£ 535 /i UNITED STATES ENVIRONMENTAL PROTECTION AGENCY
W ASH INGTCN. O C 20-160
EPA-SAB-DWC-90-015
April 26, 1990
Honorable William K. Reilly
Admi nis tra to r
U.S. environmental Protection Agency
401 tl Street, S . W.
Washington, D.C. 20460
Subject: Science Advisory Board's review of the documen'
"Reaction Kinetics and Reaction Products of Chlorine and
Chioramines in the Digestive Tract"
Dear Mr. Reilly,
The Toxicology Subcommittee of the Science Advisory Board's
Drinking Water Committee met in Washington, D.C. December 8, 1939
to review the document produced for the Office of Drinking Water
"Reaction Kinetics and Reaction Products of Chlorine and
Chloramines in the Digestive Tract".
The document reviewed focuses on the extrapolation of
information about reactions between commonly used disinfectants
and saliva and gastric juices at high dose levels to predict the
reactions of the much lower doses commonly found in disinfected
drinking water. The Subcommittee agrees that the mechanisms
involved at low doses differ from those of high doses and thus ±
simple linear relationship should not be used for extrapolation.
Using current methods the toxicity of the disinfectants may no-
be distinguishable from the toxicity of the by-products, and -_ - ; s
the' toxicity of the by-products should be studied separately.
Thus, studies are needed for the character and the toxicity of
the by-products at low dose levels both for chlorine and for
chloramines used as disinfectants. In addition several
components of a major research program are recommended.
-------
We appreciate the opportunity to conduct this particular
scientific review. We ask that the Agency formally respond to
the scientific advice provided herein.
Sincerely,
/ Q /9/ a
JjjAfyw^ C.. L^^
I\_/*' n —•—"
Raymand C. Lo£nr
Chairman
Executive Committee
William H. Graze
Chairman
Drinkiog Water Committee
-
Ve r ne A. Ray
Chai rman
Toxicology Subcommittee
-------
FPA U.S. Environmental Washington, DC
Protection Agency EPA-SAB-DWC-90-OL5
of the
Committee
of the
and
of and in the
Tract
A ADVISORY BOARD REPORT 1990
-------
1.0
The Toxicology Subcommittee of the Science Advisory Board's
Drinking Water Committee net in Washington, D.C., on December 8,
1939 to review the document produced for the Office of Drinking
Water entitled "Reaction Kinetics and Reaction Products of
Chlorine and Chloramines in the Digestive Tract", written by Dr.
Frank E. Scully, Jr. (Old Dominion University) and Dr. William
'white (University of Vermont) .
The document focuses on the extrapolation of information
about reactions of saliva and gastric juices with disinfectants
commonly found in disinfected drinking water at high dose levels
to predict the reactions that may occur of much lower dose
levels. The Subcommittee agrees that the mechanisms involved at
low dose levels differ from those as high dose levels, and that a
simple linear relationship therefore, should not be used for
extrapolation. Further, the toxicity of the disinfectants may
not be distinguishable from the toxicity of the by-products, and
thus the toxicity of the by-products should be studied
separately. Consequently» studies are needed for the
characterization and the toxicity of the by-products at low dose
levels both for chlorine and for chloramines used as
disinfectants. Although chlorine reacts quickly, chloramine
tends to be more stable in saliva and gastric juices. Due to the
stability of chloramine in saliva there is some potential for
adsorption in the mouth, esophagus, and stomach. Therefore, scne
toxicological effects of chloramine may be attributable to the
parent compound.
The Subcommittee recommends that this problem area receive
more research support. The particular research recommended hers
should include both ia_vivo and in^jnJbrjD studies. Specific
research efforts recommended include verifying the differences
between rat and human saliva, performing studies on the by-
products formed and determining the reaction chemical by-product^
that nay be formed.
-------
c. Do the findings only apply to oxidants or other chemicals
Implicated?
d. the toxicology of disinfectants be distinguished from the
toxicology of in_vlvo by-products at typical drinking water
exposure levels?
e. Are further kinetic needed to quantify rates of
decomposition that would be biologically significant? If
so, what chemical by-products (in__vivo.) studies should be
conducted to identify biologically significant products?
f. Could the Committee suggest in vivo toxicological studies
that would help to specify and quantify the toxicology of
biologically active species produced jn^vivo?
The study of reaction kinetics of oxidants such as
disinfectants is complex for several reasons. First, it is
difficult to quantify reactions mouth and stomach
leaky reaction Second, and stomach
dynamic systems and ingestion of food or the smell of food
generates acidic fluids that can change reaction kinetics
dramatically. For and other reasons discussed below, the
existing knowledge in this area is not complete. Because of the
importance of this information in regulating disinfectants and
disinfection by-products, the Committee recommends that studies
be conducted to provide the missing information.
3.0 OF THE
In the report that follows several raised by
Office of Drinking Water are addressed, followed by a
conclusion from Committee in Section 3.6. It is important to
recognize, however, that the issues tend to overlap and not
necessarily independent,
3.1 How do kinetic considerations of chlorine and chloramine
with endogenous affect the
of available on the toxicity of t'r.e
themselves?
Available information chlorine not
survive long In saliva to be available for and
to produce any direct systemic effects of toxicological
-------
significance. It is possible that very high doses of chlorine
(» 100 ppm) might be high enough to provide some chlorine,
but are no to that resulting
concentrations would be sufficient to significant
toxlcological effects. Thus Subcommittee concurs that in the
procedure of extrapolating the data obtained from animals
administered high doses of chlorine, linear extrapolation may not
be justified.
On other hand, chloramlne has sufficient stability in
saliva so that there is the potential for some chloramine
absorption via the buccal mucosa, esophagus, and stomach.
Therefore, some of the toxicological effects produced by
chloramine may be attributable to the compound..
The report reviewed here does raise concerns about the
potential role of secondary products formed from
disinfectants. As discussed in the report, there Is reason to
believe that the nature of products will vary, depending on
of disinfectant concentration. Thus, toxlcological hazards
of the disinfectants will vary, depending on.the
amounts and nature of the products formed within discrete ranges
disinfectant concentration. It is possible that products may be
formed at low which not'observed at high and that
these more toxic or less toxic than those at high
doses. Consequently, the Committee avoiding the
assumption of a simple linear relationship between production of
these secondary products and disinfectant dose. Studies should
be made of the actual dose-response relationships of the reaction
products, to the extent possible.
3.2 consideration of chlorine and chloramine reactivity
with endogenous compounds affect the characterization
of risk associated with the use of these chemicals as
disinfectants?
formation of products in the mouth, esophagus
and following the ingestlon of chlorine or chloramine
could contribute to the potential hazards associated with the use
of these disinfectants. However, the impact on the risk
characterization cannot be predicted on the basis of current
information. -The nature of by-products formed, the variability
of their formation destruction with varying concentrations of
disinfectant, and their toxicology will to be explicitly
-------
considered. It is doubtful that these problems can be adequately
dealt with using the currently available model studies that were
discussed in the report. Specifically, the model discussed
dissipation of residual disinfectant but was able to identify
reaction products only in the most general of terms.
As mentioned above, there is some possibility that some
portion of a chloramine dose could be systemically absorbed. In
this case, the biologically available absorbed dose would have to
be known and its systemic concentration correlated with the
resulting toxicity caused by all components to determine whether
the chloramine itself or by-products are of greater concern.
Further, if the carcinogenic or non-carcinogenic toxic effect :_ 3
observed only at very high levels of exposure, it is not always
the case that it can be scaled down linearly to low doses. At
high doses chloramine levels may overwhelm the ability of the
endogenous compounds in saliva or gastric juice, and of exogenous
compounds such as food, to react with the chloramine. At lower
dose levels the small amounts of chloramine would not be absorbed
to the same degree and the chloramine would be unlikely to have
much toxicological significance. This plausible non-linear
mechanism argues against using a simple linear extrapolation for
the dose-response relationship in every case.
3.3 Can the toxicity of disinfectant be distinguished from the
toxicity of in vivo by-products at typical drinking
water exposure levels?
In general, the bioassay systems cannot distinguish
differences between toxic effects produced by the parent
disinfectant compound and those produced by secondary products
generated either by chemical or biochemical processes. However,
insight into potentially toxic products generated chemically or
biochemically could allow for the more complete prediction of
toxiqity.
To pursue this avenue, we must increase our understanding z~
these by-products by studying them separately. However, in order
to do this, the by-products must be synthesized in quantities
sufficient for toxicological testing. Endpoints of toxicity ir.
these by-product toxicity studies might correlate with those see-
in toxicity studies with the disinfectant agent. In such a case.
the toxicity could be attributed to the secondary by-product.
-------
However, It should be that high levels of primary
disinfectant be producing by-products, concentrations of by-
products, and toxicity endpoints that are different from those
that result at lower level This could be to
differences in distribution, metabolic pathway kinetics
chemical characteristics of the primary disinfectant in reaction
with body fluids or tissues. At typical drinking water levels,
(i.e. low concentration of primary disinfectant) it is likely
that such low levels of secondary products are formed that it may
be impossible technically to find and prove the involvement of
secondary products in toxicity.
3.4 What mass balance studies are needed to show which
competitive kinetic rates important in the
formation decomposition of biologically significant
by-products?
Before comparative kinetic studies are undertaken,
reactions by-products that are important must be identified.
Because these disinfectants both oxidants and chlorine
substitution agents, the stoichiometry of the reactions must be
studied. Disinfectants that both oxidize form new products
by substitution typically result in different products as a
function of disinfectant dose and substrate concentration.
Products and reactions important at low concentrations becor.e
oxidized unimportant at high doses. It is therefore
recommended that reactions and by-products produced be identified
as a function of halogen concentration.
After the important by-products and the reactions producing
them have been identified, further kinetic studies are needed to
quantify rates of formation decomposition for those that are
biologically significant. studies will likely be different
for chlorine and chloramine, as discussed below.
3.4.1' Chlorination
The report being reviewed here already has identified that,
with large concentration of reducing in the
digestive tract, chlorine at normal dose levels is reduced to
chloride. The issue is the formation of biologically significant
by-products and their decomposition. The first
question is, by-products at low
doses of chlorine to compete with the reduction reactions
-------
removing free chlorine to chloride? Secondly, are the toxic by-
products by either halogenation or oxidation reactions
stable enough to be biologically active? The report did not
identify the likely additional reducing present in_viyo.
These reducing include reduced sulfur functionality, {
nitrite, iodide and ammonia. The Committee recommends further |
studies on chlorination only if studies that stable
toxic by-products produced.
3.4.2 Chloramination
The reactions which need further study following
chlorination of drinking water reactions of chloramines:
NHjCl and organic-N-Cl. These compounds are much less likely to :
be reduced to chloride, as mentioned above. The overall effect ;
of reducing in human saliva gastric fluid on reducing .
NH2C1 and organic-N-Cl can be easily studied by measuring oxidant
demand. Products of reactions with cells and secretions lining ]
gastrointestinal tract less easily modelled. The i
question which needs additional study is the extent of formation
of toxic (perhaps non-oxidant) by-products from NH2C1
organic-N-Cl reactions and potential direct systemic toxicity of
these compounds should they survive unreduced. The major gaps In
the are the reaction kinetics of chloramines,
especially under gastric fluid chloride and pH levels, and the
reactions with the cells lining the gastrointestinal tract.
Additional chloramine kinetic on both
oxidation and substitution reactions, to fill the existing data
gaps. Also studies need to be made on interactions with nitrite,
ammonia, iodide sulfur found in the
digestive tract but not are discussed in the report.
3.5 Are ilLJfJLvo toxicological studies to help specify and
quantify the toxicologically active species produced in
Performing in_^ivo studies may be difficult. On the other
hand, it will be difficult to define the risk from water
disinfection on the risk from one or even a few such active
by-products of unknowns regarding relative concentratlcr.5
under realistic conditions possible interactions.
-------
for reactions at low doses that could be different both
qualitatively and quantitatively from which occur at high
doses. Chlorine react so quickly that not
significant toxicological effects at low linearly relatable
to at Chloramine is at the
degree at low as it is at high At
chloramine overwhelm the ability of endogenous
in saliva or gastric juice of exogenous
as food to react with it. Further, it is possible
that toxic are created at low which become
subsequently destroyed at high doses. of this possible
difference in reactions (with possible differences in
toxicity), Subcommittee does not recommend that high-to-low
dose extrapolation be used for these compounds. To properly
assess their toxic potential, it is necessary to conduct studies
throughout the concentration range encountered in disinfected
waters.
-------
ROSTER
SCIENCE ADVISORY BOARD
TOXICOLOGY
8, 1989
ACTING CHAIRMAN
Dr. Verne Ray
Medical Research Laboratory
Pfizer, inc.
Groton, CT 06340
Dr. Richard Bull
College of Pharmacy
Washington State University
Pullman, WA 99164-6510
Dr. Gary Carlson
Department, of Pharmacology Toxicology
School of Pharmacy
Purdue University
West Lafayette, IN 4790?
Dr. Donald Johnson
Department of Environmental Sciences Engineering
School of Public Health
University of North Carolina
Chapel Hill, NC 27599
Dr David Kaufman
Department of Pathology
University of North Carolina
Brinkhous-Builitt, Room 515
Chapel Hill, NC 27514
Dr, Thomas Tephly
Department of Pharmacology
The Bowen science Building
University of Iowa
Iowa City, IA 52242
-------
EXECUTIVE SECRETARY
Dr. C. Richard
Science Board
A-101F
Washington^ D.C. 20460
STAFF_SECRETARY
Darlene A. Sewell
Science Advisory Board
USEPA A-101F
Washington, D.C. 20460
S TA FF_D IRECTOR
Dr. Donald G.Barnes
Science Advisory Board
A-101
Washington, D.C. 20460
------- |