ve/EPA
United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Cincinnati OH 45268
Research and Development
EPA-600/S1-81-014 Mar 1981
Project Summary
Effect of Chlorine Dioxide,
Chlorite, and Nitrite on Mice
With Low and High Levels of
Glucose-6-Phosphate
Dehydrogenase (G6PD) in
Their Erythrocytes
Gary S. Moore and Edward J. Calabrese
Since chlorination of water supplies
has come under investigation as a
source of trihalomethane formation
and possible cancer production, alter-
native disinfecting substances are
being examined. Chlorine dioxide
(CIO2) is a leading possibility for an
alternative disinfectant. However, the
health consequences of ClOa or chlo-
rite (a product formed in the disinfec-
tion process) are little known. Further,
there is little or no information available
on possible interactive effects with
other oxidant compounds in the diet or
water, such as nitrites. The effect of
these compounds might be exagger-
ated on red blood cells deficient in
glucose-6-phosphate dehydrogenase
(G6PD). This study reports on these
findings.
This Project Summary was develop-
ed by EPA's Health Effects Research
Laboratory, Cincinnati, OH, to an-
nounce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
The recognition that many public
drinking water supplies in the United
States contain carcinogenic substances
has.generated substantial controversy
within both the scientific and lay press.
The controversy has focused on the
occurrence of carcinogenic substances
(i.e. trihalomethanes) which tend to be
formed in drinking water following the
process of chlorination. Epidemiological
studies have demonstrated statistical
associations between increased cancer
mortality and the practice of chlorina-
tion of drinking water. These findings
have tended to support laboratory studies
which demonstrated the occurrence of
liver cancer in selected rat and mouse
strains exposed to chloroform. For these
reasons, the EPA is considering alterna-
tive disinfectants to the process of
chlorination which would decrease the
trihalomethane level in drinking water.
In recognition of the potential wide-
spread utilization of chlorine dioxide as
the principal disinfectant in the United
States, the intention of this study is to
evaluate the health effects of chlorine
-------
dioxide and a by-product of CIO2 disin-
fection, sodium chlorite, on mice with
high and relatively low levels of G6PD
activity in their erythrocytes.
It was predicted that such an animal
would tend to simulate the response of a
possible human high risk group to these
stressor agents.
Chlorine Dioxide and Sodium
Chlorite
There are several methods by which
chlorine dioxide may be efficiently
produced prior to its application to the
drinking water. The two principal ways
by which chlorine dioxide is produced
includes the following:
a. From chlorine and sodium chlorite
CI2 + H2O-HOCI + HCI
HOCI + HCI + 2NaCI02-
2CI02 + NaCI + H20
b. From sodium hypochlorite and
sodium chlorite
NaOCI + HCI-NaCI + HOCI
HCI + HOCI + 2NaCI02-
2CI02 + 2NaCI + H20
From a health point of view it is
necessary to control the reaction stoi-
chiometry and thereby prevent the
release of unintended and unwanted
products in the drinking water, e.g. free
chlorine, chlorite, and chlorate.
Two recent EPA reports noted that
following chlorine dioxide disinfection
of surface waters, chlorites made up as
much as 50 percent of the chlorine
dioxide demand in the pH range of 4.8 to
9.75, while chlorates contributed from
10 to 30 percent of the chlorine dioxide
demand within the same pH range.
When 1.5 mg chlorine dioxide per liter
(mg/1) was added to coagulated, settled,
and filtered Ohio River water of 7.1 pH, a
chlorite concentration of 0.72 mg/1 and
chlorate concentration of 0.41 mg/1
resulted after 42 hours of contact.
Therefore, one should expect to find
varying quantities of chlorate and chlor-
ite in water which is disinfected with
chlorine dioxide. Thus, one must consi-
der the health effects of chlorites and
chlorates as well as chlorine dioxide.
Nitrite
Studies on nitrite have demonstrated
its ability to produce MetHb, especially
in neonates. Research has also indicated
that small quantities of nitrite can
generate MetHb auto-catalytically with-
out the presence of pre-existing MetHb.
This MetHb generated by nitrite can
then serve as a catalyst for the formation
of more MetHb by chlorite. Chlorite is
known to oxidize hemoglobin more
readily in the presence of MetHb.
Glucose-6-phosphate
Dehydrogenase (G6PD)
In attempting to establish minimal
health effect levels of oxidants on
erythrocytes, it is important to consider
the health effects of such stressors on
high risk groups. The two largest groups
that fall into this category are persons
with lowered G6PD activity and neonates.
G6PD deficient cells have a reduced
ability to produce NADPH via the pentose
phosphate pathway (PPP) and conse-
quently less GSH is formed. Since GSH
is the primary mechanism of defense of
the red blood cell against oxidant stress,
then persons with deficient G6PD levels
have a lowered capacity for protection
against oxidants. Neonates have a
variety of deficiencies and differences in
their red blood cells as compared to
adults that enhance their susceptibility
to oxidant stress and methemoglobin
formation. Since nitrates are used in
agriculture, and occur naturally in
certain food items, there is substantial
opportunity for nitrates/nitrites to be
ingested and so present additional
oxidant stress to erythrocytes. The use
of ClOa as a disinfectant may exaggerate
these effects.
Based upon the hypothesis of the
potential health effects that may be
caused by chlorite and chlorite plus
nitrite combinations on these high risk
groups, it is logical to investigate the
potential health effects of an oxidant
such as ClOj. The potential use of CIO2
in public drinking water requires evalu-
ation of CIO2 alone and in combination
with nitrite, with the ultimate signifi-
cance being the effects of these agents
on humans. Due to the overwhelming
financial and ethical implications of
utilizing human subjects in this study,
the most logical possibility lies in the
use of an animal model possessing
similar enzymatic characteristics as
found in the human condition. There-
fore, a reasonable assessment might be
made concerning the effect of CI02 on
low level G6PD individuals. The animal
models selected for this study were
male A/J (high G6PD activity) and
C57L/J (low G6PD activity) mice from
Jackson Laboratories in Bar Harbor,
Maine.
Results
When mice were exposed to chlorine
dioxide for 30 days at 100 ppm, there
were no significant differences from
controls in any of the blood parameters
measured. However, significant differ-
ences did occur to MCH (mean corpuscu-
lar hemoglobin) and MCHC (mean cor-
pusclar hemoglobin concentration)
when receiving 50 ppm sodium nitrite in
their water for thirty days. There were
no strain differences nor were there any
additive or synergistic effects between
CI02 and nitrite.
Both strains of mice exposed to sodium
chlorite (100 ppm), or a combination of
chlorite and nitrite for 30 days experi-
enced a number of effects on blood
parameters. There were no strain versus
mouse interactions indicating that the
strains did not differ significantly with
respect to treatment.
When A/J (high G6PD) and C57L/J
(low G6PD) mice were exposed to vari-
ous levels of sodium chlorite (0.0, 1.0,
10.0, and 100 ppm) for 30 days in theii
drinking water, there were a number of
blood parameters that varied significant
ly with respect to treatment. However
with the exception of an increase ir
hemoglobin for the C57L/J strain at thj
1.0 and 10.0 level, all significant change!
were associated with the 100.0 pprr
exposure. There were no significan
strain versus treatment interactions fo
any level of chlorite exposure includinc
100 ppm.
The results indicate that chloriti
produces a number of effects on erythro
cytes of both A/J and C57L/J mice
Since chlorite is produced at a level o
50 percent of the chlorine dioxid
demand, caution is suggested in settin
a standard to insure a sufficient margi
of safety. Although there is no signif
cant difference between A/J and C57L/
mice with respect to treatment, th
C57L/J (low G6PD) mice may hav
sufficient G6PD to overcome the oxidar
stress of moderate levels of chlorit
and/or nitrite. Human G6PD deficienl
have even less G6PD activity than th
C57L/J mice and maybe more sensitiv
than the C57L/J mouse to equivalei
amounts of chlorite.
Conclusions
Based upon the data in this report,
appears that exposure of A/J an
C57L/J mice to 100 ppm of chlorite f(
30 days produces increases in G6P
activity, mean corpuscular volumi
-------
osmotic fragility, acanthocytosis, and
cell size (as measured by length [/(] with
transmission E.M.). The primary effect
of chlorite on erythrocytes appears to be
disruption of the cell membrane. There
is a slight but significant increase in
G6PD activity for both strains with no
decrease in GSH. Although GSH is
important to protecting the cell against
oxidative damage, the presence of GSH
within the cell doesn't preclude oxidative
damage to the surface membranes.
Membrane damage is suggested by the
evidence of increased numbers of acan-
thocytes in the treated mice. Acantho-
cytes suggest abnormal lipid content of
cells and such cells tend to become
more permeable with age than normal
cells. Membrane damage as an effect of
chlorite exposure is further supporte.d
by evidence of increased osmotic fragil-
ity. Increased osmotic fragility is normal-
ly associated with spherocytosis, and
the erythrocytes of persons with heredi-
tary spherocytosis (HS) exhibit a charac-
teristic reduction in lipid content with
age when compared with normal cells.
The HS cell also exhibits an increased
rate of Na+ flux suggesting some leaki-
ness of the membrane. The increased
number of acanthocytes along with
increased osmotic fragility in this study,
therefore, suggest that chlorite has
directly or indirectly produced damage
to the erythrocyte membrane and proba-
bly has caused alteration and/or reduc-
tion of certain lipid components of the
membrane. The evidence of increased
mean corpuscular volume (MCV) and an
increased cell length as evidenced by
transmission E.M. suggest an influx of
fluids to the erythrocyte. It is reasonable
to suppose that membrane damage
and/or reduction of ATPase activity
would encourage an osmotic imbalance
and increased fragility of the cell.
Although such changes in erythro-
cytes would likely reduce the length oi
survival time for the affected red cells,
there was no evidence of a decreased
RBC count or an increased number of
reticulocytes. Therefore, the effects are
not sufficient to produce a hemolytic
anemia in the animals tested. Even
though it was anticipated the C57L/J
strain (low G6PD) would be more sus-
ceptible to chlorite ingestion, there
were no obvious strain differences.
Thus, it would appear that the mouse
model does not reveal the enhanced
differential sensitivity at realistic and
10x higher than realistic concentrations
to offer an effective means to evaluate
the study hypothesis.
The final conclusion is that chlorite
exposure in the 100 ppm range produces
definite abnormalities of erythrocytes
that suggest membrane damage. Such
damage may occur in humans with
normal G6PD levels when similarly
exposed. It is possible that persons with
vitamin E deficiency and/or G6PD
deficiency may be at increased risk to
the effects of chlorite, but this has not
been demonstrated in this study. Chlor-
ite may be produced at a rate of 50
percent of the chlorine dioxide demand
and levels of 10 ppm have been reported
under actual disinfection conditions.
Although measurable effects may not
be seen at this level, effects may be
occurring that are not being measured;
or high risk groups may not be part of the
population being measured. In any case,
the difference of 10 ppm treatment
conditions to 100 ppm where effects on
erythrocytes are seen represents a
safety factor of 10 or less. Normal
precaution would suggest a much greater
margin of safety when the number of
people to be exposed represent a large
portion of the population.
Gary S. Moore and Edward J. Calabrese are with the School of Health Sciences,
University of Massachusetts, Amherst, MA 01003.
Paul Heffernan is the EPA Project Officer (see below).
The complete report, entitled "Effect of Chlorine Dioxide, Chlorite, and Nitrite on
Mice With Low and High Levels of Glucose-6-Phosphate Dehydrogenase
(G6PD) in Their erythrocytes," (Order No. PB 81-152 381; Cost: $9.50, subject
to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
> US GOVERNMENT PRINTING OFFICE. 1981 -757-012/7012
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
r
Region 5 Library
Chicago,
------- |