail: email@example.com
The authors declare they have no competing
Received 9 September 2008; accepted 22 October
Why Public Health Agencies Cannot Depend on Good Laboratory Practices
as a Criterion for Selecting Data: The Case of Bisphenol A
John Peterson Myers,1 Frederick S. vom Saal,2 Benson T. Akingbemi,3 Koji Arizono,4 Scott Belcher,5
Theo Colborn,6 Ibrahim Chahoud,7 D. Andrew Crain,8 Francesca Farabollini,9 Louis J. Guillette Jr.,10
Terry Hassold,11 Shuk-mei Ho,12 Patricia A. Hunt,11 Taisen Iguchi,13 Susan Jobling,14 Jun Kanno,15 Hans Laufer,16
Michele Marcus,17 John A. McLachlan,18 Angel Nadal,19 Jörg Oehlmann,20 Nicolás Olea,21 Paola Palanza,22
Stefano Parmigiani,22 Beverly S. Rubin,23 Gilbert Schoenfelder,24 Carlos Sonnenschein,23 Ana M. Soto,23
Chris E. Talsness,25 Julia A. Taylor,2 Laura N. Vandenberg,23 John G. Vandenbergh,26 Sarah Vogel,27
Cheryl S. Watson,28 Wade V. Welshons,29 and R. Thomas Zoeller30
1Environmental Health Sciences, Charlottesville, Virginia, USA; 2Division of Biological Sciences, University of Missouri, Columbia,
Missouri, USA; 3Department of Anatomy, Physiology and Pharmacology, College of Veterinary Medicine, Auburn University, Auburn,
Alabama, USA; 4Faculty of Environmental and Symbiotic Science, Prefectural University of Kumamoto, Tsukide, Kumamoto, Japan;
5Department of Pharmacology and Cell Biophysics, Center for Environmental Genetics, University of Cincinnati, Cincinnati, Ohio,
USA; 6The Endocrine Disruption Exchange, Paonia, Colorado, USA; 7Institut für Klinische Pharmakologie und Toxikologie Charité,
Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany; 8Department of Biology, Maryville College, Maryville,
Tennessee, USA; 9Dipartimento di Fisiologia, Università di Siena, Siena, Italy; 10Department of Zoology, University of Florida,
Gainesville, Florida, USA; 11School of Molecular Biosciences, Washington State University, Pullman, Washington, USA; 12Department
of Environmental Health, University of Cincinnati, Cincinnati, Ohio, USA; 13National Institutes of Natural Science, Okazaki Institute for
Integrative Bioscience, Bioenvironmental Science, Okazaki, Japan; 14Department of Biological Sciences, Brunel University, Uxbridge,
United Kingdom; 15Division of Cellular and Molecular Toxicology, National Institute of Health Sciences, Tokyo, Japan; 16Department
of Molecular and Cell Biology, University of Connecticut, Storrs, Connecticut, USA; 17Department of Epidemiology, Rollins School of
Public Health, Emory University, Atlanta, Georgia, USA; 18Center for Bioenvironmental Research, Tulane and Xavier Universities, New
Orleans, Louisiana, USA; 19Instituto de Bioingeniería and CIBERDEM, Universidad Miguel Hernández de Elche, Alicante, Spain; 20Goethe
University Frankfurt am Main, Department Aquatic Ecotoxicology, Frankfurt, Germany; 21Hospital Clínico, CIBERESP, University of
Granada, Granada, Spain; 22Dipartimento di Biologia Evolutiva e Funzionale, Universita’ di Parma, Parma, Italy; 23Tufts Medical School,
Boston, Massachusetts, USA; 24Institute of Pharmacology and Toxicology, University of Wuerzburg, Wuerzburg Germany; 25Charité
University Medical School Berlin, Berlin, Germany; 26Department of Biology, North Carolina State University, Raleigh, North Carolina,
USA; 27Chemical Heritage Foundation, Philadelphia, Pennsylvania, USA; 28Biochemistry and Molecular Biology, University of Texas
Medical Branch, Galveston, Texas, USA; 29Department of Biomedical Sciences, University of Missouri, Columbia, Missouri, USA;
30Biology Department, University of Massachusetts, Amherst, Massachusetts, USA
Background: In their safety evaluations of bisphenol A (BPA), the U.S. Food and Drug
Administration (FDA) and a counterpart in Europe, the European Food Safety Authority (EFSA),
have given special prominence to two industry-funded studies that adhered to standards defined by
Good Laboratory Practices (GLP). These same agencies have given much less weight in risk assess-
ments to a large number of independently replicated non-GLP studies conducted with government
funding by the leading experts in various fields of science from around the world.
oBjectives: We reviewed differences between industry-funded GLP studies of BPA conducted by
commercial laboratories for regulatory purposes and non-GLP studies conducted in academic and
government laboratories to identify hazards and molecular mechanisms mediating adverse effects.
We examined the methods and results in the GLP studies that were pivotal in the draft decision of
the U.S. FDA declaring BPA safe in relation to findings from studies that were competitive for U.S.
National Institutes of Health (NIH) funding, peer-reviewed for publication in leading journals,
subject to independent replication, but rejected by the U.S. FDA for regulatory purposes.
discussion: Although the U.S. FDA and EFSA have deemed two industry-funded GLP studies of
BPA to be superior to hundreds of studies funded by the U.S. NIH and NIH counterparts in other
countries, the GLP studies on which the agencies based their decisions have serious conceptual and
methodologic flaws. In addition, the U.S. FDA and EFSA have mistakenly assumed that GLP yields
valid and reliable scientific findings (i.e., “good science”). Their rationale for favoring GLP studies
over hundreds of publically funded studies ignores the central factor in determining the reliability and
validity of scientific findings, namely, independent replication, and use of the most appropriate and
sensitive state-of-the-art assays, neither of which is an expectation of industry-funded GLP research.
conclusions: Public health decisions should be based on studies using appropriate protocols with
appropriate controls and the most sensitive assays, not GLP. Relevant NIH-funded research using
state-of-the-art techniques should play a prominent role in safety evaluations of chemicals.
key words: bisphenol A, endocrine disruptors, FDA, Food and Drug Administration, GLP, good
laboratory practices, low-dose, nonmonotonic, positive control. Environ Health Perspect 117:309–315
(2009). doi:10.1289/ehp.0800173 available via http://dx.doi.org/ [Online 22 October 2008]
Myers et al.
volume 117 | number 3 | March 2009 • Environmental Health Perspectives
Administration (U.S. FDA) first issued rules
for GLP in 1978 after a 2-year federal inves-
tigation into sloppy laboratory practices of a
number of private research companies (Lublin
1978; Markowitz and Rosner 2002). What
began as serious concerns about poor quality
research expanded into a criminal investiga-
tion of Industrial Bio-Test (IBT), one of the
largest private laboratories at the time and a
subsidiary of Nalco Chemical Company. In
response to the federal investigation, the U.S.
Environmental Protection Agency (EPA)
demanded that 235 chemical companies re-
examine the > 4,000 tests conducted by the
laboratory. In 1983, three men from IBT were
found guilty of deliberating doctoring data
and were sentenced to prison (Lublin 1978;
Markowitz and Rosner 2002). The fraudulent
practices of IBT brought into question 15% of
the pesticides approved for use in the United
States. That same year, the U.S. EPA issued
similar GLP rules for regulatory testing.
Both the U.S. FDA (2008a) and European
Food Safety Authority (ESFA 2006) have
recently published documents demonstrating
that their decision to continue to declare BPA
safe at current exposure levels was based pri-
marily on the results of a few industry-funded
studies that followed GLP guidelines. These
decisions stand in stark contrast to the deci-
sions concerning the potential risks to human
health reached by a panel of 38 experts at a
U.S. National Institutes of Health (NIH)-
sponsored conference, who published The
Chapel Hill Consensus Statement (vom Saal
et al. 2007), as well as five review articles (Crain
et al. 2007; Keri et al. 2007; Richter et al.
2007a; Vandenberg et al. 2007a; Wetherill
et al. 2007). These peer-reviewed articles cov-
ered approximately 700 articles concerning
BPA and represented a comprehensive review
of the literature as of the end of 2006. In
addition, the U.S. FDA draft decision contra-
dicted the conclusions reached by the National
Toxicology Program (NTP), which had spent
2 years investigating this question (NTP 2008).
An important role of the NTP is to advise the
U.S. FDA about the science relating to toxic
chemicals in food, but in an unusual move,
the U.S. FDA chose to release its draft report
before the release of the final report on BPA
by the NTP and without indicating who at the
U.S. FDA was involved in preparing the draft
report (U.S. FDA 2008b). At a hearing on
16 September 2008 regarding the draft report
on BPA, the U.S. FDA announced that their
goal was to have a subcommittee of the U.S.
FDA Science Board complete a review of the
draft decision by the end of October 2008.
This would presumably also involve review by
the sub committee members of the approxi-
mately 1,000 articles relating to BPA.
We believe that the methods employed in
chemical industry–sponsored GLP studies are
incapable of detecting low-dose endocrine-
disrupting effects of BPA and other hormon-
ally active chemicals. Detecting endocrine-
disrupting effects at low doses of chemicals
such as BPA requires sophisticated and mod-
ern assays and analyses that have been devel-
oped in advanced, usually federally funded
laboratories over the past decade. This is espe-
cially apparent when one examines what is
now known about functional effects of BPA
on a wide range of end points (Richter et al.
2007a; Welshons et al. 2006; Wetherill et al.
2007). These end points include those medi-
ated by recently discovered estrogen response
pathways initiated in human and animal cell
membranes (nonclassical or alternative estro-
gen response mechanisms), which multiple
laboratories have shown to be equally sensitive
to BPA and estradiol in terms of activating
effects in human and animal cells at low pico-
molar through low nanomolar concentrations
(Alonso-Magdalena et al. 2008; Wetherill
et al. 2007; Wozniak et al. 2005; Zsarnovszky
et al. 2005).
The effects of BPA documented in these
studies include a diverse array for which there
are no data from GLP studies because the
end points have not been examined: altered
metabolism related to metabolic syndrome
(Alonso-Magdalena et al. 2005, 2006, 2008;
Ropero et al. 2008); altered adiponectin secre-
tion (Hugo et al. 2008), which is a condition
predicting heart disease and type 2 diabetes
(Lang et al. 2008); altered epigenetic pro-
gramming leading to precancerous lesions
of the prostate (Ho et al. 2006); differential
growth patterns in the developing prostate
(Timms et al. 2005); abnormal growth,
gene expression, and precancerous lesions
of the mammary glands (Soto et al. 2008);
and adverse effects on the female reproduc-
tive system, including uterine fibroids, para-
ovarian cysts, and chromosomal abnormalities
in oocytes (Newbold et al. 2007; Susiarjo
et al. 2007). There is also a large literature
on neuro anatomic, neurochemical, and
behavioral abnormalities caused by low doses
of BPA (Leranth et al. 2008; Richter et al.
2007a), which also are not capable of being
detected by current GLP studies conducted
for regulatory purposes because of their out-
The approaches used by academic and
government scientists to study the potential
health hazards of BPA contrast sharply with
those still used by the chemical industry that
are relied on by regulatory agencies in the
United States and Europe, including the two
studies identified by both the U.S. FDA and
European Food Safety Authority (EFSA) as
central to the decision to declare BPA safe
at current human exposure levels (Tyl et al.
2002, 2008a). By using outdated and insensi-
tive assays that were supposed to have been
replaced by a new battery of screens and tests
by 2000 [as mandated by the U.S. Congress
in 1996 in the Food Quality Protection
Act (1996), but which has, as yet, still not
occurred], these studies conducted using GLP
fail to find any adverse effects.
Reliability and Validity
Reliability and validity are separate issues,
although in the experimental research described
here, validity and reliability basically refer to
research that is credible. Golafshani (2003)
noted that “reliability” refers to the extent to
which results are consistent over time and are
an accurate representation of the total popu-
lation under study. Of central importance is
that the results of a study must be reproduced
under a similar methodology to be considered
to be reliable. “Validity” refers to whether
the research measures what it was intended
to meas ure, and valid findings are considered
to be true. In other words, reliability is deter-
mined by whether the results are replicable,
whereas validity is assessed by whether the
methods used result in finding the truth as a
result of the investigator actually measuring
what the study intended to measure.
Use of GLP in Regulatory
Despite strong evidence of aberrations caused
by low doses of BPA in animals exposed
during fetal and neonatal life in studies con-
ducted by the world’s leading academic and
government experts in the fields of endocrine
disruption, endocrinology, neuro biology,
reproductive biology, genetics, and metabo-
lism, a relatively small number of studies
reporting no adverse effects at low doses of
BPA have continued to be promoted by the
chemical industry and used by regulatory
agencies (e.g., Ashby et al. 1999; Cagen et al.
1999; Tyl et al. 2002, 2008a). According to
the U.S. FDA, these are accepted because they
used GLP (U.S. EPA 2008), with the implica-
tion that studies not employing GLP are not
reliable or valid (U.S. FDA 2008a).
GLP does not guarantee reliability or
validity of scientific results. Unfortunately,
although GLP creates the semblance of reli-
able and valid science, it actually offers no
such guarantee. GLP specifies nothing about
the quality of the research design, the skills of
the technicians, the sensitivity of the assays,
or whether the methods employed are current
or out-of-date. (All of the above are central
issues in the review of a grant proposal by
an NIH panel.) GLP simply indicates that the
laboratory technicians/scientists performing
experiments follow highly detailed U.S. EPA
requirements [or in the EU, Organization for
Economic Co-operation and Development
(OECD) requirements] for record keep-
ing, including details of the conduct of the
GLP is not a guarantee of reliable science
Environmental Health Perspectives • volume 117 | number 3 | March 2009
experiment and archiving rele vant biological
and chemical materials (U.S. EPA 2008).
These record-keeping procedures in GLP
were instituted because of widespread mis-
conduct being committed by commercial
testing laboratories (described above). These
fraudulent results were possible because con-
tract laboratory studies used in the regula-
tory process are rarely subject to the checks
and balances that peer-reviewed, replicated
scientific findings undergo. Without that
acid test of reliability (replication by other
independent scientists), other procedures
were needed. Hence GLP was implemented,
despite its severe limitations.
NIH-funded research subject to more strin-
gent reviews than GLP. Although few NIH-
funded investigators adhere to GLP-mandated
record keeping, the procedures of GLP are
actually surpassed by the procedures required
for NIH-funded science published in peer-
reviewed journals. NIH-funded studies pass
through three phases of peer review that are
far more challenging than GLP requirements.
First, the principal scientists must have dem-
onstrated competence to conduct the research,
and experimental methods, assays, and labora-
tory environment must involve use of state-of-
the-art techniques to be competitive for NIH
funding. Second, results are published in peer-
reviewed journals, with detailed evaluations
by independent experts examining all aspects
of the study. And third, the findings are chal-
lenged by independent efforts to replicate; for
example, the initial findings concerning the
stimulating effects of estrogenic chemicals on
the mouse prostate (Nagel et al. 1997; vom
Saal et al. 1997) were independently replicated
and extended by Gupta (2000), which led to
an editorial identifying “initial results con-
firmed” (Sheehan 2000).
Typically, within a laboratory, interest-
ing findings are also followed by subsequent
publications extending the prior findings;
examples include the findings of BPA effects
on β cells in the mouse pancreas (Alonso-
Magdalena et al. 2005, 2006, 2008) and the
effects of estrogenic chemicals and drugs on
the developing mouse prostate that followed
earlier findings (described above) from this
same group (Timms et al. 2005; Richter et al.
2007b). In particular, independent replica-
tion by competent, respected scientists is the
main criterion of acceptance of the findings as
having been demonstrated to be reliable and
having been validated by virtue of coming to
the same conclusion using a variety of sophis-
ticated techniques in multiple publications.
An important criticism of the approach
taken by the U.S. FDA in its assessment of
the now approximately 1,000 articles on BPA
is that it appears to have made no attempt
to connect the dots between replicated stud-
ies; instead, the U.S. FDA appears to have
assessed each study without regard to whether
it had been confirmed by other studies.
Thus, collectively, many phases used to
verify the reliability and validity of NIH-
funded published research have been com-
pletely ignored by the U.S. FDA, whereas
industry-funded GLP research is rarely, if
ever, subject to these central requirements and
yet is accepted by regulatory agencies as reli-
able and valid.
The U.S. FDA’s misguided gold standard.
In this light, the U.S. FDA’s reliance upon
GLP as the gold standard is scientifically mis-
guided. Furthermore, U.S. FDA administra-
tors are ignoring published critiques of the
GLP studies it considers reliable and valid,
such as the study by Tyl et al. (2002) and two
coordinated studies conducted at the same
time by Ashby et al. (1999) and Cagen et al.
(1999). Each was an industry-funded study
conducted using GLP. Each was harshly
criticized in peer-reviewed publications by
academic scientists and government pan-
els [Center for the Evaluation of Risks to
Human Reproduction (CERHR) 2007; NTP
2001; vom Saal and Hughes 2005; vom Saal
and Welshons 2006]. Yet, the U.S. FDA
and EFSA panels still assert that these stud-
ies represent the gold standard in toxicologic
Specifically, the studies of Cagen et al.
(1999) and Ashby et al. (1999) were recently
rejected by the NTP CERHR panel on BPA
as unusable for consideration in its evaluation
of the health hazards posed by BPA (CERHR
2007). Both the Ashby et al. (1999) and Cagen
et al. (1999) studies reported finding no effect
of their positive control [the estrogenic drug
diethylstilbestrol (DES)] on any outcome,
although these failures were not acknowledged
by the authors in either article. In experimen-
tal science, the failure of a positive control
to show an effect indicates the experiment
failed, which is the conclusion reached by the
CERHR panel (CERHR 2007).
The Tyl et al. 2002 study, which the U.S.
FDA still accepts as a major study for determi-
nation of the safety of BPA (U.S. FDA 2008a,
2008b), was criticized by an NTP panel that
met in 2000 to examine the low-dose issue
(NTP 2001), as well as in subsequent publica-
tions (vom Saal and Hughes 2005; vom Saal
and Welshons 2006), for using an insensitive
rat (the CD-SD rat) that requires extremely
high doses (≥ 50 µg/kg/day) of the potent
estrogenic drug ethinylestradiol to show
effects such as those examined in the study by
Tyl et al. (2002). This dose of ethinylestradiol
is > 100 times higher than the approximately
0.3 µg/kg/day used by women in oral con-
traceptives. The fact that Tyl et al. (2002)
adhered to GLP did not protect them from
using insensitive animals. This led the NTP
(2001) to state:
Because of clear species and strain differences
in sensitivity, animal model selection should be
based on responsiveness to endocrine-active agents
of concern (i.e., responsive to positive controls),
not on convenience and familiarity.
Thus, when reviewed by other scientists,
three prior major GLP studies of BPA have
been found to be so flawed as to be useless
for guiding regulatory agencies in decision
making. A new GLP study has now been pub-
lished by Tyl et al. (2008a). Close examina-
tion of this study also reveals fatal flaws which
render it useless for regulatory purposes, even
though it conforms to GLP.
Examples of Flaws Ignored by
the U.S. FDA and EFSA in a
Recent GLP Study of BPA
In summary, the flaws in Tyl et al. (2008a)
are as follows:
trol (estradiol) to cause an effect means the
system used by Tyl et al. (2008a), at least
in her laboratory, is relatively insensitive to
exogenous estrogens and thus inappropri-
ate for studying low-dose effects of estro-
genic compounds such as BPA. The lack
of response to low doses of estradiol or
BPA in the Tyl laboratory is puzzling, in
that the strain of mice used in these experi-
ments (the CD-1 mouse) has been reported
in > 20 other peer-reviewed publications to
show adverse effects in response to very low
doses of BPA (vom Saal 2008), as well as
many other studies showing low-dose effects
in response to the natural hormone estra-
diol, the estrogenic drugs ethinylestradiol
and DES, and to other estrogenic chemicals.
date protocols and assays that are incapa-
ble of finding many of the adverse effects
reported by more sophisticated studies
conducted by independent NIH-funded
scientists as well as scientists funded by gov-
ernment agencies in other countries.
prostate weight, Tyl et al. (2008a) reported
an abnormally high prostate weight for con-
trol animals that exceeds by > 70% the pros-
tate weights reported by other studies for
animals of the same strain and similar age
(e.g., Gupta 2000; Ruhlen et al. 2008). This
suggests that the dissection procedures for
the prostate in the Tyl laboratory included
other non prostatic tissues in the weight
measurements, rendering them unusable for
studying weight changes in the prostate in
response to BPA or estradiol; neither chemi-
cal showed any effect on the selected end
points, which directly contradicts other find-
ings concerning opposite effects of low and
high doses of estrogen on the prostate (Putz
et al. 2001; Timms et al. 2005; vom Saal
et al. 1997).
Myers et al.
volume 117 | number 3 | March 2009 • Environmental Health Perspectives
Aberrant insensitivity of CD-1 mouse to
estrogens. Tyl et al. (2008a) used estradiol as
a positive control. It was fed to female mice
before and during pregnancy and lactation at
80–220 µg/kg/day; after weaning, estradiol was
fed to offspring at doses of 80–100 µg/kg/day.
Estradiol was used as a positive control because
BPA is a man-made endocrine-disrupting
Many published findings reporting effects
of very low doses of positive control estro-
gens and BPA in CD-1 mice demonstrate
that the CD-1 mouse was somehow rendered
insensitive in the test system used by Tyl et al.
(2008a). The fact that a dose of 100–200 µg/
kg/day estradiol was necessary to show an
effect of the positive control predicts that Tyl
et al. (2008a) should not detect effects of BPA
< 10–100 mg/kg/day, far above the low-dose
range relevant to human exposures that was
supposedly of interest.
For nuclear estrogen receptor–mediated
effects via regulation of gene activity (nuclear
estrogen receptors are transcription fac-
tors whose activity is regulated by binding to
estrogen), prior studies have typically shown
a 1,000-fold lower activity for BPA relative to
estradiol or potent estrogenic drugs, includ-
ing DES and ethinylestradiol. For example,
Richter et al. (2007b) reported an increase in
androgen receptor gene activity to estradiol
at 1 pM (0.28 pg/mL) in fetal CD-1 mouse
prostatic mesenchyme cells in primary culture,
and the same response was found for BPA at
1,000 pM (228 pg/mL); the in vitro response
to estradiol was predicted by the response of
the prostate to increasing free serum estradiol
from 0.2 to 0.3 pg/mL in male mouse fetuses
via estradiol administration to the mother
(vom Saal et al. 1997). Other research showed
that a significant effect on development of
the male reproductive system in CF-1 mice
occurred at a maternal dose of 0.002 µg/kg/day
ethinylestradiol (Thayer et al. 2001), similar
to effects observed with 2–20 µg/kg/day BPA
(vom Saal et al. 1998). The research of Honma
et al. (2002) showed accelerated puberty in
CD-1 (ICR) mice at a DES dose of 0.02 µg/
kg/day (the positive control), and the same
response to BPA occurred at 20 µg/kg/day,
again revealing a 1,000-fold difference between
the positive control estrogen and BPA.
There are many other examples of findings
where a higher dose of BPA was required to
cause the same effect as the positive control
estrogen (estradiol, ethinylestradiol, or DES) in
studies where the effects were mediated by the
classical nuclear estrogen receptors, in contrast
to the more recently discovered rapid signaling
estrogen response system where BPA and these
positive control estrogens have equal potency,
as described above. In summary, CD-1 mice
have been used by a large number of academic
and government investigators and have been
reported in peer-reviewed publications to be
sensitive to positive control estrogens within
the range of human sensitivity based on in vivo
and in vitro studies via the classical estrogen
receptor α–mediated response mechanism.
The CD-1 mouse is the animal model that has
been used by the U.S. National Institute of
Environmental Health Sciences (NIEHS) for
decades, because it is considered the best ani-
mal model for predicting the effects of devel-
opmental exposure to estrogen in humans
(Newbold 1995; Newbold et al. 2007).
The failure of traditional toxicologic stud-
ies conducted by Tyl et al. (2008a, 2008b)
to detect the wide range of adverse effects of
even relatively high doses of BPA or of low
doses of estradiol that have been reported in
numerous studies by academic and govern-
ment scientists provides evidence that the GLP
protocols established long ago by regulatory
agencies to determine the toxicity of chemicals
are inappropriate for detecting the endocrine-
disrupting activities of chemicals such as BPA.
Indeed, this was the premise of the congres-
sional mandate in the Food Quality Protection
Act (1996) for the U.S. EPA to establish a new
set of assays for endocrine-disrupting chemicals,
although this process has been systemati cally
delayed and is > 8 years behind the congres-
sionally mandated date of 2000 to have these
new assays validated.
Citing Tyl et al. (2008a), the EFSA report
on BPA (EFSA 2006) stated that “the posi-
tive control substance, 17β-estradiol, resulted
in reproductive and developmental toxicity.”
This report failed to acknowledge that only a
very high dose of the positive control was suf-
ficient to elicit effects and that this meant that
the experiments conducted in the Tyl labora-
tory were for some reason very insensitive to
any estrogen and thus inappropriate for use in
a study to examine low-dose estrogenic effects
Based on the preliminary report released
by the U.S. FDA regarding BPA (U.S. FDA
2008a), it appears that the U.S. FDA has
followed the lead of the EFSA in its lack of
understanding of the importance of the dose
of the positive control estrogen required to
cause adverse effects. The consequence is that
the U.S. FDA has relied primarily on the study
of Tyl et al. (2008a, 2008b), with the result
that the U.S. FDA has assured Americans that
BPA is safe at current human exposure levels.
Several factors might account for the insen-
sitivity of the CD-1 mouse in the Tyl et al.
studies (2008a, 2008b) conducted at Research
Triangle Institute (RTI), a testing facility that
conducted these (as well as previous) studies
funded by the American Chemistry Council.
One possibility is that the diet used in these
studies may have interfered with the results.
The feed used by Tyl et al. (2008a) in this
experiment (Purina 5002) has been shown by
others to interfere with responses to exogenous
estrogenic chemi cals, blocking adverse effects
documented on other diets. For example, a
number of years ago, Thigpen et al. (2003) at
the NIEHS recommended against the use of
Purina 5002 in studies of endocrine- disrupting
chemicals. Tyl et al. (2008a) measured some
specific phyto estrogens in Purina 5002 feed
by chemical analysis; however, in a report
on NIH-sponsored meetings on this subject,
Heindel and vom Saal (2008) pointed out that
this is an insufficient control for total dietary
estrogenic contaminants that can disrupt stud-
ies involving the effects of estrogenic chemicals.
A second possibility is that there are
strain differences in sensitivity developed in
the CD-1 mouse sold by the various Charles
River Laboratories located in different regions.
We consider this unlikely, because most labo-
ratories regularly replace their CD-1 mouse
breeder stock from Charles River Laboratories,
and practices there make it unlikely that the
sensitivity of this outbred stock to estrogens
has changed dramatically over a very short
period of time. Also, because RTI, where the
Tyl studies were conducted, is very near the
laboratories of the NIEHS, it is likely that the
CD-1 mice used by these two programs were
purchased from the same breeding facility.
Use of insensitive, out-of-date protocols and
assays. Another serious concern about the two
recent studies by Tyl et al. (2008a, 2008b) is
the experimental approach used, thus raising
questions about the validity of the studies.
The study design used by Tyl et al. (2008a,
2008b) has been super seded by advances in
both experimental design and analytical tools
developed by NIH-funded scientists (and their
counterparts in Europe and Asia) since the
mid-1990s. The methods used by Tyl et al.,
primarily wet weight changes of tissues, gross
histologic changes, and developmental land-
marks such as vaginal opening, were estab-
lished procedures by the 1950s. Thus, a major
limitation of the Tyl studies is the failure to
measure more meaningful and sensitive end
points in order to detect the effects of low-dose
BPA exposure, which are often not macro-
scopic in nature. Indeed, in 2001, the director
of the reproductive division of the National
Health and Environmental Effects Research
Laboratory at the U.S. EPA stated that the
inconclusive results concerning effects of BPA
on reproductive toxicology can only be solved
by understanding the mechanisms (Triendl
2001). With current GLP standards it is not
possible to study mechanisms because they
still rely on out-of-date assays.
As one example of a comparison between
the approach by Tyl et al. (2008a) and inde-
pendent government-funded academic scien-
tists, extensive research has been conducted
by Soto et al. (2008) and by other indepen-
dent academic and government scientists
GLP is not a guarantee of reliable science
Environmental Health Perspectives • volume 117 | number 3 | March 2009
describing effects of exposure of female mice
and rats to very low doses of BPA during peri-
natal development on the mammary glands
(Jenkins et al. 2009). Although Tyl et al.
(2008a) reported no low-dose effects of BPA
on the mammary glands using conventional
histologic analysis, there have been consistent
findings of adverse effects of low doses of BPA
from studies that used more sophisticated and
sensitive analysis of whole mounted mam-
mary glands to facilitate detection of micro-
scopic lesions, coupled with immunos taining
for regulatory proteins as well as techniques
for determination of aberrant gene expression
associated with progression to cancer. These
peer-reviewed studies have reported detect-
ing changes during embryonic development
of mammary glands as well as abnormalities
detected during adolescence through adult-
hood that are indicative of mammary gland
cancer as well as other developmental abnor-
malities (Colerangle and Roy 1997; Durando
et al. 2007; Jenkins et al. 2009; LaPensee et al.
2008; Markey et al. 2001, 2005; Moral et al.
2008; Munoz-de-Toro et al. 2005; Murray
et al. 2007; Nikaido et al. 2004; Vandenberg
et al. 2006, 2007b; Wadia et al. 2007).
Similar to the findings for the mammary
gland, Ogura et al. (2007) reported that if
tissues were analyzed by conventional his-
tologic methods (staining with hematoxalin
and eosin), prenatal exposure to low doses of
BPA or DES showed no effects on prostate
development, whereas if the sections were
analyzed using antibodies that identified basal
cells and basal cell squamous metaplasia, then
significant effects were revealed. Squamous
metaplasia of basal cells indicates abnormal
proliferation and function of the prostate stem
cell population that is thought to transform
into neoplastic cells; Ho et al. (2006) reported
that neonatal exposure to very low doses of
BPA caused 100% of male rats to develop
high-grade prostatic intra epithelial neoplastic
lesions later in life. All of these studies were
rejected by the U.S. FDA as not adequate for
making regulatory decisions about the safety
of BPA. Instead, the U.S. FDA relied upon
Tyl et al. (2008a), even though the study used
techniques that Ogura et al. (2007) showed
lacked the sensitivity of 21st century experi-
Although findings regarding changes in
brain structure, brain chemistry, and behav-
ior represent the largest portion of the litera-
ture on low-dose BPA, Tyl et al. (2008a) did
not examine any neuro behavioral end points.
The NTP (2008) and the NIEHS confer-
ence consensus reports (vom Saal et al. 2007)
both indicated concern about neuro behavioral
effects of low doses of BPA. Thus, the absence
of studies that included neuro behavioral
end points is a glaring omission of Tyl et al.
Flawed prostate dissection. Data presented
by Tyl et al. (2008a) raise questions about the
adequacy of techniques used in their BPA stud-
ies. Specifically, Tyl et al. (2008a) reported that
the prostate in 3.5-month-old control male
CD-1 mice weighed > 70 mg [see Table 3
in Tyl et al. (2008a) for data on F1 retained
males]. This average control weight contrasts
sharply with those reported from other labo-
ratories. Specifically, the weight of the prostate
in 2- to 3-month-old CD-1 mice using the dis-
section technique based on both Ruhlen et al.
(2008) and Gupta (2000) and at the NIEHS
(Newbold RR, personal communication) is
about 40 mg. Several studies have reported that
prenatal exposure to very low doses of BPA and
positive control estrogens increased prostate
size, prostatic androgen receptors, and pros-
tate androgen receptor gene activity (Gupta
2000; Richter et al. 2007b; Thayer et al. 2001;
Timms et al. 2005; vom Saal et al. 1997), but
the enlarged prostate of experimental animals
exposed to BPA in these laboratories weighed
less than the prostates in the control animals of
Tyl et al. (2008a). This raises serious questions
about the procedures and/or animals used by
Tyl et al. The weight of prostate reported by
Tyl et al. (2008a) suggests that the technique
used for dissecting the prostate resulted in non-
prostatic tissue being weighed along with pros-
tate. The seminal vesicle, coagulating gland,
and dorso lateral prostate all merge together
where the ejaculatory ducts enter the urethra,
and there are also fat deposits on the prostate.
This poses a challenge for those without proper
training in distinguishing these different tissues
during dissection in mice.
Alternatively, as male rodents age, they
are prone to develop prostatitis. Although
this inflammatory disease leads to an increase
in prostate size and could thus account for
the very large prostate weights reported by
Tyl et al. (2008a), anyone familiar with the
appearance of prostatitis would detect this
abnormality upon histologic examination,
which Tyl et al (2008a) supposedly con-
ducted. Also, prostatitis is rare in young-adult
mice or rats (Cowin et al. 2008), and the size
of the prostates in the Tyl et al. (2008a) study
were similar to those for middle-aged and old
The findings regarding effects of BPA on
the prostate presented by Tyl et al. (2008a)
are thus suspect and cannot be used as evi-
dence that other earlier studies (Gupta 2000;
Timms et al. 2005; vom Saal et al. 1997) are
not replicable. Given these problems in pros-
tate weight measurements, it is not surprising
that even very high doses of BPA or estradiol
reported by Tyl et al. (2008a) had no effect on
the prostate, in sharp contrast to other studies
that showed stimulation of the prostate at low
doses of estrogen and inhibition at high doses
(Putz et al. 2001; Timms et al. 2005).
In addition to the problem associated
with the high prostate weight reported by Tyl
et al. (2008a), in a separate measurement the
authors combined the anterior prostate (coagu-
lating gland) and seminal vesicle, presenting
these two organs as one combined outcome
meas ure. This is wrong and misleading. The
coagulating glands emerge as the anterior ducts
of the prostate from the dorso cranial region
of the uro genital sinus, whereas the seminal
vesicles bud from the proximal region of the
Wolffian ducts. Elevated estrogen is associ-
ated with an increase in prostate size associated
with an increase in prostate androgen recep-
tors, whereas a decrease in seminal vesicle size
is associated with a reduction in 5α-reductase,
an enzyme that converts testosterone to the
more potent androgen 5α-dihydrotestosterone
(Nonneman et al. 1992). Low doses of BPA
have been shown to decrease the size of organs
that differentiate from the embryonic Wolffian
ducts (epididymides and seminal vesicles)
while increasing the size of regions of the pros-
tate that develop from the uro genital sinus
(vom Saal et al. 1998). Combining these dif-
ferent organs (it is technically not difficult to
separate them) was thus inappropriate because
they develop from different embryonic tis-
sues that show markedly different responses to
estrogenic chemicals during development. In
fact, Ogura et al. (2007) reported that the ante-
rior prostate (coagulating glands) showed the
greatest expression of ER-α, and also showed
the most pronounced indication of basal cell
squamous metaplasia in response to develop-
mental exposure to low doses of DES and BPA
relative to other regions of the prostate.
Because the control data of Tyl et al. (2008a)
were not consistent with the prior published
literature for prostate weight of young-adult
CD-1 male mice and because their methods
were inappropriate for revealing an extensive
body of adverse effects detected using more
sophisticated approaches, we deem the find-
ings by Tyl et al. to be invalid. Hundreds of
studies show adverse effects of BPA in ani-
mals, with many conducted at concentrations
equivalent to current human levels of BPA
exposure; thus, it is unlikely that academic sci-
entists would bother to replicate the outdated
approaches used by Tyl et al. (2008a, 2008b).
This lack of replication is typical of GLP stud-
ies, which tend to involve unnecessarily large
numbers of animals [Tyl et al. (2002) used
> 8,000 rats], and reliability appears to be
accepted because of the numbers of animals
that were used. Although using excessive
numbers of animals is accepted as good sci-
ence by the U.S. FDA, the use of arbitrarily
large numbers of animals per group (> 20 ani-
mals per treatment group is common) actually
violates guidelines in the NIH Guide for the
Myers et al.
volume 117 | number 3 | March 2009 • Environmental Health Perspectives
Care and Use of Laboratory Animals (Institute
of Laboratory Animal Research 1996) that
govern research conducted by academic and
government scientists. For research with ani-
mals to be approved by any university animal
care and use committee, group sizes must be
based on power analysis conducted using his-
toric data. Based on this criterion in the NIH
Guide, all of the studies by Tyl et al. were
significantly over powered and thus in direct
violation of federal guidelines for conducting
animal research, a fact about which U.S. FDA
regulators seem unaware.
Each of the four main industry-funded
GLP studies of BPA (Ashby et al. 1999; Cagen
et al. 1999; Tyl et al. 2008a, 2008b) is flawed
and not appropriate for use in setting health
standards. Clearly, meeting GLP standards is
not a guarantee of reliable or valid science. It is
of great concern that the U.S. and EU regula-
tory communities are willing to accept these
industry-funded, antiquated, and flawed stud-
ies as proof of the safety of BPA while rejecting
as invalid for regulatory purposes the findings
from a very large number of academic and gov-
ernment investigators using 21st- century scien-
tific approaches. The basis for these decisions
by U.S. and EU regu la tory agencies should
be thoroughly investigated, particularly since
the NTP (2008) concluded that BPA expo-
sure to human infants was in the range shown
to cause harm in experimental animals and
since both the Canadian Ministry of Health
and the Ministry of the Environment recently
concluded that BPA was a toxic chemical
(Environment Canada 2008).
Problems inherent with reliance on GLP
as the standard for choosing data are com-
pounded by the process used by federal agen-
cies to determine membership on science
advisory panels. Leading experts qualified by
specific experience on the chemical or end
points under consideration are often specifi-
cally excluded from membership. For example,
the U.S. FDA’s BPA review panel was identi-
fied as an expert panel, when in fact the panel
was composed largely of scientists lacking any
experience in research with BPA. This process,
which appears to consider almost any scientist
knowledgeable about a chemical to create bias,
makes it vastly more difficult for the panel
to integrate scientific data from the relevant
literature, especially since, as with BPA, there
are almost 1,000 relevant studies and the
review panel is provided with very little time
to become knowledgeable about the details.
It means that the depth of knowledge pres-
ent on this and similarly constituted govern-
ment regulatory agency panels is unlikely to
be sufficient to subject draft assessments to the
scrutiny that peer review by experts normally
entails. Combined with reliance on GLP data,
this process has a high potential to yield flawed
assessments that jeopardize public health.
We are not suggesting that GLP should
be abandoned as a requirement for industry-
funded studies. We object, however, to regu-
latory agencies implying that GLP indicates
that industry-funded GLP research is some-
how superior to NIH-funded studies that are
not conducted using GLP. This argument
demonstrates a lack of understanding of the
profound difference between the use of repli-
cation as a mechanism to assess reliability and
the methods used to assess validity for peer-
reviewed published academic studies, whereas
GLP was instituted with the expectation that
this type of verification would not occur.
Public health decisions should be based
on studies using appropriate protocols and
the most sensitive assays. They should not be
based on criteria that include or exclude data
depending on whether or not the studies use
GLP. Simply meeting GLP requirements is
insufficient to guarantee scientific reliability
Alonso-Magdalena P, Laribi O, Ropero AB, Fuentes E, Ripoll C,
Soria B, et al. 2005. Low doses of bisphenol A and diethyl-
stil bestrol impair Ca2+ signals in pancreatic α cells through
a nonclassical membrane estrogen receptor within intact
islets of Langerhans. Environ Health Perspect 113:969–977.
Alonso-Magdalena P, Morimoto S, Ripoll C, Fuentes E, Nadal A.
2006. The estrogenic effect of bisphenol A disrupts pancrea-
tic β-cell function in vivo and induces insulin resistance.
Environ Health Perspect 114:106–112.
Alonso-Magdalena P, Ropero AB, Carrera MP, Cederroth CR,
Baquie M, Gauthier BR, et al. 2008. Pancreatic insulin
content regulation by the estrogen receptor ERα. PLoS
ONE 3(4):e2069; doi:10.1371/journal.pone.0002069 [Online
30 April 2008].
Ashby J, Tinwell H, Haseman J. 1999. Lack of effects for low
dose levels of bisphenol A (BPA) and diethylstilbestrol
(DES) on the prostate gland of CF1 mice exposed in utero.
Regul Toxicol Pharmacol 30:156–166.
Cagen SZ, Waechter JM, Dimond SS, Breslin WJ, Butala
JH, Jekat FW, et al. 1999. Normal reproductive organ
development in CF-1 mice following prenatal exposure to
bisphenol A. Toxicol Sci 11:15–29.
CERHR (Center for the Evaluation of Risks to Human
Reproduction). 2007. NTP-CERHR Expert Panel Report
on the Reproductive and Developmental Toxicity of
Bisphenol A. NTP-CERHR-BPA-07. Available: http://cerhr.
pdf [accessed 28 July 2008].
Colerangle JB, Roy D. 1997. Profound effects of the weak
environmental estrogen-like chemical bisphenol A on the
growth of the mammary gland of Noble rats. J Steroid
Biochem Mol Biol 60(1–2):153–160.
Cowin PA, Foster P, Pedersen J, Hedwards S, McPherson SJ,
Risbridger GP. 2008. Early-onset endocrine disruptor-
induced prostatitis in the rat. Environ Health Perspect
Crain DA, Eriksen M, Iguchi T, Jobling S, Laufer H, LeBlanc GA,
et al. 2007. An ecological assessment of bisphenol-A:
evidence from comparative biology. Reprod Toxicol
Durando M, Kass L, Piva J, Sonnenschein C, Soto AM, Luque EH,
et al. 2007. Prenatal bisphenol A exposure induces pre-
neoplastic lesions in the mammary gland in Wistar rats.
Environ Health Perspect 115:80–86.
EFSA (European Food Safety Authority). 2006. Opinion of the
Scientific Panel on Food Additives, Flavourings, Processing
Aids and Materials in Contact with Food on a request
from the commission related to 2,2-bis(4-hydroxyphenyl)
propane (bisphenol A). EFSA J 428:1–75. Available: http://
78620772817.htm [accessed 23 January 2009].
Environment Canada. 2008. Screening Assessment for
the Challenge Phenol, 4,4’-(1-Methylethylidene)bis-
(Bisphenol A), Chemical Abstracts Service Registry Number
80-05-7. Available: http://www.ec.gc.ca/substances/ese/
15 August 2008].
Food Quality Protection Act of 1996. 1996. Public Law 104-170.
Golafshani N. 2003. Understanding reliability and validity in
qualitative research. Qualitative Rep 8(4):597–607.
Goldman D. 1988. Chemical aspects of compliance with Good
Laboratory Practices. In: Good Laboratory Practices: An
Agrochemical Perspective (Garner WY, Barge SB, eds).
Washington DC: American Chemical Society, 13–23.
Gupta C. 2000. Reproductive malformation of the male offspring
following maternal exposure to estrogenic chemicals.
Proc Soc Exp Biol Med 224(2):61–68.
Heindel JJ, vom Saal FS. 2008. Meeting report: batch-to-batch
variability in estrogenic activity in commercial animal
diets—importance and approaches for laboratory animal
research. Environ Health Perspect 116:389–393.
Ho SM, Tang WY, Belmonte de Frausto J, Prins GS. 2006.
Developmental exposure to estradiol and bisphenol A
increases susceptibility to prostate carcinogenesis and
epigenetically regulates phosphodiesterase type 4 variant 4.
Cancer Res 66(11):5624–5632.
Honma S, Suzuki A, Buchanan DL, Katsu Y, Watanabe H,
Iguchi T. 2002. Low dose effect of in utero exposure to
bisphenol A and diethylstilbestrol on female mouse repro-
duction. Reprod Toxicol 16:117–122.
Hugo ER, Brandebourg TD, Woo JG, Loftus J, Alexander JW,
Ben-Jonathan N. 2008. Bisphenol A at environmentally rele-
vant doses inhibits adiponectin release from human adipose
tissue explants and adipocytes. Environ Health Perspect
Institute of Laboratory Animal Research. 1996. Guide for
the Care and Use of Laboratory Animals. Washington,
DC:National Academy Press.
Jenkins S, Raghuraman N, Eltoum I, Carpenter M, Russo J,
Lamartiniere CA. 2009. Oral Exposure to Bisphenol A
Increases Dimethylbenzanthracene-Induced Mammary
Cancer in Rats. Environ Health Perspect doi:10.1289/
ehp.11751 [Online 7 January 2009].
Keri RA, Ho SM, Hunt PA, Knudsen KE, Soto AM, Prins GS. 2007.
An evaluation of evidence for the carcinogenic activity of
bisphenol A. Reprod Toxicol 24(2):240–252.
Lang IA, Galloway TS, Scarlett A, Henley WE, Depledge M,
Wallace RB, et al. 2008. Association of urinary bisphenol
A concentration with medical disorders and laboratory
abnormalities in adults. JAMA 300(11):1303–1310.
LaPensee EW, Tuttle TR, Fox SR, Ben-Jonathan N. 2009.
Bisphenol A at low nanomolar doses confers chemo-
resistance in estrogen receptor-α–positive and –negative
breast cancer cells. Environ Health Perspect 117:175–180.
Leranth C, Hajszan T, Szigeti-Buck K, Bober J, MacLusky NJ. 2008.
Bisphenol A prevents the synaptogenic response to estradiol
in hippocampus and prefrontal cortex of ovariectomized non-
human primates. Proc Nat Acad Sci 105(37):14187–14191.
Lublin JS. 1978. Safety problems. Wall Street Journal (New
York) 21 February: A1.
Markey CM, Luque EH, Munoz De Toro M, Sonnenschein C,
Soto AM. 2001. In utero exposure to bisphenol A alters the
development and tissue organization of the mouse mam-
mary gland. Biol Reprod 65(4):1215–1223. [Erratum in Biol
Reprod 71:1753 (2004) reported that the actual doses used
were 25 and 250 ng/kg/day].
Markey CM, Wadia PR, Rubin BS, Sonnenschein C, Soto AM.
2005. Long-term effects of fetal exposure to low doses of
the xenoestrogen bisphenol-A in the female mouse genital
tract. Biol Reprod 72(6):1344–1351.
Markowitz GE, Rosner D. 2002. Deceit and Denial: The Deadly
Politics of Industrial Revolution. Berkeley, CA:University of
Moral R, Wang R, Russo IH, Lamartiniere CA, Pereira J,
Russo J. 2008. Effect of prenatal exposure to the endocrine
disruptor bisphenol A on mammary gland morphology and
gene expression signature. J Endocrinol 196(1):101–112.
Munoz-de-Toro M, Markey CM, Wadia PR, Luque EH, Rubin BS,
Sonnenschein C, et al. 2005. Perinatal exposure to bisphe-
nol-A alters peripubertal mammary gland development in
mice. Endocrinology 146(9):4138–4147.
Murray TJ, Maffini MV, Ucci AA, Sonnenschein C, Soto AM.
2007. Induction of mammary gland ductal hyperplasias and
carcinoma in situ following fetal bisphenol A exposure.
Reprod Toxicol 23(3):383–390.
GLP is not a guarantee of reliable science Download full-text
Environmental Health Perspectives • volume 117 | number 3 | March 2009
Nagel SC, vom Saal FS, Thayer KA, Dhar MG, Boechler M,
Welshons WV. 1997. Relative binding affinity-serum modified
access (RBA-SMA) assay predicts the relative in vivo bio-
activity of the xenoestrogens bisphenol A and octylphenol.
Environ Health Perspect 105:70–76.
Newbold R. 1995. Cellular and molecular effects of develop-
mental exposure to diethylstilbestrol: implications for
other environmental estrogens. Environ Health Perspect
Newbold RR, Jefferson WN, Padilla-Banks E. 2007. Long-term
adverse effects of neonatal exposure to bisphenol A on
the murine female reproductive tract. Reprod Toxicol
Nikaido Y, Yoshizawa K, Danbara N, Tsujita-Kyutoku M, Yuri T,
Uehara N, et al. 2004. Effects of maternal xenoestrogen
exposure on development of the reproductive tract and
mammary gland in female CD-1 mouse offspring. Reprod
Nonneman DJ, Ganjam VK, Welshons WV, Vom Saal FS. 1992.
Intrauterine position effects on steroid metabolism and
steroid receptors of reproductive organs in male mice.
Biol Reprod 47(5):723–729.
NTP (National Toxicology Program). 2001. National Toxicology
Program’s Report of the Endocrine Disruptors Low
Dose Peer Review. Available: http://ntp.niehs.nih.gov/
23 January 2009].
NTP (National Toxicology Program). 2008. NTP-CERHR
Monograph on the Potential Human Reproductive and
Developmental Effects of Bisphenol A. Available: http://
[accessed 28 January 2009].
Ogura Y, Ishii K, Kanda H, Kanai M, Arima K, Wang Y, et al.
2007. Bisphenol A induces permanent squamous change in
mouse prostatic epithelium. Differentiation 75(8):745–756.
Putz O, Schwartz CB, Kim S, LeBlanc GA, Cooper RL, Prins GS.
2001. Neonatal low-and high-dose exposure to estradiol
benzoate in the male rat: 1. Effects on the prostate gland.
Biol Reprod 65:1496–1505.
Richter CA, Birnbaum LS, Farabollini F, Newbold RR, Rubin BS,
Talsness CE, et al. 2007a. In vivo effects of bisphenol A in
laboratory rodent studies. Reprod Toxicol 24(2):199–224.
Richter CA, Taylor JA, Ruhlen RR, Welshons WV, vom Saal FS.
2007b. Estradiol and bisphenol A stimulate androgen recep-
tor and estrogen receptor gene expression in fetal mouse
prostate cells. Environ Health Perspect 115:902–908.
Ropero AB, Alonso-Magdalena P, Garcia-Garcia E, Ripoll C,
Fuentes E, Nadal A. 2008. Bisphenol-A disruption of the
endocrine pancreas and blood glucose homeostasis. Int J
Ruhlen RL, Howdeshell, KL, Mao J, Taylor JA, Bronson FH,
Newbold RR, et al. 2008. Low phytoestrogen levels in feed
increase fetal serum estradiol resulting in the “fetal estro-
genization syndrome” and obesity in CD-1 mice. Environ
Health Perspect 116:322–328.
Sheehan DM. 2000. Activity of environmentally relevant low doses
of endocrine disruptors and the bisphenol A controversy: ini-
tial results confirmed. Proc Soc Exp Biol Med 224(2):57–60.
Soto AM, Vandenberg LN, Maffini MV, Sonnenschein C.
2008. Does breast cancer start in the womb? Basic Clin
Pharmacol Toxicol 102(2):125–133.
Susiarjo M, Hassold TJ, Freeman E, Hunt PA. 2007. Bisphenol A
exposure in utero disrupts early oogenesis in the mouse.
PLoS Genet 3(1):63–70.
Thayer KA, Ruhlen RL, Howdeshell KL, Buchanan DL, Cooke PS,
Preziosi D, et al. 2001. Altered prostate growth and daily
sperm production in male mice exposed prenatally to sub-
clinical doses of 17alpha-ethinyl oestradiol. 16(5):988–996.
Thigpen JE, Haseman JK, Saunders HE, Setchell KDR, Grant MG,
Forsythe DB. 2003. Dietary phytoestrogens accelerate the
time of vaginal opening in immature CD-1 mice. Comp Med
Timms BG, Howdeshell KL, Barton L, Bradley S, Richter CA,
vom Saal FS. 2005. Estrogenic chemicals in plastic and oral
contraceptives disrupt development of the mouse prostate
and urethra. Proc Natl Acad Sci U S A 102:7014–7019.
Triendl R. 2001. Genes may solve hormone-disrupter debate.
Tyl RW, Myers C, Marr M, Sloan CS, Castillo N, Veselica MM,
et al. 2008a. Two-generation reproductive toxicity study of
dietary bisphenol A (BPA) in CD-1 (Swiss) mice. Toxicol
Tyl R, Myers C, Marr M, Sloan CS, Castillo N, Veselica MM, et al.
2008b. Two-generation reproductive toxicity evaluation of
dietary 17β-estradiol (E2; CAS No. 50-28-2) in CD-1 (Swiss)
mice. Toxicol Sci 102(2):392–412.
Tyl RW, Myers CB, Marr MC, Thomas BF, Keimowitz AR,
Brine DR, et al. 2002. Three-generation reproductive toxicity
study of dietary bisphenol A in CD Sprague-Dawley rats.
Toxicol Sci 68(1):121–146.
U.S. EPA. 2008. Good Laboratory Practices Standards.
programs/fifra/glp.html [accessed 15 August 2007].
U.S. FDA. 2008a. Draft assessment of bisphenol A for use in food
contact applications. Available: http://www.fda.gov/ohrms/
Draft%20Assessment.pdf [accessed 14 August 2008].
U.S. FDA. 2008b. Statement of Norris Alderson, Ph.D., Associate
Commissioner for Science, Food and Drug Administration,
Department of Health and Human Services, before the
Subcommittee on Commerce, Trade and Consumer
Protection, Committee on Energy and Commerce. U.S. House
of Representatives. June 10, 2008. Available: http://www.fda.
gov/ola/2008/BPA061008.html [accessed 23 January 2009].
Vandenberg LN, Hauser R, Marcus M, Olea N, Welshons WV.
2007a. Human exposure to bisphenol A (BPA). Reprod
Vandenberg LN, Maffini MV, Wadia PR, Sonnenschein C,
Rubin BS, Soto AM. 2007b. Exposure to environmen-
tally rele vant doses of the xenoestrogen bisphenol-A
alters development of the fetal mouse mammary gland.
Vandenberg LN, Wadia PR, Schaeberle CM, Rubin BS,
Sonnenschein C, Soto AM. 2006. The mammary gland
response to estradiol: monotonic at the cellular level, non-
monotonic at the tissue-level of organization? J Steroid
Biochem Mol Biol 101(4–5):263–274.
vom Saal FS. 2008. Bisphenol A: update of current published
studies. Available: http://endocrinedisruptors.missouri.
edu/vomsaal/vomsaal.html [accessed 31 July 2008].
vom Saal FS, Akingbemi BT, Belcher SM, Birnbaum LS, Crain DA,
Eriksen M, et al. 2007. Chapel Hill bisphenol A expert panel
consensus statement: integration of mechanisms, effects
in animals and potential to impact human health at current
levels of exposure. Reprod Toxicol 24(2):131–138.
vom Saal FS, Cooke PS, Buchanan DL, Palanza P, Thayer KA,
Nagel SC, et al. 1998. A physiologically based approach to
the study of bisphenol A and other estrogenic chemicals
on the size of reproductive organs, daily sperm production,
and behavior. Toxicol Ind Health 14(1–2):239–260.
vom Saal FS, Hughes C. 2005. An extensive new literature
concerning low-dose effects of bisphenol A shows the
need for a new risk assessment. Environ Health Perspect
vom Saal FS, Timms BG, Montano MM, Palanza P, Thayer KA,
Nagel SC, et al. 1997. Prostate enlargement in mice due to
fetal exposure to low doses of estradiol or diethylstil bestrol
and opposite effects at high doses. Proc Nat Acad Sci USA
vom Saal FS, Welshons WV. 2006. Large effects from small
exposures. II. The importance of positive controls in low-
dose research on bisphenol A. Environ Res 100:50–76.
Wadia PR, Vandenberg LN, Schaeberle CM, Rubin BS,
Sonnenschein C, Soto AM. 2007. Perinatal bisphenol A
exposure increases estrogen sensitivity of the mammary
gland in diverse mouse strains. Environ Health Perspect
Welshons WV, Nagel SC, vom Saal FS. 2006. Large effects
from small exposures. III. Endocrine mechanisms mediat-
ing effects of bisphenol A at levels of human exposure.
Endocrinology 147(suppl 6):S56–S69.
Wetherill YB, Akingbemi BT, Kanno J, McLachlan JA, Nadal A,
Sonnenschein C, et al. 2007. In vitro molecular mecha-
nisms of bisphenol A action. Reprod Toxicol 24(2):178–198.
Wozniak AL, Bulayeva NN, Watson CS. 2005. Xenoestrogens at
picomolar to nanomolar concentrations trigger membrane
estrogen receptor-alpha-mediated Ca2+ fluxes and prolac-
tin release in GH3/B6 pituitary tumor cells. Environ Health
Zsarnovszky A, Le HH, Wang HS, Belcher SM. 2005. Ontogeny
of rapid estrogen-mediated extracellular signal-regulated
kinase signaling in the rat cerebellar cortex: potent non-
genomic agonist and endocrine disrupting activity of the
xeno estrogen bisphenol A. Endocrinology 146(12):5388–5396.