   SEQ CHAPTER \h \r 1 UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

					WASHINGTON, D.C.  20460

OFFICE OF              

PREVENTION, PESTICIDES

AND TOXIC SUBSTANCES 

TXR# 0054554

Date:		April 2, 2007

MEMORANDUM

SUBJECT:	Endosulfan.  Hazard Characterization and Endpoint Selection
Reflecting Receipt of a Developmental Neurotoxicity Study and Subchronic
Neurotoxicity Study. 

		PC Code: 079401.  

		DP Barcode D338576.

FROM:	Elissa Reaves, Ph.D., Toxicologist

			Reregistration Branch 2

		Health Effects Division (7509P)

THRU:	William J. Hazel, Ph.D., Chief

		Reregistration Branch 2

		Health Effects Division (7509P)

TO:		Tracy Perry, Chemical Review Manager

		Special Review Branch

		Special Review and Reregistration Division (7508P)

This memo outlines the rationale and characterization for endpoint
selection for 

the endosulfan risk assessment following the review of a developmental 

neurotoxicity study (MRID 46968301) and subchronic neurotoxicity study
(MRID 

46444401).

The 5/30/03 endosulfan risk assessment (D250471) retained the 10X FQPA
Safety Factor for the following reasons:

Data gaps of the subchronic neurotoxicity and developmental
neurotoxicity studies, 

Concerns for endosulfan-induced toxicity to the male reproductive system
(e.g., effects on sperm parameters) reported in the published
literature,

Lack of evaluation of sperm parameters in the guideline studies, 

Testicular lesions seen in the chronic rat study, 

Increased pituitary and uterine weight in the two-generation
reproduction study 

Each of these concerns is addressed below.  

Fulfillment of Data Gaps.  Since this 5/30/03 evaluation, OPP has
received and evaluated both a subchronic neurotoxicity and a
developmental neurotoxicity study, thus fulfilling the data gaps
identified in the previous risk assessment.  

In a subchronic neurotoxicity study, endosulfan (96.55-98.1% a.i) was
administered via the diet to 12 Wistar, Hanover Crl:Wl[GIx/BRL/Han]IGS,
rats/sex/group at dose levels of 0, 40, 225 or 600 ppm (equivalent to
Males: 0, 2.11, 13,7, 37.2 mg /kg/day; Females: 0, 2.88, 16.6, 45.5 mg
a.i./kg/day) for 13 weeks.  Neurobehavioral assessment (functional
observational battery and motor activity testing) was performed on12
animals/sex/group at pretreatment, and weeks 4, 8 and 13. No dose
related or treatment effects were seen in clinical observations, FOB,
motor activity, locomotor activity, measured grip strength or foot
splay, and there was no evidence of neuropathology (D. Anderson,
4/18/05, TXR # 0053078, D312303, MRID 46444401). 

In the developmental neurotoxicity study, Wistar rats (30/group) were
administered endosulfan (99.1% pure) in the diet at dose levels of 0,
50, 150, or 400 ppm (equivalent to 0, 3.74, 10.8 or 29.8 mg/kg/day) from
gestation day 6 through postnatal day 21; doses were adjusted to nearly
constant mg/kg/day throughout lactation. The maternal NOAEL was 10.8
mg/kg/day and the LOAEL was 29.8 mg/kg/day based on decreases in body
weight, food consumption and food efficiency. In the offspring toxicity
manifested as decreased body weight in males (8.3-13.2%) and females
(8.0-13.7%) at all dose levels on PND 11 and in males only (6.9-8.8%) on
PND 17.  The male and female pup weight decrement at all dose levels on
PND 11 was reflected in a decrement in the pup weight gain from PND
4-11.  There were no treatment-related effects on offspring survival,
clinical signs, sexual maturation, FOB, motor activity, auditory startle
response, learning and memory, neuropathology, brain weight or brain
morphometrics at any dose level. 

In addition, because of concern about the developing male reproductive
system as reported in the literature (see below), EPA required that
sperm parameters be measured in young adult males in the DNT study. 
Sperm parameters were evaluated in male offspring at terminal sacrifice
(day 75 +/- 5).  No treatment-related effects were seen in testes or
epididymis weights or in any of the sperm parameters (i.e., sperm and
spermatid numbers, sperm morphology) evaluated nor were there any
histopathological lesions in any of the male reproductive organs. The
offspring LOAEL was 3.74 based on decreased pup weight on PND11 and
weight gain on PND 4-11; an offspring NOAEL was not established (D.
Anderson, 3/15/07, TXR # 0054486, D327215, MRID 46968301).

Concerns for public literature reports on endosulfan-induced toxicity to
the male reproductive system.  In the 2003 risk assessment, studies by
Dalsenter et al. (1999) and Sinha et al. (2001) raised concerns for
toxicity to the male offspring of rats exposed to endosulfan.  However,
these concerns no longer exist due to a more recent study published by
Dalsenter et al., (2003) which does not show adverse effects in the male
offspring of dams following pre and post natal exposures to low doses of
endosulfan. These studies are discussed below.

Sinha et al. (2001) gave oral intubation of endosulfan (95% pure) at 1
or 2 mg/kg/day in peanut oil to Druckery rats from gestation day 12
through parturition (the approximate time of gonadal differentiation in
the rat). At birth, the offspring were fostered to naïve (i.e.,
untreated) dams. Sperm parameters were evaluated on postnatal day 100
(i.e., young adults).  Treatment at both dose levels caused reductions
of sperm and spermatid counts, decreases in testes, seminal vesicle, 
and epididymis weights.  In addition the following effects were observed
on two testicular enzyme markers: an increase in lactate dehydrogenase
(LDH), a decrease in sorbitol dehydrogenase (SDH) activities.

Dalsenter et al (1999) investigated the effects of endosulfan in the
offspring of male Wistar rats following oral administration of 0, 1.5,
or 3.0 mg/kg/day of endosulfan (97% pure) in Tween 80 to dams (24/group)
from day 15 of pregnancy to postnatal day 21 of lactation. Male
offspring rats were examined on postnatal days 65 and 140, which
corresponded to the pubertal and adult stages of development.  Maternal
toxicity, at the high dose, was manifested as decreased body weight
during pregnancy, but litter size and mean birth weights were not
affected. Offspring toxicity, at the high dose (3 mg/kg/day), was
characterized as a decrease in both daily sperm production and the
percentage of seminiferous tubules with complete spermatogenesis in male
offspring, both at puberty and adulthood. At 1.5 mg/kg/day, daily sperm
production was significantly reduced only at puberty.  No
treatment-related effects were seen on developmental landmarks or on the
weight of reproductive sex organs at either dose level.

In a follow-up study, Dalsenter et al (2003) examined the effects of low
doses of endosulfan (97% pure) on the reproductive system of male
offspring Wistar rats following prolonged (pre and postnatal) exposures.
 Dams (27/group) were given oral administration of endosulfan in
sunflower oil at 0.5 or 1.5 mg/kg/day for 21 days prior to mating,
during mating, pregnancy and lactation.  Reproductive endpoints measured
at adulthood (post natal day 140) included: sex organ weights, daily
sperm production, spermatid number, sperm transit, sperm morphology, and
testosterone level.  No maternal toxicity was seen at either dose.  No
adverse effects were seen in sperm production; sperm count in cauda
epididymis, sperm transit and sperm morphology and serum testosterone
concentrations.  These results show that pre and postnatal exposure to
low doses of endosulfan do not induce significant adverse effects in the
reproductive system of male offspring Wistar rats at adulthood. 

Results of these studies indicate that there may be differences in
susceptibility of the male reproductive system to endosulfan depending
on:

the strain of rats (Wistar vs. Druckrey); including the sperm count
difference (both Dalsenter et al and Sinha used the same method [Robb et
al., 1978 as reported in both articles] to determine sperm count, yet
the Druckrey rats of Sinha have a 38% lower control value than the
Wistar rats of Dalsenter)

exposure duration (gestation period vs. pre and postnatal),  and 

the purity of endosulfan used (95% with Druckrey rats vs. 97% with
Wistar rats).  

Lack of measurement of male reproductive parameters in guideline
studies. As mentioned above in the DNT discussion, because of concern
about the developing male reproductive system as reported in the
literature, EPA required that sperm parameters be measured in young
adult males in the DNT study.  No treatment-related effects were seen in
testes or epididymis weights or in any of the sperm parameters (i.e.,
sperm and spermatid numbers, sperm morphology) evaluated nor were there
any histopathological lesions in any of the male reproductive organs. 
The DNT study design required exposure to pregnant dams from gestation
day 6 through lactation day 21.  Therefore, male pups were exposed in
utero and for their entire weaning period.  Lack of sperm effects at
post-pubertal ages (approximately PND 70-80) at the high dose level of
29.8 mg/kg/day reduces the uncertainty for this concern.

Testicular lesions in the early chronic rat study.  There is low
confidence in the testicular atrophy observed in male rats in a 1978
toxicity study (MRID No. 00004256) since these lesions were seen at high
doses (20.4 and 48 mg/kg/day; the only doses tested) and were not
replicated in another chronic study (1989) following dietary
administration of endosulfan to Sprague-Dawley rats at the low dose
levels of 0.1, 0.4, 0.7 or 3.8 mg/kg/day for 104 weeks.  In addition,
the lack of sperm effects in the DNT study at PND 70-80 at the dose of
29.8 mg/kg/day needs to be taken into account.  Finally, the 1978 study
used another strain of rat (Osborne-Mendel) which may also explain some
differences in the data.

Pituitary and uterine weight effects in two-generation reproductive
study. There is no concern for the increased relative (to body weight)
pituitary weights seen in the F0 generation (seen in at the first mating
measurement but not at the second mating measurement) and increased
(absolute) uterine weights seen in the F1b generation (again, seen at
the first mating time point but not at the second mating time point) in
the two-generation study because they were seen only at the highest dose
tested (6.2 mg/kg/day).  This is approximately 10-fold higher than the
dose (0.6 mg/kg/day) used for deriving the chronic Reference Dose. 
Also, in both cases, there is no dose response for these relative
(pituitary) and absolute (uterine) organ weight changes

Conclusion regarding the retention of the FQPA 10X factor. The above
data demonstrate that the uncertainty associated with the data gap has
been addressed with the submission of the requested studies (subchronic
neurotoxicity and developmental neurotoxicity studies). The concern for
endosulfan-induced adverse effects in male offspring as assessed in a
guideline study is also alleviated since no adverse effects in sperm
parameters, testes weights, histopathology of the testes were seen in
the DNT study at doses up to 29.8 mg/kg/day.  

In terms of the sperm effects observed in the open literature, the two
studies performed by Dalsenter et al., (1999, 2003) demonstrate a NOAEL
of 0.5 mg/kg/day and possible LOAEL of 1.5 mg/kg/day for these effects. 
Importantly, these studies used Wistar rats (a common strain in
toxicology testing and the one used in the DNT), a long exposure period
(in utero and during lactation), and a 97% pure test material. In
contrast, the Sinha et al., study which showed effects at the 1 m/kg/day
used a less familiar or commonly used strain, (Druckery), less pure test
material (95.32%), a shorter exposure duration (gestation day 12 to
parturition), and peanut oil as the vehicle.   Also, in these studies,
the mode of administration was oral (gavage) intubation where as the
mode of administration was dietary in the DNT.

Additionally, there was no quantitative or qualitative evidence of
increased susceptibility following in utero exposures of rats or rabbits
to endosulfan during gestation. In the two generation reproduction
study, there was evidence for qualitative increase in susceptibility;
increased ovarian weights were seen in the female offspring compared to
body weight decreases in the parental animals.  However, there is low
concern for this finding since the dose (0.6 mg/kg/day) used for chronic
dietary risk assessment is approximately 10-fold lower than the dose
that caused these effects and would address these concerns.

Based on the above discussed data, there is no need for the FQPA Safety
Factor (i.e., 1X) since there are no residual uncertainties for pre
and/or post natal toxicity.

Acute RfD:  NOAEL = 1.5 mg/kg/day – Acute Neurotoxicity study. Effects
of concern (clinical signs of neurotoxicity) were seen after a single
oral dose and thus appropriate for this exposure scenario.  Decreases in
body weight were seen in the offspring at the lowest dose tested (3.7
mg/kg/day) in the DNT. However, because the decrease in body weight is
the result of repeated exposure (not at birth or during post-weaning;
only during pre-weaning), this dose and endpoint cannot be used for this
exposure scenario. 

Chronic RfD: NOAEL = 0.6 mg/kg/day – chronic/carcinogenicity study in
rats.  The endpoint of concern (decrease in body weight, nephrotoxicity
and blood vessel aneurysm in male rats) was seen after long-term (104
weeks) exposure and is appropriate for this exposure scenario. 
Additionally, this dose is lower than the extrapolated NOAEL from the
DNT study (i.e., 3.7 ÷ UF3 = 1.2 mg/kg/day) and thus would protect the
decreased offspring body weight seen in the DNT.

DERMAL (Short-& Intermediate Term):   Previously, two available 21-day
dermal toxicity rat studies were the basis of quantifying dermal risk. 
The dermal NOAEL was 12 mg/kg/day and LOAEL of 48 mg/kg/day was based on
mortality.  However, the results of the recently submitted DNT shows
concern for offspring toxicity (decreased pre-weaning body weight) which
is not evaluated in the 21-day dermal study (conducted in adult animals
only).  Additionally, the DNT was examined to address the concern for
changes in the uterine and pituitary weights that were seen in adults at
the highest dose only (6.2 mg/kg/day) in the two-generation reproduction
study.

Consequently, the appropriateness of these three studies for dermal risk
assessments were considered as discussed below:

In the dermal studies, the NOAEL was 12 mg/kg/day and the LOAEL was 48
mg/kg/day;

In the two generation reproduction study, the NOAEL was 1.2 mg/kg/day
and the LOAEL was 6.2 mg/kg/day;

In the DNT, the LOAEL was 3.7 mg/kg/day, LDT; a NOAEL was not
established.  

In trying to effectively protect worker exposure to endosulfan the DNT
was deemed to be the most appropriate study.  This decision was based on
the fact that the DNT study evaluates the effect of concern, namely the
pup weight decrements due to nursing.  Since a NOAEL was not established
in the DNT, there is a need to estimate one to derive a point of
departure.  The standard uncertainty factors usually considered are
either 10x or 3x.  In this case, a 3x is most appropriate for the
following reasons:

the original data gaps identified (DNT and subchronic neurotoxicity
studies) have been filled;

the use of an offspring endpoint from the DNT study is the most
sensitive endpoint for the most sensitive population (female workers)
for regulatory purposes;

the pup weight decrements were only seen during lactation (i.e., the pup
body weights were not affected at birth AND the pups recovered after
post-weaning (> PND 22) (NOTE: this same phenomenon was observed in the
offspring from the two-generation study at a LOAEL of 6.2 mg/kg/day and
a NOAEL of 1.2 mg/kg/day); and 

use of the DNT study also addresses the concern for mortality seen via
the dermal route in the 21-day dermal study. 

Thus, the point of departure of 1.2 mg/kg/d (which is a reasonable
estimate for a NOAEL in the DNT study [3.7 mg/kg/d divided by a LOAEL to
NOAEL uncertainty factor of 3] and is similar to the NOAEL from the
two-generation study for the same effect (pup body weight loss during
lactation only). 

Because an oral value is used, a dermal absorption factor of 45% should
be used for route-to-route extrapolations.

A Margin of Exposure of 300 is appropriate since a LOAEL to NOAEL factor
is required for use in risk assessments. 

For the endosulfan occupational risk assessments the body weight is 60
kg.  Additionally, dermal and inhalation exposures should be combined
since both the rat DNT and 21-day rat inhalation studies showed body
weight decreases as an adverse effect.

INHALATION (Short-& Intermediate Term):  Inhalation NOAEL= 0.2 mg/kg/
day- 21 day inhalation toxicity study. The endpoints (decreased body
weight and alterations in hematology and clinical chemistry parameters)
are appropriate since they were seen after the route (inhalation) and
duration (21 days) of concern.  

A Margin of Exposure of 100 is adequate since a NOAEL is used for risk
assessments.   For occupational assessments with endosulfan, dermal and
inhalation exposures should be combined since both the rat DNT and
21-day rat inhalation studies showed body weight decreases as an adverse
effect.

Table 3.  Summary of Toxicology Endpoint Selection- Endosulfan

Exposure

Scenario	Dose Used in Risk Assessment, UF	Endpoint & Level of Concern
for Risk Assessment	Study and Toxicological Effects

Acute Dietary

(general population including infants and children)

	NOAEL = 1.5 mg/kg/day

UF = 100

FQPA=1x

	

aRfD = 0.015

mg/kg/day

aPAD = 0.015

mg/kg/day

	Acute neurotoxicity study in rats;

MRID 44403101

Oral LOAEL= 3 mg/kg/day; based on increased incidence of convulsions
seen in female rats within 8 hours after dosing.

Chronic Dietary

(all populations)	NOAEL = 0.6 mg/kg/day

UF = 100

FQPA=1x

	cRfD = 0.006

mg/kg/day

cPAD = 0.006

mg/kg/day

	Combined chronic/carcinogenicity study in rats; MRID 41099502

LOAEL = 2.9 mg/kg/day, based on reduced body weight gain, increased
incidences of marked progressive glomerulonephorsis & blood vessel
aneurysms in male rats.

Dermal, Short (1-30 days)- & Intermediate

(1-6 mos)-Term	LOAEL = 3.7 mg/kg/day

45% dermal absorption factor	Occupational LOC

MOE = 300	Developmental Neurotoxicity- rats: MRID 46968301

LOAEL = 3.74, based on decreased pup weight; NOAEL not established.

Dermal

Long-term

(> 6 months)	Long-term dermal exposure is not expected for endosulfan.

Inhalation

Short- and Intermediate-term

(1 - 30 days)	NOAEL = 0.2 mg/kg/day

 (0.001 mg/L)	LOC for Occupational MOE = 100	21-Day inhalation toxicity
study in male & female rats

MRID 00147183, 41667501

LOAEL = 0.002 mg/L, based on decreased body weight gain and alterations
in hematology and clinical chemistry parameters 

Inhalation

Long-term

(> 6 months)	Long-term inhalation exposure is not expected for
endosulfan.

Cancer	Group E- Evidence of non-carcinogenicity for humans	Q1* not
required	Chronic toxicity/carcinogenicity study in rats (MRID 41099502)
& carcinogenicity study in mice (MRID 40792401)

No increased in the frequency of tumors in either the rat or mouse.



 PAGE   

Page   PAGE  3  of   NUMPAGES  8 

