-------
b
e
fli
41
ac
^
|
led Ian Response Concentration
ur 48 -Hour 96 -Hour
e
t\i
-J
«t
u
i
«
u
S2 2 2 2 2 S ° § §
^(DgttStte&oD 0*0*
* • 19 yg
1 1 1 1 1 1 1 1 *i' '*!' =5 -= -= 1*
J v§ ,8 J ^ ^§ §
oeooeoe" e
tf>i-nusur>iriknuiif%5.S.
^^ P*> ^^ CT ^^ ^0 C^ ^^ CL C^
1 1 1 1 1 1 1
2S*2**»*^*" • *
MWrtrtwrtrjwiiN «IBI
iiuiiniiiiiiiiuu ^*
•B »c »e *e »e *e *c *e • * *it *u
S- o— o— o— e— o— o- o- o o o o o o
<~ x"— *fi" >f>^ *n>- «•>»- *n^- Kir- m S *« K' 5?p S
s, <
5 *
sssssssssss^fs
sssssssssssssis
0 O 0 S"
•« r- — tf «e — » («• — •! «i -»
oil o|i oi< efli e. & K
Q*v ^ § 5 5i£
•531 S— S f2«/I 5*^ ^cce— 'S*£ oi<«$
a. o i3 u M uj >. « i^4 iTix— •
e
u
4i*
ac
u
*••#
VI
II
I/I
a>
I
o
e
n
ee
9
it
fT = flowthrough; H = M
0157d
4-9
04/04/89
-------
Among the most tolerant, studies with fathead minnows and goldfish produced
96-hour LC50s ranging from 150-235 mg/l under certain test conditions,
while tests with sunflsh and bass produced LC5Qs ranging from 140-527
mg/l (Palachek and Tomasso, 1984b; Tomasso, 1986). A detailed review of
the toxlclty of nitrite to fish and the role of environmental variables 1n
modifying the acute effects of nitrite In fish Is presented by Lewis and
Morris (1986).
In other studies assessing the toxlclty of nitrite to aquatic
vertebrates, Colt and Tchobanoglous (1976) determined LTcns for channel
DU
catfish, Ictalurus punctatus. exposed to various concentrations of nitrite
at three different test temperatures. Fish were exposed to nitrite under
static conditions In 40 a. glass aquaria for 9 days. Test solutions were
aerated and fish were not fed during the assays. LTc0s for f1sh exposed
to 18, 32, 56 and 100 mg n1tr1te/l at 22°C were 259.5, 156.8, 67.9 and
18.9 hours, respectively. LT5Qs for fish exposed to 56, 75 and 100 mg
nltrlte/l at 26°C were 376.6, 16.7 and 27.8 hours, respectively. LT5Qs
for fish exposed to 32, 39, 50, 56 and 100 mg n1tr1te/t at 30°C were 240,
124, 30.9, 57.9 and 19.1 hours, respectively. HortalHy curves for nitrite
In channel catfish were linear and did not exhibit a threshold for toxlclty.
Arlllo et al. (1984) assessed the biochemical and ultrastructural
effects of nitrite In rainbow trout, Salmo qalrdnerl. Fish were exposed to
450 jig nitrite/1 of nitrite-nitrogen at 12°C under flowthrough condi-
tions. Chloride content of diluent water was 2 mg/l. Fish were fed twice
dally during the exposure period. Investigators noted significant decreases
1n liver ATP and sugar levels after 12 and 24 hours, respectively, and
significant Increases In liver a-glycerophosphate, lactate and sucdnate
concentrations after 24, 48 and 40 hours, respectively. There were no
0157d
4-10
04/04/89
-------
significant effects on any of these parameters In brain tissues from
exposure to nitrite. Significant changes In the levels of lactate, sugars
and ATP were observed In fish that were overturned after 40 hours of
treatment. Significant Increases In methemoglobln levels were observed In
treated fish after 12 hours. Damage to the hepatic mitochondria was the
most notable ultrastructural effects observed In nitrite-exposed trout.
Hasan and Macintosh (1986) assessed the toxUHy of nitrite to common
carp, Cyprlnus carplo. under the Influence of Increasing chloride concentra-
tions In the assay water. Assays were conducted with reconstituted water at
28°C. Test solutions were aerated but not renewed during the course of
exposure. The presence of chloride In test solutions Increased the toler-
ance of carp to nitrite. Exposure of carp to nitrite In test solutions with
1, 5, 10.5, 27.5 and 45 mg Cl~/l produced 120-hour lC5Qs of 2.3, 5.8,
13.4, 26.4 and 45.2 mg nitrite/8., respectively. The respective 168-hour
LC5Qs were 2.2, 5.7, 12.0, 24.5 and 43.9 mg n1tr1te/l, respectively.
Daniels and Boyd (1987) assessed the toxlclty of nitrite to spotted
seatrout, Cynosdon nebulosus. Eggs were exposed to nitrite In 750 ms.
glass containers with 75 eggs/container. Fertilized eggs were obtained from
hatchery-spawned adults. Test solutions were aerated during the assay.
Temperature and salinity of test solutions during the assay ranged from
26-27'C and 13-14 o/oo, respectively. Total exposure time for eggs was 6
hours. Eggs exposed to the highest concentration of nitrite tested (1200
mg/i) experienced a hatching success rate of 92% vs. a hatching success
rate In controls of 96X.
A variety of Investigators assessed the relationship between exposure of
fish to nitrite and various biochemical parameters from tissues of treated
fish. Smith and Williams (1974) reported significant Increases 1n the
0157d
4-11
07/18/89
-------
levels of methemoglobin of rainbow trout, Salmo galrdnerl. and Chinook
salmon, Oncorhynchus tschawytscha. exposed to 0.15-0.55 mg nitrite/a for
24-48 hours. Brown and McLeay (1975) reported that rainbow trout, Sal mo
galrdnerl. experienced a significant reduction In hemoglobin levels at >0.10
mg nitrite/1 and a significant Increase In methemoglobln at all concentra-
tions tested (X3.015 mg/l) after 96 hours of treatment. The percentage of
methemoglobln to total hemoglobin rose from 0.9% 1n control fish to 78.5X 1n
fish exposed to 0.30 mg/l. Crawford and Allen (1977) reported that the
percentage of methemoglobin 1n Chinook salmon, Qnchorynchus tshawytscha.
Increased when fish were exposed to >4 mg nltrlte/i In freshwater, but was
unaffected In salmon In saltwater (salinity = 32.5 o/oo) until the nitrite
exposure concentration reached 100 mg/l.
Perrone and Mead (1977) reported Increases In the levels of methemo-
globln of coho salmon, Onchorynchus klsutch. exposed to 4-30 mg n1tr1te/l
for 72 hours. Methemoglobln levels In salmon exposed to nitrite In the
presence of chloride (148-260 mg/l) were lower than levels In fish exposed
to nitrite In low chloride water. Raju and Rao (1979) reported only slight
Increases In the percent methemoglobln (1.2-3.OX) In blood of catfish,
Clarlas batrachus. exposed to 5-15 ppm of nitrite. Blanco and Heade (1980)
reported that ascorbic acid In the diet of steelhead trout, Salmo galrdnerl.
moderated the development of methemoglob1nem1a In fish exposed to nitrite.
They also reported that the percentage of methemoglobln Increased when
assays were conducted at elevated temperatures and that larger fish
(37-42 g) exhibited greater sensitivity to exposure to nitrite than smaller
fish (3.2-3.5 g).
Tomasso et al. (1980) reported decreasing levels of methemoglobln 1n
channel catfish, Ictalurus punctatus. exposed to nitrite for 24 hours 1n the
0157d
4-12
07/18/89
-------
presence of Increasing concentrations of chloride. Tomasso et al. (1981)
reported Increases 1n the plasma cortlcosterold concentrations 1n channel
catfish, Ictalurus punctatus. exposed to 1-5 mg nltrlte/i. Elevated
chloride concentrations (303 mg/i) were effective In preventing elevated
cortlcosterold levels In catfish exposed to 5 mg nltrlte/l. Mensl et al.
(1982) reported significant decreases 1n cathepsln B, cathepsln C, leucyl-
amlnopeptldase and total protease activity from livers of rainbow trout,
Salmo galrdnerl. exposed to 0.45 mg n1tr1te/l for >48 hours.
Bowser et al. (1983) reported that an Increasing trend In the chloride
to nitrite ratio resulted In decreasing levels of methemoglobln 1n blood of
channel catfish, Ictalurus punctatus. treated for 48 hours. The sodium and
calcium salts of chloride were equally effective 1n preventing the rise In
methemoglobln levels. When based on weight of monovalent 1on. NaHCCL was
not as effective as either of the salts of chloride In preventing the
development of methemogloblnemla. Mortality of catfish Increased with
Increasing nitrite and decreasing chloride and oxygen levels. Eddy et al.
(1983} reported concomitant rises In the levels of nitrite and methemoglobln
and declines In the percent hematocrlt In plasma of rainbow trout, Sal mo
galrdnerl. exposed to 0.7 and 22.5 mmol/l 1n freshwater and seawater (16
o/oo), respectively, for 24 hours. Increasing the concentration of chloride
In the exposure media reduced the plasma nitrite concentration observed 1n
nitrite-exposed fish. Concomitant declines 1n plasma nitrite and methemo-
globln levels were observed when trout were transferred to nitrite-free
water.
Harglocco et al. (1983) monitored levels of hemoglobin and methemoglobln
In the blood of rainbow trout, Salmo galrdnerl. exposed to 0.45 mg
nltrlte/l for 72 hours. The percentage of methemoglobln was elevated
0157d
4-13
07/18/89
-------
significantly after 12 hours of treatment, while the percentage of hemo-
globin was reduced significantly after 48 hours of treatment. Raju and Rao
(1983) reported significant Increases 1n the activities of succlnate,
glutamate and lactate dehydrogenases from tissues of mosqultoflsh, Gambusla
affInls, exposed to 10, 6 and 4 mg sodium nHrHe/l, respectively, for 96
hours. Palachek and Tomasso (1984a) reported a positive relationship
between methemoglobin levels (0-90X) In channel catfish, Ictalurus punc-
tatus, and tllapla, Tllapla aurea, and environmental nitrite concentrations
(0-25 mg/i) over a 24-hour exposure period. In contrast, methemoglobln
levels 1n largemouth bass, Hlcropterus salmoIdes, were not related to
environmental nitrite concentrations until the exposure concentration of
nitrite exceeded 48.7 mg/i, suggesting the possibility of a nitrite
exclusion mechanism In largemouth bass.
Scarano and Saroglla (1984) reported that sea bass, Dlcentrarchus
labrax. recovered from functional anemia within 24 hours after termination
of exposure to 150 mg nltrlte/i that had lasted for 18 hours. While
methemoglobln levels of treated fish were approximately those of control
fish, hemoglobin levels of treated fish were severely depressed (41-46% of
control) and required 24 days to return to levels within 20X of the control
Msh. Nagaraju and Rao (1985) reported significant Increases In the activi-
ties of aspartate and alanlne amlnotransferase from tissues of mosqultoflsh,
Gambusla afflnls. exposed to 6 and 8 mg sodium nltnte/l, respectively,
for 96 hours. Watenpaugh and Beltlnger (1986) reported a significant
relationship between nitrite exposure concentrations (0.0, 9.3, 18.3 and
27.7 mg/a) and respiration of fathead minnows, Plmephales promelas. F1sh
exposed to the highest nitrite treatments experienced a significant Increase
1n respiration following 24 hours of treatment.
0157d
4-14
07/18/89
-------
Tomasso (1986) reported a wide divergence In the methemoglobin levels
from blood of fish exposed to 40 mg nitrite/a for 24 hours. Green sun-
fish, Lepomls cyanellus. channel catfish, Ictalurus punctatus. and tllapla,
Tllapla aurea. displayed methemoglobln levels ranging -65-85X. Blueglll
sunflsh, Lepomls macrochlrus. and largemouth bass, Hlcropterus salmoldes.
displayed levels of ~10 and <5X, respectively. Tomasso (1986) reported a
significant correlation between plasma nitrite levels and percent methemo-
globln for these five species. Matenpaugh and Beltlnger (1986) reported a
significant relationship between nitrite exposure concentrations (0.0, 0.5,
1.0 and 1.5 mg/4) and blood methemoglobln levels (13.0, 43.6, 65.3, 78.5X,
respectively) 1n channel catfish, Ictalurus punctatus. after a 24-hour
treatment period. Catfish exposed to 1.0 and 1.5 mg nltrlte/i were
significantly more sensitive to anoxlc conditions than fish exposed to <1.0
mg nltrlte/l.
Almendras (1987) reported significant Increases In methemoglobln levels
of freshwater-adapted mllkflsh, Chanos chanos. exposed to nitrite levels
>0.875 mg/a after 48 hours of exposure. A significant Increase was also
observed 1n bracklsh-water-adapted mllkflsh exposed to 14-896 mg
nltrlte/i. H1lmy et al. (1987) reported significant reductions In eryth-
rocytes, hemoglobin, hematocrlt and serum total proteins and a significant
Increase In methemoglobln of the teleost, ClaMas lazera. exposed to 28 and
32 mg n1tr1te/l for 96 hours. Jensen et al. (1987) reported that the
fraction of methemoglobln 1n blood of carp, Cyprlnus carp^o. rose from
4.9-83.3X after 48 hours of exposure to 1 mM environmental nitrite. This
Increase corresponded to a decline In the oxygen saturation of functional
hemoglobin, significant Increases 1n plasma bicarbonate, lactate and
potassium concentrations, and a decline 1n plasma chloride concentration.
0157d
4-15
07/18/89
-------
4.1.1.2. INVERTEBRATES — The acute toxlclty of nitrite to aquatic
Invertebrates as expressed by the LC5_ Is presented In Table 4-2. Among
crustaceans, the late naupl11 stage of the prawn, Penaeus 1nd1cus, was the
least tolerant of exposure to nitrite, with a 24-hour 1C of 10.2 mg/i
(Jayasankar and Muthu, 1983). The water flea, Oaphnla magna. was the most
tolerant of the freshwater Crustacea, with a 24-hour LC of 87 mg/l
(Brlngmann and Kuehn, 1982). Larval prawn, Hacrobrachlum rosenberqll.
exposed to nitrite produced 24-hour LC5Qs of ~70 and 250 mg/l In 12 o/oo
salinity dilution water (Armstrong et al., 1976). Marine molluscs were
highly tolerant of exposure to sodium nitrite. The 96-hour LCrns for
DU
juvenile and adult clams and oysters ranged from 3240-5870 mg/l (Eplfanlo
and Srna, 1975).
In studies with Invertebrates addressing endpolnts other than LC5Qs,
Eplfanlo and Srna (1975) assessed the effect of nitrite on clearance rates
of algae from suspension by the hard clam, MercenaMa mercenarla. and the
American oyster, Crassostrea vlrglnlca. The ability of 2 adults and 10
juveniles In each aquarium to clear a suspension of the alga, Isochrysls
galbana, (IxlO5 cells/ml) was assessed at five concentrations of nitrite
over a 20-hour period. Studies were conducted 1n the dark to minimize algal
growth during the test period. Algal cells were counted with a Coulter
Counter. Clearance of algae from suspension was reduced by <15% 1n
comparison with controls by clams and oysters exposed to 5xlO~3 mol/l.
Exposure to lx!0~a mol/i resulted 1n minimal effects on clearance by
clam adults and oyster juveniles. Clearance by clam juveniles and oyster
adults was reduced by 31.9 and 21.6%, respectively. Exposure of clams and
oysters to 2xlO~2 mol/l resulted In reductions 1n clearance of algae by
Il.5-47.l5i. Exposure to 4xlO~2 and 8xlO~2 mol/l resulted In reduced
clearance rates of 50.9-100% and 93.5-100%, respectively.
0157d
4-16
07/18/89
-------
O
a
w
01
e
e
41
Re
4)
U
9<
O>i—
§ -
.
e
o
>.
WR§
^ r> u
o" S
Vf> O I—
8
CB
ra
tn
rw
cj U
«
nr\iv>
«
V
o
e
o
(M
o
(Si
ss^
u u <
iS
X M
s>
»i
CM » op
»
m
» <^
<— o.x. e
—~^~- 5
IB
ifl IB
ee
«> 41
I
ii
5g
u w
w _ w
1*1=
0157d
4-17
04/04/89
-------
•*"••
!*•*
B
e
w
rw
t
•^
^
a
P*.
0)
o
e
41
^
•>
i^
u
««
•*rf
c
i
(J
t.
a
o
X
c <
o >e
^ ^
•«
L.
•»*
e
«•
w
§ w
LJ a
e
»i ac
»rt '
1 5
wl
0>
ac
e
f
« L.
* I
*
»
CM
-?
IA £
«l «<
-£
ffl
u
1
O
iff
0)
u
«l
,&
? ?
•9 kfl 19 kfl
r« (••
em e w>
^ c ^
^ • «9 »
hk (V 1^ IQ
— C — C
*4_ l_ ^H i
^k ^ ^L ^
Unt C/) dbl (/)
>! >?
^^ ^^
o o
f— p-
1 »
H» VI
3 3
2 -
L. a
— —
e
^-
I
c
Of
t-
u
**
•o
I/I
II
ac
(A
• •
U
«*
^
^«
Iff
II
I/I
i*
1
«^
§
u
Si
••
•«
u
L.
a
i/i
1
«
K
0157d
4-18
04/04/89
-------
Wlcklns (1976) determined LT5Qs for the prawn, Hachrobrachlum rosen-
bergll. which was exposed to the sodium salt of nitrite. Tests were
conducted 1n water at 22.2°C, 3 o/oo salinity and pH 7.4. Exposure to 204,
304 and 419 mg n1tr1te/l resulted 1n LT5Qs of 880, 510 and 410 minutes,
respectively.
Beltlnger and Huey (1981) assessed the toxldty of nitrite to the
crayfish, Procambrus slmulans. Groups of crayfish (10) were exposed to 100
mg nitrite/a, In 30 a aquaria at 25°C for 96 hours. Test solutions were
renewed dally. There were no deaths among crayfish exposed to nitrite 1n
solutions containing 300 mg Cl~/l at pH 7.0. Tests conducted at pH 5.6
under these conditions produced 50% mortality among treated crayfish after
48 and 96 hours. Exclusion of Cl~ (5 mg/i) from test solutions resulted
1n 80, 90 and 100% mortality after 12, 24 and 96 hours, respectively, at pH
7.0 and 5.6.
4.1.2. Chronic Effects on Fauna.
4.1.2.1. TOXICITY -- Westln (1974) assessed the toxldty of nitrite
to chlnook salmon, Oncorhynchus tshawtscha. 1n 10-day static assays. Tests
were conducted 1n reconstituted water at 13.6-15.6°C, and test solutions
were aerated during the assays. The 10-day TL (and 95% confidence
limits) was 2.4 ppm (1.87-3.07).
Yamagata and N1wa (1976) reported that chronic exposure of eels,
Angullla japonlca and Angullla angullla. to 30 ppm nitrite retarded growth,
decreased feed consumption, and reduced erythrocyte counts and hemoglobin
concentration. No effects were observed at 10 ppm nitrite.
Thurston et al. (1978) assessed the toxldty of nitrite to cutthroat
trout, Salmo clarkl. under flowthrough conditions with a flow rate of -125
mi/minute 1n 62 I volume tanks. Test temperatures and pHs ranged from
0157d 4-19 07/18/89
-------
11.8-12.4°C and 7.80-7.88, respectively. Mean sizes of fish used 1n the
assays were 1.0 and 3.1 g. Exposure of trout for 11 days produced LC5Qs
ranging from 0.39-0.54 mg/8.. Exposure of trout for 36 days produced
LC5Qs of 0.37 and 0.38 mg/a. Size of test fish did not Influence the
LC5Qs generated.
Wedemeyer and Yasutake (1978) assessed the effects of nitrite exposure
on growth, blood parameters and ability of steelhead trout, Salmo galrdnerl.
to adapt to seawater. Fish were exposed to nitrite at concentrations
ranging from 0.015-0.06 mg nitrite/!, for 6 months In a freshwater flow-
through system. F1sh were transferred to seawater for 2 months following
treatment 1n freshwater. Nitrite concentrations were measured weekly.
There was no significant correlation between growth In fresh or saltwater
and nitrite concentrations, although there was a significant but biologi-
cally mild methemoglob1nem1a In treated fish (3-4% Increase In methemo-
globin).
Colt et al. (1981) assessed the effects of nitrite on the growth of
channel catfish In a 31-day study. F1sh were exposed to nitrite In flow-
through 40 B. glass aquaria. Flow rate to the aquaria was 150 ml/minute.
Diluent was unchlorlnated well water with a chloride content of 22 mg/a.
Fish were acclimated to the test temperature (28°C) before starting the
experiment, and were fed to satiation twice dally during the course of the
exposure to nitrite. Concentrations of nitrite ranged from 0.012-4.78 mg
nltrlte/l. Investigators reported significantly lower body weights for
fish In treatments >1.62 mg nltrlte/l. There were no significant effects
on final moisture content of fish or gill damage at any treatment level.
0157d 4-20 07/18/89
-------
Grizzle (1983) assessed the effects of nitrite exposure on channel
catfish, Ictalurus punctatus. over 21 days under flowthrough conditions. It
was reported that exposure of channel catfish to 5 mg n1tr1te/l Increased
the chances for bacterial Infections, but that fish were able to acclimate
to the presence of nitrite and were subsequently able to avoid a reduction
In growth rate. Methemoglobln of fish reached 68% of the total hemoglobin
within 8 hours of exposure to 5 mg n1tr1te/l but returned to normal by the
end of the exposure period (21 days). Low hematocMts observed during the
course of the study also returned to normal levels by the end of the
exposure period. Glycogen content of hepatocytes was reduced but growth
rate was not affected by exposure to <2.76 mg n1tr1te/a. Clearance time
for bacteria Injected Into nitrite-exposed catfish Increased.
Holt and Arnold (1983) assessed the effects of nitrite on growth and
survival of red drum, Sclaenops ocellatus. eggs and larvae. Eggs were
obtained from laboratory spawned stock. Eggs and larvae were exposed to
nitrite under static conditions, but nitrite concentrations were measured
and adjusted throughout the assay to maintain exposure concentrations to
within 10% of target concentrations. Salinity and pH ranged from 28-32 o/oo
and 8.0-8.2, respectively. Temperature was maintained at 25-26°C. Percent
hatch of red drum eggs was not affected by exposure to nitrite at any of the
test concentrations (<500 mg/l). Percent survival and growth of larvae
was not affected at <100 mg/i after 14 days of treatment. Survival of
larvae at 500 mg/i after 4 and 14 days was 14 and 0%, respectively.
de L.G. Solbe et al. (1985) assessed the toxldty of nitrite to common
carp, Cyprlnus carplo. and roach, Rutllus rutHus. 1n flowthrough assays.
3157d 4-21 07/18/89
-------
Mean test temperatures for assays with carp and roach were 14.2 and 16.1°C,
respectively. The chloride content of assay water 1n tests with both
species was ~20 mg/i. The 10-day LC5_ (and 95% confidence limits) for
carp was 15.6 mg/l (13.3-17.9). The 14-day LC5Q (and 95X confidence
limits) for roach was 10.1 mg/8, (8.98-11.2). The LT5Qs for carp exposed
to 21, 40, 66 and 96 mg nitrite/a were >527, 111.5, 74.8 and 41.7 hours,
respectively. The LT s for roach exposed to 7, 12, 19 and 38 mg
nltrlte/l were >504, 417, 107 and 21.7 hours, respectively.
H1lmy et al. (1987) reported significant reductions In erythrocytes,
hemoglobin, hematocrlt and serum total proteins and a significant Increase
In methemoglobln of the teleost, Clarlas lazera. exposed to 2.8 and 3.2 mg
nitrite/1 for 6 months.
Armstrong et al. (1976) assessed the effects of exposure of larval
prawns of Hacrobrachlym rgsenberqll to the sodium salt of nitrite. Assays
were conducted 1n 250 mi beakers with 15 organisms In each test beaker.
Prawns were fed brine shrimp and test solutions were renewed every 24 hours.
The 120- to 168-hour LC5Qs ranged from 5-10 mg/l. The 192-hour LC5Q
was 5 mg/t. There were no mortalities In solutions <1.0 mg/i after 168
hours. Larvae exposed to 1.8 mg/s, were significantly smaller than control
prawns at the conclusion of the 8-day exposure period.
Mlcklns (1976) assessed the effects of chronic exposure of prawns,
Penaeus 1nd1cus and Hachrobrachlum rosenbergll. to the sodium salt of
nitrite. Exposures were conducted 1n a flowthrough apparatus at a tempera-
ture of 28°C and salinities of 30-34 o/oo (P. IndUus) and 0.5-4.0 o/oo (M.
rosenbergll). Growth of £. jndlcus was reduced to -50% of that observed In
controls on exposure of prawns to 6.4 mg nitrite/a, after 3-5 weeks of
treatment. LTcns for L Indie us exposed to ~20, 40 and 60 mg/l were
0157d 4-22 07/18/89
-------
~333, 250 and 167 hours, respectively. Growth of M. rosenbergll was not
affected by exposure to <27.94 mg n1tr1te/l. The 4-week LC5Q for M.
rosenberflll exposed to sodium nitrite was 15.4 mg n1tr1te/i.
Jayasankar and Muthu (1983) assessed the toxlclty of sodium nitrite to
larval stages of the prawn, Penaeus Indlcus. Exposure of early nauplU
through the Mysls III stage produced a 10-day IC™ of 0.78 mg
n1tr1te/i. Exposure of late-naupHus stage larvae for 9 days produced an
LC&0 based on mortality of 3.28 ppm. A 9-day EC™ value that Included
both mortality and failure to metamorphose as test endpolnts was 1.8 mg
n1tr1te/i.
4.1.2.2. BIOACCUMULATION/BIOCONCENTRATION — Eddy et al. (1983)
monitored the concentrations of nitrite 1n tissues of rainbow trout, Sal mo
galrdnerl. exposed to 0.7 and 22.5 mmol nltrlte/8, In freshwater and
saltwater (16 o/oo), respectively, for 24 hours. Investigators reported
linear Increases In the plasma concentrations of nitrite for trout 1n both
systems. Levels rose from 0 to -7 mmol/l In freshwater trout and from 0
to -8 mmol/8. 1n trout exposed to nitrite 1n seawater for 24 hours.
Depuration was rapid upon cessation of exposure. Nitrite was depurated
completely within 20-28 hours.
Harglocco et al. (1983) monitored the concentrations of nitrite 1n
tissues of rainbow trout, Salmo galrdnerl. exposed to 0.45 mg nltrlte/8.
for 72 hours. Significant levels of nitrite were found In liver and brain
tissues (2.0 and 2.3 yg/g, respectively) after 12 hours and In blood (7.9
yg/mi) after 24 hours.
Palachek and Tomasso (1984a) monitored the concentrations of nitrite In
plasma of channel catfish, Ictalurus punctatus. tllapla, Tllapla aurea. and
large-mouth bass, Hlcropterus salmoldes. exposed to nitrite concentrations
0157d 4-23 07/18/89
-------
ranging from 0-200 mg/l for 24 hours. Concentrations of nitrite 1n plasma
of catfish and tllapla Increased linearly with respect to exposure concen-
tration, reaching plasma levels of ~80 and 60 mg/l, respectively, for fish
exposed to 25 mg/l. Plasma nitrite levels 1n largemouth bass did not
begin to accumulate until the exposure concentration was >50 mg/l. Bass
exposed to 200 mg/l achieved plasma levels of only -30 mg/l after 24
hours of exposure.
Tomasso (1986) compared the environmental and plasma levels of nitrite
1n green sunflsh, Lepomls cyanellus. channel catfish, Ictalurus punctatus.
tllapla, T1lap1a aurea. blueglll sunflsh, Lepomls roacrochlrus. and
largemouth bass, Hlcropterus salmoldes. exposed to 40 mg n1tr1te/l for 24
hours. Green sunflsh, catfish and tllapla displayed ratios of plasma to
environmental concentrations of 2-3. Ratios for bluegllls and bass were
<0.25.
Jensen et al. (1987) assessed the accumulation of nitrite by rainbow
trout, Salmo galrdnerl. exposed to 1 mM nitrite for 48 hours. Plasma
nitrite levels reached the exposure level within 6 hours and continued to
accumulate 1n plasma to 5.4 mM after 48 hours.
4.1.3. Effects on Flora.
4.1.3.1. TOXICITY -- Admlraal (1977) assessed the tolerance of 10
species of estuarlne benthlc diatoms to various concentrations of nitrite.
Diatoms were Isolated from field samples and grown on an artificial medium.
Salinity of the media was either 15 or 30 o/oo, reflecting the salinity of
the waters from which the diatom species were collected. Tolerance was
assessed by the relative growth of treated cultures In comparison with
control cultures and by Inhibition of photosynthesis. Tests were conducted
In 100 ml Erlenmeyer flasks with a thin layer of analytically clean sand
0157d
4-24
07/18/89
-------
and 40 ma. of culture medium. Tests were conducted at 12°C. There were no
effects on growth of any of the diatom cultures exposed to 1 mmol
nHrlte/i. Growth was Inhibited strongly (88-100% reduction) In cultures
of Navlcula cryptocephala. NUzschla slgma and Stauronels constMcta;
moderately Inhibited (46-63% reduction) In cultures of Navlcula arenaMa.
NUzschla c.f. dlsslpata. NUzschla dublformls and Navlcula sallnarum; and
relatively uninhibited (<20% reduction) 1n cultures of NUzschla closterlum.
Amphlprora c.f. paludosa and Gyroslgma spencer 11 exposed to 10 mmol
nitrite/1. All species experienced >90% Inhibition on exposure to 50 mmol
nitrite/1 except for Amphlprora c.f. paludosa (77% Inhibition). Effects
of nitrite on net photosynthesis was assessed for N. slgma. S. constrlcta.
N. arenarla. N. c.f. dlsslpata. N. closterlum and A. c.f. paludosa. Net
photosynthesis was Inhibited strongly at 50 mrnol nUrlte/i for each of
these species, and Inhibited only moderately at 10 mmol n1tr1te/a. for one
species (N. c.f. dlsslpata).
Hodzlnskl et al. (1977) assessed the effects of exposure of numerous
species of algae to nitrite. Algae were grown In Bristol's solution.
harvested by centrlfugatlon and resuspended In fresh medium augmented with 1
mmol nltrlte/t. After a 40-mlnute Incubation period In the light at 25°C,
NaH1*C03 was added. Cultures were Incubated for an additional 30
minutes In the light before processing the cells to determine the uptake of
radioactivity by liquid scintillation. Photosynthetlc activity was
Inhibited strongly (80-100% Inhibition) 1n cultures of Bummer 1a exllls.
Draparnaldla pulmosa. Staurastrum sp., Oedogonlum foeolarum. SchlzomeMs
Ie1ble1n11. Gloecvstls veslculosa. Anklstrodes-mus falcatzus. Chlamydomonas
relnhardtVK Ulothrlx flmbrlata and Scenedesmus quadrlcauda. Photosynthetlc
0157d 4-25 07/18/89
-------
activity was Inhibited mildly (23% Inhibition) In cultures of Synechococcus
cedrorum and not Inhibited 1n cultures of CyHndrospermum sp., Flscherella
musclcola. Calothrlx anomala. Sch1zothr1x sp., QsclllatoMa sp. Anabaena
flosaquae and Lyngbya sp.
4.1.3.2. BIOCONCENTRATION — Pertinent data regarding the bloconcen-
tratlon potential of nitrite In aquatic flora were not located In the avail-
able literature cited In Appendix A.
4.1.4. Effects on Bacteria. Pertinent data regarding the effects of
exposure of aquatic bacteria to nitrite were not located In the available
literature cited In Appendix A.
4.2. TERRESTRIAL TOXICOLOGY
4.2.1. Effects on Fauna. Stoewsand (1970) assessed the effects of
dietary nitrite on Japanese quail, Cotrunlx coturnlx Japonlca. The diet of
male and female quail at 15 weeks of age was augmented with 0.5X dietary
nitrite as the sodium salt for 1 week. Inclusion of nitrite In the diet
resulted 1n reductions In both food Intake and growth of males and females
compared with controls. Inclusion of nitrite In the diet also resulted In a
decrease In the blood hemoglobin of males and a 2- to 4-fold Increase In the
levels of methemoglobln 1n blood as a percentage of the hemoglobin levels of
both males and females.
4.2.2. Effects on Flora. Pertinent data regarding the effects of
exposure of terrestrial flora to nitrite were not located In the available
literature cited In Appendix A.
4.3. FIELD STUDIES
Pertinent data regarding the effects of nitrite on flora and fauna In
the field were not located In the available literature cited 1n Appendix A.
0157d 4-26 04/04/89
-------
4.4. AQUATIC RISK ASSESSMENT
The lack of pertinent data regarding the effects of freshwater fauna and
flora exposure to nitrite prevented the development of a freshwater crite-
rion {Figure 4-1). Development of a criterion by the method of U.S.
EPA/OWRS (1986) requires the results of acute assays with a planktonlc
crustacean, an Insect, a nonarthropodX-chordate and a species from an Insect
family or phylum not represented previously. Criterion development also
requires the results from acceptably conducted algal assays, chronic assays
with species for which acute data were available, and bloaccumulatlon/
bloconcentratlon studies.
Data available for rainbow trout, Salmo galrdnerl. and common carp,
Cyprlnus carplo. demonstrated that the toxIcHy of nitrite to these species
depends on the chloride concentration In the test medium. The regression of
96-hour LCj-gS for trout against the corresponding chloride concentration
generated a slope and Y-1ntercept of 0.247783 and 1.599, respectively. The
regression of 96-hour LC5Qs for carp against the corresponding chloride
concentration generated a slope and Y-1ntercept of 1.025413 and 1.48578,
respectively. The lack of a comparable study with a freshwater Invertebrate
demonstrating that nitrite toxlclty depends on chloride concentration
prevented the development of an acute equation using a common slope.
Criteria recommended by Calamarl et al. (1984), however, appear adequate for
the protection of these species. Comparing the predicted 96-hour LC5Qs
for either of these species with the criteria recommended by Calamarl et al.
(1984) Indicates that the criteria are -100-fold below the predicted 96-hour
toxlclty values.
The lack of pertinent data regarding the effects of marine fauna and
flora exposure to nitrite prevented the development of a saltwater criterion
0157d 4-27 07/18/89
-------
Family
«1
Chordate tSalmonid-f i»h)
f£
Chord *t> (warmwater fi«h)
«3
Chordate
Crustacean (bent hie)
*b
1 nsect an
«/
non- Ar t hropod / -Chord at •
«tb
New Insectan or phylum
represent at ive
«y
algae
ttiu
Vascular plant
IfcST TYPt
GMftV*
co»
CD*
J.09*
N«
6.1*
NH
IMP
MM
xxxxxxxxxxxx
xxxxxxxxxxxx
xxxxxxxxxxxx
xxxxxxxxxxxx
OMCV*
NO
NA
NA
Ntt
NA
NM
NA
NA
NA
NA
BCF«
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
•NH-Not Available »CD«chloride dependent 96-hour LCB*« for rainbow
trout Salmo nairdneri *CD»chloride dependent 96-hour LC««s for common
carp Cvorinus earoio • 96-hour LC«« in mg/L for salamander flmbvstoima
texanum in water with a chloride concentration of S mg/L '96-hour Ld«
in mg/L for crayfish Procambaru* simulans in Mater with a chloride
concentration of j,& mg/L
FIGURE 4-1
Organization Chart for Listing GMAVs, GMCVs and BCFs Required
to Derive Numerical Mater Quality Criteria by the Method
of U.S. IPA/OWRS (1986) for the Protection of Freshwater
Aquatic Life from Exposure to Nitrite
0157d
4-28
07/18/89
-------
(Figure 4-2). Development of a criterion by the method of U.S. EPA/OWRS
(1986) requires the results of acute assays with species of chordates from
two different families, a mysld or panaeld, two additional nonchordate
species from two different families, and a species from a family not yet
represented. Development of a saltwater criterion by this method also
requires the results from acceptably conducted algal assays, chronic assays
with species for which acute data were available, and b1oaccumulat1on/b1o-
concentratlon studies.
The available acute toxldty data for marine species were Inadequate to
calculate regression equations between salinity and nitrite toxldty In the
same manner that regression equations were calculated for rainbow trout and
common carp.
4.5. SUMMARY
The toxldty of nitrite to fish and amphibians depends significantly on
the concentration of chloride or on salinity of the test medium. This
dependency was demonstrated for Chinook salmon, Onchorhynchus tshawytscha
(Crawford and Allen, 1977), rainbow trout, Salmo galrdnerl (Russo and
Thurston, 1977) and carp, CypMnus carplo (Hasan and Haclntosh, 1986).
Additional data demonstrated that larger trout (10 g) were consistently less
sensitive (1.2- to 4-fold) than smaller trout (5 g), while trout were more
sensitive to nitrite at lower pHs (6-7) than fish 1n tests conducted at a
higher pH (8) (Wedemeyer and Yasutake, 1978). Salmonlds were the aquatic
vertebrates most sensitive to exposure to nitrite, with 96-hour LC5Qs for
rainbow trout ranging from 0.1-1 mg/l (Russo and Thurston, 1977; Wedemeyer
and Yasutake, 1978). Fathead minnows, Plmephales promelas. and goldfish,
Carraslus auratus. were among the most tolerant, with 96-hour LCcns
0157d 4-29 07/18/89
-------
Family
• «1
Chordat »
«c'
Chordat*
«3
non-Arthropod / -Chordat •
«4
Lrustac*an (Mysid/Pana*id>
415
non-Chordat*
«b
non-thordat *
«•/
non-i;hordat e
«B
Oth*r
«y
• lg*e
«1O
Vascular plant
TtST TYPE
BMRV*
NA
NA
S.7£»
NA
3.57-
NA
NA
NA
xxxxxxxxxxxx
xxxxxxxxxxxx
xxxxxxxxxxxx
xxxxxxxxxxxx
6MCV-
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
BCF*
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
•NA-Not Available •96-hour TL. in a/L for th* hard clam M»re«naria
fl»re»naria at a salinity of £?•/.. '96-hour TL. in g/L for th* American
oyster Cr*"°*tr>* virninica at a salinity of £?•/..
FIGURE 4-2
Organization Chart for Listing GHAVs, GHCVs and BCFs Required
to Derive Numerical Hater Quality Criteria by the Method
of U.S. EPA/OURS (1966) for the Protection of Saltwater
Aquatic Life from Exposure to Nitrite
0157d
4-30
07/18/89
-------
ranging from 150-235 mg/l, while 96-hour LC5Qs with sunflsh, Lepomls
macrochlrus. and bass, Hlcropterus sp., ranged from 140-527 mg/l (Palachek
and Tomasso, 1984b; Tomasso, 1986).
Biochemical and ultrastructural effects of nitrite among a variety of
fish Included significant decreases In erythrocytes, hemoglobin, hematocMt,
serum total proteins, liver ATP and sugar levels, cathepsln B, cathepsln C,
leucylamlnopeptldase and total protease activity, and significant Increases
1n the activities of aspartate and alanlne amlnotransferase, sucdnate,
glutamate and lactate dehydrogenases. Test specimens also experienced
significant Increases In respiration and liver a-glycerophosphate, plasma
cortlcosterold concentrations, methemoglobln levels, plasma bicarbonate,
lactate and potassium concentrations, declines 1n plasma chloride concentra-
tion and damage to the hepatic mitochondria (Arlllo et al., 1984; Smith and
Williams, 1974; Brown and HcLeay, 1975; Crawford and Allen, 1977; Perrone
and Head, 1977; Raju and Rao, 1979; Blanco and Heade, 1980; Tomasso et al..
1980; Mensl et al., 1982; Eddy et al., 1983; Marglocco et al., 1983; Raju
and Rao, 1983; Palachek and Tomasso, 1984a; Scarano and Saroglla, 1984;
Nagaraju and Rao, 1985; Watenpaugh and Beltlnger, 1986; Jensen et al., 1987;
Almendras, 1987; H1lmy et al., 1987). Other Investigators reported signifi-
cant correlations between plasma nitrite levels and percent methemoglobln In
blood of several species of fish, although methemoglobln levels In large-
mouth bass, Hlcropterus salmoldes. were not related to environmental nitrite
concentrations until the exposure concentration of nitrite was >48.7 mg/l
(Tomasso, 1986; Hatenpaugh and Beltlnger, 1986; Palachek and Tomasso, 1984a).
Among crustaceans, the late nauplH stage of the prawn, Penaeus Indlcus.
was the least tolerant of exposure to nitrite, with a 24-hour LC5Q of 10.2
mg/l (Jayasankar and Muthu, 1983). The water flea, Daphnla maqna. was the
0157d 4-31 07/18/89
-------
most tolerant of the freshwater Crustacea, with a 24-hour LC5Q of 87
mg/l (Brlngmann and Kuehn, 1982). Larval prawn, Macrobrachlum rosen-
berqll. exposed to nitrite produced 24-hour LC5Qs of ~70 and 250 mg/l In
12 o/oo salinity dilution water (Armstrong et al.t 1976). Marine molluscs
were highly tolerant of exposure to sodium nitrite. The 96-hour LC5Qs for
juvenile and adult clams and oysters ranged from 3240-5870 mg/l (Eplfanlo
and Srna, 1975).
The 10-day TL for nitrite In Chinook salmon, Oncorhynchus tshawtscha.
was 2.4 ppm (Westln, 1974). The NOEL for nitrite 1n eels, Anqullla japonlca
and AnguUla anqullla. was >10 and <30 ppm (Yamagata and N1wa, 1976). The
11- and 36-day Lc5gs 1n cutthroat trout, Salmo clarkl. ranged from
0.39-0.54 and 0.37-0.38 mg/l, respectively (Thurston et al., 1978).
Channel catfish experienced significantly lower body weights for fish
exposed to >1.62 mg n1tr1te/l for 31 days (Colt et al., 1981).
Percent hatch of red drum eggs of Sclaenops ocellatus was not affected
by exposure to nitrite at concentrations <500 mg n1tr1te/l. Percent
survival and growth of larvae was not affected at <100 mg/i after 14 days
of treatment. Survival of larvae at 500 mg/l after 4 and 14 days was 14
and 0%, respectively (Holt and Arnold, 1983). The 10-day LC5Q for common
carp, CypMnus carplo. was 15.6 mg/l, while the 14-day LC5Q for roach,
Rutllus rutllus. was 10.1 mg/l (de L.G. Solbe et al., 1985).
The 192-hour LC5Q for larval prawns of Hacrobrachlum rosenberqll was 5
mg/l. There were no mortalities In solutions <1.0 mg/l after 168 hours.
Larvae exposed to 1.8 mg/l were significantly smaller than control prawns
at the conclusion of the 8-day exposure period (Armstrong et al., 1976).
Growth of Penaeus Indlcus was reduced to ~50% of that observed In controls
on exposure of prawns to 6.4 mg n1tr1te/l after 3-5 weeks of treatment.
0157d
4-32
07/18/89
-------
Growth of M. rosenbergll was not affected by exposure to <27.94 mg
n1tr1te/i. The 4-week LC5Q for H. rosenberqll was 15.4 mg nltrHe/fc
(Wlcklns, 1976). Exposure of early naupl11 of the prawn, Penaeus Indlcus.
through the Mysls III stage produced a 10-day LC5_ of 0.78 mg n1tr1te/i.
Exposure of late-naupHus stage larvae for 9 days produced an LC5Q based
on mortality of 3.28 ppm. A 9-day EC™ value that Included both mortality
and failure to metamorphose as test endpolnts was 1.8 mg nitrite/I
(Jayasankar and Muthu, 1983).
Uptake of nitrite by fish was rapid and reflected exposure concentration
and duration. In general, depuration of nitrite from fish during recovery
periods was equally rapid (Eddy et a!., 1983; Marglocco et al., 1983;
Tomasso, 1986; Jensen et al., 1987). Plasma nitrite levels 1n largemouth
bass, Mlcropterus salmoldes. however, did not begin to accumulate until the
exposure concentration was >50 mg/i, and reached only -30 mg/i In the
plasma of bass exposed to 200 mg/a for 24 hours (Palachek and Tomasso,
1984a).
Growth of benthlc diatoms was Inhibited strongly (38-100% reduction) In
cultures of Navlcula cryptocephala. NUzschla slgma and Stauronels
constrlcta: Inhibited moderately (46-63X reduction) 1n cultures of Navlcula
arenaMa. NUzschla c.f. dlsslpata. NUzschla dublformls and Navlcula
sallnarum; and relatively uninhibited (<20% reduction) 1n cultures of
NUzschla closterlum. Amphlprora c.f. paludosa and Gyros 1qma spencer 11
exposed to 10 mmol n1tr1te/l. All species experienced >90X Inhibition
upon exposure to 50 mmol nitrite/I except for Amphlprora c.f. paludosa
(77X Inhibition). Net photosynthesis was Inhibited strongly at SO mmol
nltrlte/l for N. slqma. S. constrlcta. N. arenarla. N. c.f. dlsslpata. N.
closterlum and A. c.f. paludosa. and Inhibited only moderately at 10 mmol
0157d
4-33
04/04/89
-------
nUrUe/t for N. c.f. dlsslpata. There were no effects on growth of any
of the diatom cultures exposed to 1 mmol nUrUe/l (Admlraal, 1977).
Photosynthetlc activity was Inhibited strongly (80-100X Inhibition) In
cultures of BumllleMa exllls. Draparnaldla pulmosa. Staurastrum sp.,
Oedoqonlum foeolarum. Schlzomerls lelblelnll. Gloecvstls veslculosa.
Anklstrodesmus falcatzus. Chlamydomonas relnhardtll. Ulothrlx flmbrlata and
Scenedesmus quadrlcauda Incubated with 1 ramol nltrlte/i for 70 minutes.
Photosynthetlc activity was Inhibited mildly (23X Inhibition) 1n cultures of
Synechococcus cedrorum. and not Inhibited 1n cultures of Cyllndrospermum
sp., Flscherella rousclcola. Calothrlx anoroala. Schlzothrlx sp., Osclllatorla
sp. Anabaena flosaquae and Lvnqbva sp. (Wodzlnskl et al.. 1977).
Inclusion of 0.5X nitrite In the diet of Japanese quail. Cotrunlx
coturnlx Japonlca. for 1 week resulted 1n reduced food Intake and growth of
males and females compared with controls, and a decrease In the blood hemo-
globin of males with a 2- to 4-fold Increase 1n the levels of methemoglobin
as a percentage of the hemoglobin levels of both sexes (Stoewsand, 1970).
Ol57d
4-34
04/04/89
-------
5. PHARMACOKINETICS
5.1. ABSORPTION
Instability of the Ion In acid and acid-catalyzed reactions with the
diet complicates quantification of nitrite absorption from the GI tract
(Mlrvlsh et al.. 1975), Friedman et al. (1972) administered 150 v>9 of
sodium nitrite In water by gavage to groups of 13-18 young adult male Swiss
ICR/Ha mice to measure disappearance of nitrite from the stomach. Nitrite
disappearance was observed to fit a second-order kinetic model. By 10
minutes after administration, 85% of the nitrite had disappeared from the
stomach. Ugatlon at the pylorus to prevent normal stomach emptying had no
effect on the rate of nitrite disappearance.
in vitro experiments were performed with Isolated mouse stomachs to
which sodium nitrite had been added (Friedman et al., 1972). After a
30-mlnute Incubation, 63% of the administered nitrite had disappeared. Of
that, 40% had been converted to nitrate. The Investigators concluded that
absorption Into the bloodstream 1s the major pathway by which nitrite leaves
the stomach. They stated that the second-order mechanism observed J_n vivo
1s consistent with conversion of nitrite to Np03 followed by rapid
absorption of N2°3' not1n9 tnat uncharged molecules are generally
absorbed more rapidly than charged molecules.
Witter et al. (1979) administered an unspecified dose of iaN-n1tr1te
by gavage to Sprague-Dawley rats (sex not reported) with or without
ligatures of the pylorls and measured the disappearance of radioactivity
from the stomach. In contrast to the conclusions of Friedman et al. (1972)
regarding mice, these Investigators concluded that some gastric absorption
of radlolabel had occurred, but that most disappearance from the stomach
Involves passage Into the duodenum. Using Ui situ rat preparations, Frltsch
0158d
5-1
07/18/89
-------
et al. (1980a) estimated that SOX of a nitrite dose was absorbed from the
small Intestine. Without providing data or a reference, Ishlwata and
Tanlmura (1982) stated that nitrite Is absorbed from the Ugated stomach and
proximal small Intestine of guinea pigs.
Parks et al. (1981) administered Intravenous or Intratracheal doses of
l3N-n1tr1te of 10-100 ng/kg to groups of 10-12 BALB/C mice (sex
unreported), which were sacrificed at 5-30 minutes for measurement of radio-
activity In selected organs. Although quantitative time vs. organ concen-
tration data were not provided, the Investigators stated that distribution
following Intratracheal administration was time-dependent, slightly slower
than following Intravenous administration and appeared to be perfuslon-
Umlted, In subsequent reference to this work, Parks and Krohn (1983)
stated that l3N-n1tr1te was cleared from the lungs within 30 minutes.
5.2. DISTRIBUTION
Experiments In several species suggest widespread and rapid distribution
of nitrite to all soft tissues. Schneider and Yeary (1975a,b) administered
20 mg/kg of sodium nitrite Intravenously Into seven dogs (several breeds,
both sexes), seven sheep (cross-bred females) and seven ponies (both sexes).
Based on the plasma decay curves, distribution half-lives of 47.8, 11.9 and
5.3 minutes were estimated for the dogs, sheep and ponies, respectively,
suggesting rapid distribution. Vds were estimated at 1624, 278 and 192
ml/kg for these species. The large Vd In dogs, coupled with low plasma
concentration, suggested that the nitrite had been rapidly sequestered by
the erythrocytes.
Parks and Krohn (1983) Investigated the distribution of 13N from
l3N-n1tr1te In mice (Table 5-1). They noted that dynamic equilibrium was
reached within 5 minutes after administration by any of the three routes
0158d
5-2
04/05/89
-------
TABLE 5-1
Tissue Distribution of 13N In BALB/C M1cea»b
Tissue
Carcass
Lung
Kidneys
Liver
Stomach
Small Intestine
Large Intestine
Intravenous
58.8/4.11
1.55/5.6
3.51/7.82
7.93/6.95
4.04/7.00
7.00/5.25
5.83/7.14
Route of Adm1n1strat1onc
Intraesophageal
48.4/2.28
1.67/6.90
2.95/6.12
6.19/4.89
23.5/66.5
8.10/6.31
3.47/3.71
Intratracheal
61.1/3.43
5.06/16.25
3.56/7.23
9.76/7.28
4.40/6.73
8.45/5.10
6.46/5.95
aSource: Parks and Krohn, 1983
^Groups of 10-12 adult mice given single 10-100 ng/kg doses of
sacrificed 5-30 minutes after treatment. Values are adjusted means for all
sacrifice times when time covarlance 1s Included, expressed as percent of
administered dose per organ/percent of administered dose/g of organ.
cRad1oact1v1ty measured by gamma-counting
0158d
5-3
04/05/89
-------
used (gavage, Intratracheal, Intravenous). WHh the exception of the
stomach after gavage treatment and the lungs after Intratracheal treatment,
there were no significant differences In tissue concentrations that were due
to administration route. Radioactivity concentrations expressed as percent
of administered dose/g of tissue Indicated that there were no differences 1n
tissue affinity for 13N. The lower values for the eviscerated carcass
probably reflect the low vascularlty and minimal uptake by the skeleton.
Parks and Krohn (1983) stated that Increasing the dosage of nitrite to 60
mg/kg had no effect on organ distribution compared with the much smaller
doses used In this study.
Parks et al. (1981) and Parks and Krohn (1983) also studied radio-
activity distribution from Intravenously administered 13N-n1tMte 1n New
Zealand white rabbits with an Auger scintillation camera. Parks et al.
(1981) stated that radioactivity distribution was rapid and homogeneous
throughout the soft tissues. Equilibrium between Intra- and extravascular
compartments was reached within 5 minutes. At 30-45 minutes after
treatment, the urinary bladder contained 2-3X of the administered dose of
radioactivity.
Parks et al. (1981) subjected plasma from which the protein fraction had
been removed to HPLC to Identify the radioactive moiety. The plasma was
obtained 10 minutes after Intratracheal administration of lsN-n1tr1te to
mice or Intravenous administration to rabbits. In mice, 70X of the
radioactivity existed as nitrate, 21% as nitrite and 3X as nonanlonlc
N-contalnIng compounds. Rabbit plasma contained 51% of the radioactivity as
nitrate, 46X as nitrite and 3X as nonanlonlc compounds. The Investigators
concluded that nitrite 1s oxidized to nitrate In blood.
0158d
5-4
07/18/89
-------
Parks and Krohn (1983) stated that 13N-nHr1te radioactivity crossed
the placental barrier of pregnant rats. Large nitrite doses crossing the
placenta resulted In fetal methemogloblnemla (Gruener et at., 1973; Shuval
and Gruener, 1977).
5.3. METABOLISM
There are three facets of nitrite metabolism pertinent to Its potential
toxlclty: (1) reduction of nitrate to nitrite In the GI tract; (2) reaction
of nitrite with amines present 1n the GI tract; and (3) the conversion of
hemoglobin to methemoglobln accompanied by oxidation of nitrite to nitrate.
Hlcroflora In the GI tract reduce nitrate to nitrite. Jji vitro studies
show that the reduction of nitrate to nitrite by bacteria In human saliva
appears to be strongly pH-dependent, with maximum activity at pH 6-6.4 and
complete Inhibition of activity at pH <4 or >9 (Goaz and Blswell, 1961).
The normal pH of the adult human stomach averages <3 (U.S. EPA, 1985). The
stomach pH of breast-fed Infants averages 3.75, while that of Infants
accustomed to cow's milk (which has a high buffering capacity) averages
4.75. Therefore, Infants fed cow's milk, adults and Infants with disease
conditions, and those under treatment with medications that raise stomach pH
may efficiently reduce nitrate Ingested In food or secreted 1n the saliva to
nitrite.
The fate of orally-administered nitrate and nitrite has been studied In
germ-free and conventional rats. Witter and Ballsh (1979) provided tap
water containing additional nitrate (from sodium nitrate) at 0 or 1000
yg/mt (1000 ppm) to conventional and germ-free Sprague-Dawley rats and
measured the nitrate and nitrite concentrations 1n the stomach and small
Intestine. Low levels of nitrate were found In the stomach, but not In the
small Intestine, of both germ-free and conventional control rats. The
0158d
5-5
07/18/89
-------
Investigators attributed this to nitrate In the diet and 1n tap water
offered as drinking water. Germ-free rats given nitrate had Increased
levels of nitrate, compared with germ-free controls, In both the stomach and
small Intestine. Conventional rats given nitrate had elevated levels of
nitrate In the stomach and small Intestine, and Increased levels of nitrite
In the stomach. When tap water containing additional nitrite (1000 ppm from
sodium nitrite) was given, elevated nitrite and nitrate levels were found 1n
the stomach and small Intestine of both germ-free and conventional rats,
compared with controls. The Investigators hypothesized that nitrite had
been oxidized In the add environment of the stomach. The Investigators
concluded that reduction of nitrate to nitrite Is accomplished by the micro-
flora of the gut and that oxidation of nitrite to nitrate Is accomplished by
the mammalian host.
Ward et al. (1986) provided distilled water containing 2% potassium
nitrate to conventional, germ-free and gnotoblotlc Porton-Hlstar rats (the
latter contaminated with one unidentified yeast strain). The Investigators
measured blood methemoglobin concentration as an Indicator of reduction of
nitrate to nitrite. (Nitrite oxidizes hemoglobin to methemoglobln; see
below.) Markedly elevated methemoglobln levels, as compared with pretreat-
ment values, were measured In all three groups of rats, with little
difference between groups. The Investigators then anaeroblcally Incubated
mucosal scrapings from the stomachs or small Intestines of germ-free rats
with nitrate to determine If reduction to nitrite would occur In the absence
of microorganisms. Much more nitrate-reducing activity was found 1n the
Intestinal preparation than 1n the stomach preparation. The nitrate-
reducing activity In the Intestinal preparation was largely heat-labile,
suggesting to the Investigators that the reaction 1s enzyme-catalyzed. In
0158d
5-6
04/05/89
-------
contrast to Witter and Ballsh (1979), Hard et al. (1986) concluded that
reduction of nitrate to nitrite can be accomplished by the GI mucosa as well
as by the resident mlcroflora. They speculated that the nitrate concentra-
tion 1n the drinking water In the Witter and Ballsh (1979) study may have
been too low to result In detectable levels of nitrite.
Nitrite In the GI tract can react with primary amines with Immediate
decomposition to molecular nitrogen and the corresponding alcohol or olefln,
or with secondary and tertiary amines to form N-n1trosam1nes that may be
subject to subsequent oxidation and reduction reactions to yield molecular
nitrogen (Frank et al., 1985). The formation of N-nHrosamlnes In the GI
tract Is a concern because several nltrosamlnes thus formed have been shown
to be potent animal carcinogens (Section 6.2.2.). The proximate carcinogen
1s probably the alkyldlazohydroxlde, which can alkylate nucleophlllc groups
on macromolecules with simultaneous release of molecular nitrogen (Frank et
al.. 1985).
Frank et al. (1985) administered lsN-sod1um nitrite or l5N-d1methyl-
amlne and 14N-n1irHe by gavage to male Sprague-Dawley rats maintained In
a closed system In an atmosphere of helium and oxygen. 15N-n1tr1te was
expected to react with primary amines In the diet to form 1SN-14N.
lsN-d1methylam1ne was expected to react with 14N to form 15N-n1troso-
dlmethylamlne, which would undergo mlcrosomal metabolism with release of
1SN-14N. Molecular nitrogen released by either reaction would be
exhaled and available for measurement as an Indication of the extent to
which the reaction occurred. When l5N-n1tr1te was administered (0.71
mmol/kg or 50 mg/kg), 14N-15N was expired at the rate of 0.11
pmol/mln/kg for the first 3 hours. Considerable variation was observed,
which the Investigators attributed to availability of primary amines
0158d
5-7
07/18/89
-------
(probably dependent upon nutritional status), gastric pH and loss of
nitrogen from the closed system during flushing with the helium-oxygen
mixture. Detection of 14N-15N following treatment with lsN-d1methyl-
amlne {1.1 mmol/kg, -50 rag/kg) and l«N-n1tr1te (0-2.2 mmol/kg, -0-100
mg/kg) followed a lag time of ~1 hour. The amount of 14N-15N recovered,
expressed as percent of l5N-d1methylam1ne administered. Increased 1n a
dose-related manner over -10 hours. A maximum of ~6X of the dose was
recovered when nitrite was given at 2.2 mmol/kg.
U.S. EPA (1985) and NCI (1982) summarized a large body of literature
regarding the j£ vivo and jn. vitro formation of N-nltrosamlnes from the
reaction of nitrite with many chemical classes of secondary or tertiary
amines. The extent to which nltrosatlon occurred depended on many factors.
The structure of the amlne appeared to be an Important determiner; weakly
basic amines were nltrosated more rapidly by several orders of magnitude
than strongly basic amines. In most cases, optimum pH appeared to be
between 1 and 3. For ±t± vivo formation of N-nltroso compounds In detectable
amounts, an exogenous nitrite source appeared to be necessary. Simultaneous
addition of alpha-tocopherol or ascorbic add reduced formation of N-n1troso
compounds, apparently by reducing the concentration of available nitrite
Ions.
Hard et al. (1986) compared the rate of formation of N-n1trosoprol1ne In
germ-free and conventional Porton-Wlstar rats provided with drinking water
containing 0.5-1% prollne (a secondary amlno acid) and 1-2X sodium nitrate
for 6-7 days. Small numbers of animals were used and the Individual varia-
tion was substantial. However, nltrosatlon appeared to occur more rapidly
1n conventional than In germ-free rats, as evaluated by recovery of the
N-n1trosoprol1ne 1n urine. The Investigators postulated that nltrosatlon
0158d
5-8
04/05/89
-------
occurred without direct mlcroblal Involvement, although Ui vitro data
Indicate that microbes can catalyze N-nltrosaUon reactions, and that lower
gastric pH 1n the conventional state enhanced the reaction rate.
The classical syndrome of acute nitrate (or nitrite) toxlclty Is anoxia
resulting from methemogloblnemla. The biochemical reactions Involved In the
conversion of hemoglobin to methemoglobln are not entirely understood. Lee
(1970) stated that methemoglobln 1s formed when nitrite oxidizes the ferrous
Iron In hemoglobin to the ferric state. Parks et al. (1981) postulated that
the mechanism Involved 1s dependent on the nitrite concentration In the
blood. At low concentrations, the normal spontaneous oxidation of oxyhemo-
globln results In the formation of methemoglobln and superoxlde Ion, which
1s converted to hydrogen peroxide by superoxlde dlsmutase. The hydrogen
peroxide then forms a complex with catalase for which nitrite 1s a sub-
strate. The final step 1s oxidation of nitrite to nitrate with the release
of water. At higher concentrations nitrite reacts directly with oxyhemo-
globln, with the formation of methemoglobln and nitrate. Methemoglobln Is
Incapable of releasing oxygen to the tissues.
Burrows (1979} administered sodium nitrite Intravenously to groups of 4
mature crossbred female sheep to Investigate the action of nitrite on blood
hemoglobin. Blood methemoglobln concentrations reached maxima (expressed 1n
terms of percent conversion of hemoglobin) of 13X at 15 minutes, 43X at 45
minutes and 63X at 60 minutes at doses of 6.6, 22 and 35 mg/kg, respec-
tively. A dose of 50 mg/kg was fatal, resulting In 80% conversion of
hemoglobin to methemoglobln at 60 minutes, just before death. At nonfatal
doses, a reductase enzyme system 1n the erythrocytes converted methemoglobln
back to hemoglobin. l£ vitro studies Indicate marked species differences 1n
nitrite's ability to reduce hemoglobin to methemoglobln. Calabrese et al.
0158d
5-9
04/05/89
-------
(1983) Incubated rat, sheep and human erythrocytes 1n a system containing 3
mH sodium nitrite for 2 hours. Conversion of hemoglobin to methemoglobin
was 14.2, 47.8 and 48.6% 1n the three species, respectively. Hethemoglobln
reductase activity, which converted methemoglobln back to hemoglobin, was
present 1n rat erythrocytes at a level ~5 times that 1n human erythrocytes.
U.S. EPA (1985) reported that human blood normally contains methemo-
globln equivalent to 0.5-2.5% of total hemoglobin. Higher values have been
measured In pregnant women, with a maximum of 10.5% at the 30th week of
gestation {SkMvan, 1971). Pregnant women, therefore, may represent a
segment of the population unusually sensitive to the toxic effects of
nitrite. Lee (1970) noted that Infants are unusually sensitive to the toxic
effects of nitrite. Fetal hemoglobin, which may constitute up to 80% of the
hemoglobin 1n neonates, Is more readily converted to methemoglobln. Also,
the enzymatic methemoglobln reductase system may not be as efficient In
Infants as In adults.
Data In several species suggest that mammals rapidly oxidize nitrite to
nitrate. Parks et al. (1981) and Parks and Krohn (1983) administered
l9N~n1tr1te (dosage not specified) Intratracheally to mice and Intra-
venously to rabbits to determine the partition of the Ions In fractions of
the blood. In mice, 75% of the nitrite had been oxidized to nitrate In a
plasma sample taken 10 minutes after administration. The site of oxidation
was hypothesized to be the erythrocyte, because 100% of the 13N located In
cell lysate was nitrate. In rabbits, however, 51% of the Intravenously
administered nitrite had been converted to nitrate. The Investigators
concluded that there are species differences In the rate of conversion of
nitrite to nitrate 1n mammalian blood.
0158d 5-10 04/05/89
-------
Yoshlda et al. (1983) administered 15N-sod1um nitrite (2 mg
lsN/an1mal) by Intraperltoneal Injection Into male Wlstar rats and
Identified metabolites In urine collected over 48 hours. Total urinary
radioactivity accounted for about 53% of the dose. Of the radioactivity
recovered In urine, 49-61% was Identified as nitrate, 23-27% was Identified
as urea and the remainder was unidentified nitrogenous compounds. Male JCL
mice were similarly treated with 0.617 mg 15N from sodium nitrite and
placed In metabolism cages for 48 hours for collection of urine, feces and
expired air. Total recovery of radioactivity averaged 70.6%. Expressed In
terms of administered dose, 61% was recovered 1n urine, 7.8% 1n feces, 0.3%
1n expired air and 1.6% In the carcass. Of the radioactivity recovered In
urine, -80% occurred as nitrate, 14% as urea and 0.6% as protein.
5.4. EXCRETION
In the study described In Section 5.2., Schneider and Yeary (1975a,b)
estimated an elimination half-time of Intravenously-administered nitrite
from the plasma of dogs, sheep and ponies of -0.5-0.6 hours. The Investi-
gators suggested that this extremely rapid half-time of elimination Involved
metabolic conversion to nitrate rather than renal clearance alone. Half-
times of nitrate elimination from the plasma were 44.7, 4.2 and 4.8 hours In
the dog, sheep and pony, respectively. The Investigators suggested that the
elimination half-times for nitrate represent excretion, because nitrate Is
not expected to undergo further metabolism.
Yoshlda et al. (1983) administered "N-sodlum nitrite by Intraperlto-
neal Injection to rats and mice (see Section 5.3.). Urinary excretion of
radlolabeled metabolites accounted for -53% of the dose 1n rats and 61% of
the dose In mice. Although 7.8% of the radioactivity In mice was recovered
In feces and 0.3% In expired air, most of the radioactivity 1n feces was
0158d 5-11 04/05/89
-------
attributed to contamination with urine. The radioactivity 1n expired air
was attributed to ammonia generated by decomposition of radlolabeled urea.
Ishlwata and Tanlmura (1982) studied the excretion of nitrate 1n
Japanese people whose dietary Intake of nitrate was estimated at 218-408
mg/day. Saliva presumably recovered from the mouth contained 73£78 ppm
nitrate and 16+21 ppm nitrite. Ouctal saliva contained only nitrate,
suggesting that the nitrite In samples recovered from the mouth are the
result of mkroblal nitrate reduction. Urine, presumably from the same
persons from whom the aforementioned saliva samples were obtained, contained
74*42 ppm nitrate and no nitrite. Thus, nitrate produced in vivo from
oxidation of absorbed nitrite may also be excreted through saliva and urine,
although quantitative estimation 1s not possible.
Donahoe (1949) reported five cases of Infant methemogloblnemla. One
occurred 1n a breast-fed newborn and one 1n an Infant fed milk from cows
that drank nitrate-contaminated water. Davlson et al. (1964) reported
nitrate levels of 5, 9 and 15 ppm In milk from dairy heifers treated orally
with nitrate at dosages of 0, 440 and 660 mg/kg/day, respectively. These
data suggest that Ingestlon of high levels of nitrate results In excretion
via the mammary gland. On the other hand, Crowley et al. (1974) reported
nitrate levels of 5.6 and 5.7 ppm 1n milk from dairy herds drinking water
containing 19 and 374 ppm nitrate, respectively. Assuming that dairy cows
drink 45 l of water/day and weigh 545 kg [crudely estimated from data
presented by Atkeson and Warren (1934)], the 374 ppm nitrate concentration
In drinking water 1s equivalent to a dosage of -30 mg/kg/day. This dosage
1s much lower than those used by Davlson et al. (1964).
0158d
5-12
07/18/89
-------
5.S. SUMMARY
Quantification of nitrite absorption from the GI tract 1s complicated by
the Instability of the 1on 1n add and reactions with diet components
(Mlrvlsh et al., 1975). The data suggest, however, that the majority of
orally administered nitrite Is absorbed from the GI tracts of mice (Friedman
et al., 1972} and rats (Frltsch et al., 1980b). Intratracheally adminis-
tered nitrite appears to be rapidly and almost completely cleared from the
lungs of mice (Parks et al., 1981; Parks and Krohn, 1983). Data In several
mammalian species strongly Indicate that, following absorption, nitrite Is
rapidly distributed throughout the body (Schneider and Yeary, 1975a,b; Parks
and Krohn, 1983; Parks et al., 1981). Nitrite has also been reported to
cross the placental barrier In rats (Gruener et al., 1973; Shuval and
Gruener, 1977; Parks and Krohn, 1983). WHh the exception of the dog
erythrocyte (Schneider and Yeary, 1975a,b), no tissue appeared to sequester
nitrite.
Ingested nitrate can be reduced to nitrite by the mlcroflora of the GI
tract (Goaz and Blswell, 1961; WUter and Ballsh, 1979) and probably by
mammalian tissues as well (Hard et al., 1986). Nitrite In the gut can react
with primary amines to cause Immediate decomposition to molecular nitrogen,
or with secondary or tertiary amines to form N~n1trosocompounds (Frank et
al., 1985; U.S. EPA, 1985; Hard et al.f 1986). The latter reactions
generally proceed faster at pH 1-3 (U.S. EPA, 1985). An exogenous source of
nitrite greater than that normally present 1n experimental animal diets
appears to be required for production of detectable levels of N-nltroso
compounds (U.S. EPA, 1985), probably because the reaction of nitrite with
primary amines Is the preferential pathway.
0158d
5-13
04/05/89
-------
Absorbed nitrite can oxidize hemoglobin to methemoglobln, resulting In
tissue anoxia (Lee, 1970; Burrows, 1979; U.S. EPA, 1985). Infants appear to
be especially sensitive to this phenomenon because their hemoglobin Is more
easily converted to methemoglobln, and their methemoglobln reductase
activities are less efficient, compared with adults. Absorbed nitrite Is
rapidly oxidized to nitrate, probably In the erythrocyte (Parks et al.,
1981; Yoshlda et al., 1983). Nitrate Is excreted largely through urine
{Yoshlda et al., 1983; Ishlwata and Tanlmura, 1982), although substantial
amounts may be recycled through saliva (Ishlwata and Tanlmura, 1982).
Excretion through the mammary gland appears to occur when high doses of
nitrate are Ingested (Donahoe, 1949; Davlson et al., 1964).
0158d
5-14
04/05/89
-------
6. EFFECTS
6.1. SYSTEMIC TOXICITY
6.1.1. Inhalation Exposure. Data were not located regarding exposures to
aerosols of nitrite Ion; however, data regarding Inhalation exposure of
animals to nitrogen dioxide gas are plentiful. Parks and Krohn (1983)
stated that Inhaled nitrogen dioxide appears to react with llplds In the
pulmonary cell membranes, abstracting a hydrogen Ion to form nitrous acid
and damaging the cell membrane. The formed nitrous add Is rapidly neutral-
ized by the abundant buffering capacity of the pulmonary fluid leaving the
nitrite anlon. If this becomes long-term exposure, the chemically Induced
damage to the membranes of the pulmonary cells may lead to an Immune
response and splenic enlargement, which Is seen 1n laboratory animals
exposed to nitrogen dioxide gas. Topical application of nitrite 1on to the
respiratory epithelium, either from Inhalation of nitrogen dioxide gas or
from Inhalation of a nitrite 1on aerosol, would probably damage the endo-
thellum of the postcaplllary venules permitting the entrance of cells from
the bloodstream Into the lungs. Regarding systemic effects, absorbed
nitrite would convert hemoglobin to methemoglobln. In addition, nitrite
absorbed from the respiratory tract and delivered by circulation to the
lumen of the Intestine could participate In nltrosatlon reactions with
secondary or tertiary amines and cause the formation of potentially carcino-
genic compounds.
As Indicated above. Parks and Krohn (1983) speculated that Inhalation of
nitrite Ion would result 1n the same local effects on the lung and systemic
effects seen with nitrogen dioxide gas. However, neither short-term nor
longer-term toxldty data are available to test this hypothesis. Further-
more, for reasons discussed later In Section 8.2.1., 1t 1s Inappropriate to
0159d
6-1
07/18/89
-------
derive an RfD for Inhalation exposure to nitrite from data on nitrogen
dioxide gas. For this reason, Inhalation data for nitrogen dioxide gas are
not reviewed herein.
6.1.2. Oral Exposure. Recent analyses (U.S. EPA, 1985, 1986a) concluded
that laboratory animals are not acceptable models for nitrite-Induced
conversion of hemoglobin to methemoglobln because animals are more resistant
than humans. This conclusion Is supported by Calabrese et al. (1983), who
demonstrated marked species differences In nitrite-Induced methemogloblnemla
and noted specifically that rats are poor models for this condition. U.S.
EPA (1985) further Indicated that the conversion of hemoglobin to methemo-
globln 1s an acute phenomenon and that Us effect does not Intensify with
continued exposure. Lljlnsky (1976) noted that oral dosages slightly lower
than those associated with acute lethality, when delivered over time 1n
drinking water, do not result 1n adverse effects even If exposure continues
over the animal's lifetime. Therefore, the oral exposure section Is divided
Into human data and animal data, rather than subchronlc and chronic data.
6.1.2.1. HUMAN DATA — Risk assessment for oral exposure to nitrite
was recently performed by the Agency 1n the derivation of HAs (U.S. EPA,
1985, 1987) and a verified oral RfD (U.S. EPA, 1986a). The derivation of
the RfD required the application of several assumptions. The first assump-
tion 1s that Ingested nitrate 1s reduced to nitrite In the GI tract of the
human (U.S. EPA, 1985). The nitrite thus formed 1s completely absorbed and
causes conversion of hemoglobin to methemoglobin. The second assumption Is
that the Infant (newborn to ~3 months old, weighing -4-6 kg) Is the most
sensitive member of the human population to nitrite-Induced methemoglobln-
emla (U.S. EPA, 1985, 1986a).
0159d
6-2
07/18/89
-------
There are several reasons for the Infant's greater sensitivity. The
higher pH of the Infant's GI tract, compared with that of other members of
the human population, permits colonization with microorganisms that more
efficiently reduce Ingested nitrate to nitrite (Swann, 1975; Lljlnsky, 1976;
Fan et a!., 1987), although the extent of this reduction has not been
experimentally quantified. U.S. EPA (1985) assumed that Infants may reduce
100% of Ingested nitrate to nitrite and that nonlnfant humans effectively
reduce about 10%. In addition, 60-80% of the circulating hemoglobin of the
newborn exists as fetal hemoglobin, which 1s more readily oxidized to
methemoglobln than 1s the adult form (Swann, 1975). Furthermore, the
methemoglobln reductase system, which enzymatlcally converts methemoglobln
back to hemoglobin, Is not as efficient 1n the Infant as 1n the nonlnfant
(Swann, 1975; Fan et al., 1987). Finally, fluid consumption of Infants
approximates 160 ml/kg/day, which Is >5 times greater than consumption of
29 ml/kg/day, or 2000 ml/day for a 70 kg adult.
Adequate data regarding the toxlclty of nitrite Ingested by humans were
not located. The laws of thermodynamics, however, dictate that all
nitrogenous substances 1n water tend to convert to nitrate (MAS, 1977b).
Acceptance of the assumptions discussed above permits derivation of an RfD
for oral exposure to nitrite from epIdemlologUal data regarding methemo-
globlnemla In Infants exposed to drinking water containing nitrate. A
comprehensive study of this nature was conducted by Walton (1951), who
analysed data for 214 cases of Infant methemogloblnemla reported 1n 17
states for which water nitrate N data were available. No cases were asso-
ciated with concentrations of nitrate N ranging from 0-10 ppm. A total of 5
cases (2.3%) were associated with concentrations of 11-20 ppm. 36 cases
(16.8%) with 21-50 ppm, 81 cases (37.8%) with 51-100 ppm and 92 cases
0159d 6-3 09/13/89
-------
(43.1%) with >100 ppm. Ohio, Oklahoma and Texas reported a substantial
number of water samples exceeding 10 ppm nitrate N, but no cases of Infant
methemogloblnemla.
Shuval and Gruener (1972) measured conversion of hemoglobin to methemo-
globln In 1702 Infants of a large region In Israel exposed to drinking water
with medium high nitrate levels averaging 50-90 ppm (11-20 ppm nitrate N).
A control group of 758 Infants In Jerusalem was exposed to drinking water
containing low nitrate levels averaging 5 ppm (1.1 ppm nitrate N). Data
were grouped for Infants 1-60 days of age, 61-90 days, >90 days and all
ages. There were no significant differences In methemoglobln levels between
Infants from the medlum-hlgh-nltrate and low-nitrate regions. There were no
significant age-related differences within regions. Infants 1n the medlum-
hlgh-nltrate region fed on powdered formula reconstituted with tap water had
somewhat higher methemoglobln levels than breast-fed Infants or those fed
whole cow's milk. The differences were not significant, however, and the
reverse trend was reported In the low-nitrate region. Infants 1n either
region suffering from diarrhea had slightly higher methemoglobln levels than
those not afflicted, but the differences were not significant. Infants <90
days of age In the medium-high region that consumed dtrus or tomato juice
had significantly lower methemoglobln levels than those not fed the Juices.
No differences were noted 1n Infants >90 days of age.
A well-controlled experiment with newborn Infants In a hospital supports
the NOAEL of 10 ppm nitrate N In the Walton (1951) study. Gruener and
Toeplltz (1975) fed Infants formula made with water containing 108 ppm
nitrate (24.4 ppm nitrate N) for 3 days. Previously, formula had been made
with water containing 15 ppm nitrate (3.4 ppm nitrate N). Conversion of
hemoglobin to methemoglobln averaged 0.9% during feeding with formula
0159d
6-4
07/18/89
-------
prepared from water containing 15 ppm nitrate, compared with 1.3% during
feeding with formula prepared from water containing 108 ppm nitrate. In
most cases methemoglobln dropped to "normal levels" on the second day of
exposure to high nitrate, suggesting that some adaptation had occurred.
Although the methemoglobln level reached 6.9, 13.9 and 15.9% In three
Infants on the second day of exposure to high nitrate, there was no clinical
evidence of methemoglob1nem1a.
Reports from the Soviet Union claimed that school children exposed to
drinking water containing high levels of nitrate N developed methemoglobln-
emla. Prompted by these reports, Craun et al. (1981) Investigated the Inci-
dence of methemogloblnemla In children 1-8 years old of selected Illinois
families whose drinking water sources were private wells. Of the partici-
pating families, 36 (with a total of 64 children) used water with 22-111 ppm
of nitrate N; 14 (38 children) used water with <10 ppm of nitrate N.
Methemoglobln expressed as percent of total hemoglobin was 0.98% 1n the
low-nitrate groups and 1.13% In the h1gh-n1trate groups. The differences
were neither biologically nor statistically significant. Methemoglobln
level did not appear to relate to gender, level of nitrate N 1n the water,
or dose of nitrate N Ingested 1n 2 or 24 hours before blood was drawn for
analysis. A small but significantly higher methemoglobln level was found 1n
children aged 1-4 years compared with those aged 5-8 years, regardless of
water nitrate N levels.
Ulnton et al. (1971) measured hemoglobin conversion to methemoglobln 1n
a group of 111 Infants aged <2 weeks-6 months that received nitrate from
drinking water at dosages of <1 mg/kg/day (n=63), 1-4.9 mg/kg/day (n=3),
5.0-9.9 mg/kg/day (n-20) and 10-15.5 mg/kg/day (n=5). From data provided by
the Investigators, It was estimated that the lowest water concentration
0159d
6-5
07/18/89
-------
associated with a dosage of 10 rag/kg/day was -50 ppm nitrate (11 ppm nitrate
N). Mean methemoglobln levels were 1.6%, compared with a normal range of
1.0-2.9%. Methemoglobln levels did not exceed the normal range except In
three Infants 1n the high-dose group. The highest level (5.3%) was measured
In a 30~day~old Infant whose nitrate dosage was estimated at 15.5 mg/kg/day.
There was no clinical evidence of methemogloblnemla.
A Soviet study suggested that CNS effects may occur 1n children who,
because of Ingestlon of h1gh-n1trate water, developed methemogloblnemla.
Petukhov and Ivanov (1970) reported slowed conditioned motor reflexes In
response to auditory and visual stimuli In 39 children exposed to drinking
water containing 105 ppm nitrate (23.7 ppm nitrate N). Average methemo-
globln levels were reported at 5.3%. Control children were exposed to water
containing 8 ppm nitrate (1.8 ppm nitrate N). Levels of methemoglobln 1n
control children were not reported.
U.S. EPA (1985) reviewed several other reports of elevated blood
methemoglobln levels In Infants, but these reports Involved complication
with diarrhea or various Infectious diseases, which may result In relatively
high levels of circulating methemoglobln. These reports are not discussed
herein.
6.1.2.2. ANIMAL DATA — Animal data are useful for characterizing the
toxldty of chronic exposure to nitrite, and suggest that conversion of
hemoglobin to methemoglobln does not Increase with Increased duration of
exposure.
Chow et al. (1980) provided drinking water containing sodium nitrate at
4000 ppm (659 ppm nitrate N), sodium nitrite at 2000 ppm (406 ppm nitrite
N), or drinking water containing no added nitrite or nitrate to groups of
9-12, 2-month-old male Sprague-Oawley rats for 14 months. Within the first
0159d 6-6 07/18/89
-------
6 months, mortality attributed to respiratory Infection occurred 1n 4/9
controls, 6/10 nitrate-exposed rats and 7/12 nitrite-exposed rats. In rats
surviving to terminal sacrifice, lung lesions were observed In 1/5 controls,
4/4 nitrate-exposed and 5/5 nitrite-exposed rats. Lesions reported 1n the
nitrite-exposed rats Included mlcroabscesses and congestion. These lesions
were more severe than those In the nitrate-exposed or control rats. It 1s
unclear 1f this evaluation was based on gross or microscopic pathological
examination. Other effects attributed to treatment Included significantly
(p<0.05) reduced body and absolute liver weights, Increased lung weights and
reduced plasma vitamin E concentration. These effects were more Intense In
the nitrite-exposed rats than In the nitrate-exposed rats. Effects observed
only 1n nitrite-exposed rats (p<0.05) were methemogloblnemla (1-35% conver-
sion of hemoglobin, compared with 0-1X In controls) and elevated erythrocyte
reduced GSH concentration, which the Investigators hypothesized may reflect
a younger population of erythrocytes.
In a second experiment, Chow et al. (1980) exposed groups of eight
1-month-old male rats to drinking water containing no test compound
(controls), 400 ppm nitrate from sodium nitrate (90 ppm nitrate N) or 200
ppm nitrite from sodium nitrite (61 ppm nitrite N) for 16 weeks to determine
effects on blood. There was no effect on body weight. Elevated lung
weights were reported In both treated groups, with the greater effect
observed In nitrite-exposed rats. Conversion of hemoglobin to methemoglobln
was <1.2% In controls and 1n nitrate-exposed rats, and 0.5-3.IX In nitrite-
exposed rats. There was no effect on erythrocyte GSH or plasma vitamin E
levels. Statistical analysis was not performed 1n this experiment, but the
Investigators concluded that nitrate and nitrite at these levels had no
effects other than those on the lungs. Fecal extracts from rats In the
0159d 6-7 07/18/89
-------
first experiment and liver extracts from rats In the second experiment were
assayed for the presence of mutagenU factors In the Ames assay with
Salmonella typhlmurlum strains TA98 and TA100 with negative results. The
Investigators concluded that the rats In the Chow et al. (1980) study were
relatively resistant to reduction of nitrate to nitrite, as well as to the
effects of nitrite.
Further evidence that rats are resistant to nitrite-Induced methemo-
globlnemla was provided by Csallany and Ayaz (1978), who exposed female
Sprague-Dawley rats to drinking water containing nitrate followed by nitrite
for 25 weeks. Groups of five 5-week-old rats on diets containing three
levels of vitamin E (none, low or high) were provided drinking water con-
taining nitrate at 200-1600 ppm (45-361 ppm nitrate N) over a 5-week period.
At the end of this period, nitrate was removed and nitrite was substituted
at 200 ppm (60 ppm nitrite N). The level of nitrite was Increased to 3000
ppm (912 ppm nitrite N) for the last 4 weeks of exposure. A control group
was maintained on a low vitamin E diet and was given drinking water without
test chemical. Body weight loss occurred In filtrate-nitrite treated rats on
diets low or lacking 1n vitamin E, but not 1»i treated rats on the high
vitamin E diet. Percent conversion of hemoglobin to methemolobln, measured
nine times during the experiment, was significantly elevated only at the
eighth sample time, during which the rats were exposed to nitrite at 2000
ppm (608 ppm nitrite N). There appeared to be no effect of vitamin E on
methemoglobin level. The Investigators concluded that methemoglobln
reductase had been Induced In the treated rats, affording substantial
protection against the effects of nitrate and nitrite.
Shuval and Gruener (1977) Investigated the effects of nitrite on blood
methemoglobln levels and motor activity In mice. Groups of fifteen
0159d 6-8 07/18/89
-------
50-day-old male C57b1/6J mice were provided with drinking water containing
sodium nitrite at 0, 100, 1000, 1500 or 2000 ppm (0, 20.3, 203, 305 or 406
ppm nitrite N, respectively) for 3 weeks. The Investigators estimated
dosages of sodium nitrite of 0, 8.8, 88, 133 or 178 mg/kg/day, respectively.
Significant (p<0.05) conversion of hemoglobin to methemoglobln was observed
at 1500 and 2000 ppm. A significant decrease In motor activity was reported
at 2000 ppm. The Investigators hypothesized that this was due to reduced
oxygen capacity of the muscles resulting from conversion of myoglobln to
metmyoglobln.
Shuval and Gruener (1977) evaluated the effects of nitrite on the brain
electrical activity of male Sabra strain rats. Groups of four rats were
provided with drinking water containing sodium nitrite at 0, 100, 300 or
2000 ppm (0, 20.3. 60.9 or 406 ppm nitrite N) for 3 weeks followed by
observation periods of 2.5-4.5 months. Sodium nitrite Intake was estimated
at 0, 14, 42 and 280 mg/kg/day In the control, low-, middle- and high-dose
groups, respectively. Elevated conversion of hemoglobin to methemoglobln
(12%) occurred only at the high dose and only during the exposure period.
High-dose rats also appeared to be more sedate than controls. The authors
reported altered brain-wave activity, the frequency of which appeared to be
dose-related, In all treatment groups. In a repeat of this experiment 1n
different facilities with different staff, there was no evidence of altered
brain-wave activity. The authors were unable to explain this discrepancy.
A French study with rats linked transient weight loss and hlstopatho-
loglc lesions of the spleen, anterior portion of the GI tract, thyroid and
kidneys with dietary concentrations of nitrate of 5X (11,300 ppm nitrate N)
or nitrite of 0.5X (1522 ppm nitrite N) (Frltsch et al., 1980b).
0159d
6-9
07/18/89
-------
Shuval and Gruener (1972) provided drinking water containing sodium
nitrite at 0, 100, 1000, 2000 or 3000 ppm (0, 20.3, 203, 406 or 609 ppm
nitrite N) to groups of eight 3-month-old male rats (strain not reported)
for 24 months for hlstopathologlc evaluation of heart, lungs, kidneys,
liver, spleen, pancreas, adrenal and brain. The authors reported sodium
nitrite dosages of 0, 13, 133, 267 or 400 mg/kg/day. A dose-related
Increase In the frequency and Intensity of lesions In the lungs, liver,
spleen and kidneys was observed at concentrations of >1000 ppm. Lesions 1n
the lungs Included bronchial dilatation and Infiltration with lyphocytes,
the presence of purulent bronchial exudate, and atrophy of the mucosal and
muscle cells. In addition, there was the occasional presence of Inter-
stitial round cells, flbrosls and emphysema. Lesions In other organs
Included congestion of the liver and spleen, and focal Inflammation and
degeneration of the kidneys. Cardiac lesions, Including thinning and
dilatation of the Intramural coronary arteries, were observed In all treated
groups, but the lesions were most striking 1n the high dose group. Rats 1n
this group also had degenerative foci In the cardiac muscle.
Shuval and Gruener (1977) exposed groups of 52 male Sabra rats to drink-
Ing water containing sodium nitrite at 0, 200, 1000, 2000 or 3000 ppm (0,
40.6, 203, 406 or 609 ppm nitrite N), or sodium nitrate at 2000 ppm (330 ppm
nitrate N) for 24 months to evaluate conversion of hemoglobin to methemo-
globln and to evaluate the hlstopathologlcal appearance of the heart. The
Investigators reasoned that methemoglobin levels might be highest at night,
because greatest consumption of food and water by rats occurs at night, and
levels might return to normal by morning because of the rapidity with which
methemoglobln 1s reduced to hemoglobin. Therefore, the light-dark cycle of
the animal rooms was reversed so that blood samples were drawn at the rats'
0159d
6-10
07/18/89
-------
"midnight." Interim sacrifices were performed every 3 months. Slight
retardation of growth and development occurred at 3000 ppm sodium nitrite.
The highest conversion rates of hemoglobin to methemoglobin were 4.44% at
1000 ppm, 11.43% at 2000 ppm and 23.9X at 3000 ppm sodium nitrite.
Dilatation and thinning of the coronary arteries was observed 1n most rats
receiving >10QQ ppm sodium nitrite and sodium nitrate and 1n -1/2 the rats
receiving sodium nitrite at 200 ppm.
Oral studies using mice suggest that chronic exposure to nitrate or
nitrite Increased the Incidence of amyloldosls In this species. Suglyama et
al. {1979) fed groups of 50 male and 50 female mice sodium nitrate at
dosages of 0, 2500 or 5000 mg/kg/day for over 1 year and reported Incidences
of amyloldosls of 25, 42 and 37% 1n survivors 1n the control, low- and high-
dose groups, respectively. The method of administration (diet or drinking
water) was not specified. Inal et al. (1979) provided groups of 50 male and
50 female mice with drinking water containing sodium nitrite at 0, 1250,
2500 or 5000 ppm (0, 254, 508 or 1015 ppm nitrite NJ for >18 months.
Dosages of sodium nitrite were estimated at 0, 206, 416 or 833 mg/kg/day In
the control, low-, middle- and high-dose groups, respectively. The authors
did not provide quantitative dose-response data but stated that the liver
appeared to be the target organ and showed marked atrophy and hemoslderosls.
Amyloldosls was reported In the liver, kidney, spleen and adrenals.
6.1.3. Other Relevant Information. Oral LD5Q values for nitrate and
nitrite In laboratory animals are compiled 1n Table 6-1. There appear to be
no marked species or cation differences 1n the acute toxlclty of nitrate or
nitrite. In rats and rabbits, nitrite Is -1 order of magnitude more toxic
than nitrate. The Intraperltoneal L05Q for nitrite of 119 mg/kg In mice
(Masukawa and Iwata, 1979) 1s similar to the oral LD5_. Tissue anoxia
that convert from hemoglobin to methemoglobin results In death. In horses
0159d
6-11
07/18/89
-------
TABLE 6-1
Oral LD50 Data for Nitrate and Nitrite
Species
Rat
Rat
Rat
Rata
Ratb
Ratc
Rat
Rat
Rabbit
Rabbit
Rabbit
Mouse
House
Mouse
Cation
Na*
Na=
Na*
Na*
Na+
Na+
K+
K+
Na+
K+
K*
Na*
K+
K +
Nitrate
(mg/kg/day)
NR
NR
NR
NR
NR
NR
1986
1986
1955
1166
1849
NR
NR
NR
Nitrite
(mg/kg/day)
100
120
57
51
87
73
NR
NR
124
108
NR
143
119
95
Reference
Ima1zum1 et al
Wlndholz, 1983
Sax, 1984
Druckrey et al
Oruckrey et al
Druckrey et al
WHO, 1962
WHO, 1962
DollahUe and
DollahHe and
Sax, 1984
Sax, 1984
WHO, 1962
WHO. 1962
., 1980
., 1963
., 1963
., 1963
Rowe, 1974
Rowe, 1974
al-year fasting
b!-year fed
c3-month fasting
NR = Not reported
0159d
6-12
04/05/89
-------
treated Intravenously with sodium nitrite, the signs of toxldty are
hypotension and elevated pulse rate, which occurred within 5 minutes of
treatment In the study by BartU (1964). These signs reflect the smooth
muscle relaxant and vasodilator properties of the nitrite 1on. Peak levels
of methemoglobln did not occur until 30 minutes after treatment.
Burden (1961} reported lethal doses of potassium nitrate and sodium
nitrite In humans at 54-462 mg/kg and 32-154 mg/kg, respectively. These
doses correspond to doses of nitrate of 33-283 mg/kg and doses of nitrite of
21-103 mg/kg.
As discussed In Section 6.1.2.1., neonatal Infants exposed to high
levels of nitrate develop methemogloblnemla from conversion of hemoglobin to
methemoglobln. Other sensitive subgroups Include pregnant women (WHO,
1984b), Individuals with low erythrocyte glucose-6-phosphate dehydrogenase
activity (Calabrese et al., 1980), and Individuals with achlorhydMa,
Including those under treatment for gastric ulcer and those suffering from
chronic gastritis or pernicious anemia (Fan et al., 1987). In addition,
those with hereditary deficiencies of methemoglobln reductase or those with
hereditary hemogloblnopathles may be unusually sensitive (Fan et al., 1987).
Although they may develop methemogloblnemla, adults acutely Intoxicated
with nitrate or nitrite also experience effects on their cardiovascular
systems. Nitrites have been used therapeutlcally at dosages of 30-300 mg to
Induce vasodllatlon (Wolff and Wasserman, 1972). Weiss et al. (1937)
experimentally Induced reversible cardiovascular collapse In a normal male
human given 2.6 mg/kg of sodium nitrite orally. Effects were not noted
while the subject remained horizontal. Raising the subject to a 75° upright
position resulted In typical cardiovascular collapse and loss of conscious-
ness. The subject promptly regained consciousness when lowered to the
horizontal position.
0159d
6-13
07/18/89
-------
Harris et al. (1979) reported cyanosis and generalized seizures In three
men derma 11y exposed to a mixture of molten sodium and potassium nitrate In
an Industrial accident. One man succumbed to Irreversible cardiac arrest.
The other two recovered, following treatment with methylene blue and massive
exchange blood transfusions. Aquanno et al. (1981) reported weakness,
sweating, nausea, throbbing and a roaring sound In the ears, palpitations,
numbness and tingling 1n two laboratory workers who accidentally salted
their breakfasts with sodium nitrite. Conversion of hemoglobin to methemo-
globin was reported at 34 and 54%. Treatment with methylene blue returned
methemoglobin levels to normal within 1 hour. Hal ley and Flanagan (1987)
reported nausea, weakness, unconsciousness and cyanosis accompanied by
hypoxla and methemogloblnemla (methemoglobln levels of 7.7-66%) In three
adults who had consumed pickled pork that contained 10,000-15,000 ppm
nitrite (3044-4567 ppm nitrite N). Treatment with oxygen and methylene blue
reversed the symptoms.
Several substances have been found to antagonize the effects of nitrite
In humans and animals. Methylene blue has been used effectively to reverse
anoxia associated with methemogloblnemla, as mentioned In the three case
reports discussed above. Methylene blue acts as a coenzyme In an alternate
NADPH-dependent methemoglobln reductase pathway (Wailey and Flanagan, 1987).
Burrows et al. (1977) reported that tolonlum chloride was more effective
than methylene blue In reversing nitrite-Induced methemogloblnemla 1n sheep.
Masukawa and Iwata (1979) reported a significant reduction 1n the toxldty
of sodium nitrite administered to mice by Intraperltoneal Injection when
treatment was accompanied by subcutaneous administration of selenlte. In
vitro studies Indicated that selenlte did not retard nitrite-Induced oxida-
tion of hemoglobin to methemoglobln, and the Investigators suggested that
0159d
6-14
04/05/89
-------
the protection observed j_n vivo was due to accelerated reduction of methemo-
globln to hemoglobin. Calabrese et al. (1983) reported a dose-related
decrease from ascorbic add 1n nitrite-Induced In vitro formation of
methemoglobln In human and rat erythrocytes, but not 1n sheep erythrocytes.
NHrlte, usually 1n combination with thlosulfate, has been used success-
fully as an antidote for cyanide poisoning (Way et al., 1984). The
rationale Is that nitrite converts hemoglobin to methemoglobln, which forms
a relatively stable complex with cyanide and reduces the cyanide available
to bind cytochrome oxldase. More recent Investigations, however, suggest
that other unidentified mechanisms exist whereby nitrite antagonizes cyanide
toxldty (Hay et al., 1984).
An Indirect mechanism of nitrite toxldty Involves reaction with
secondary and tertiary amines and amides to form N-nltroso compounds, many
of which have been shown to be hepatotoxlc (Swann, 1975; U.S. EPA. 1985).
It Is beyond the scope of this document to review the toxlclty of N-n1troso
compounds. However, many researchers have reported that ascorbic acid
significantly reduced hepatotoxIcHy or carclnogenlcHy associated with
simultaneous oral administration of nitrite and nltrosatable amines In rats
(Cardesa et al., 1974; Kamm et al., 1975; M1rv1sh, 1986; Garcia et al.,
1987). Ascorbic acid also reduced or eliminated the formation of mutagenlc
compounds In mice given oral doses of an am1nopyr1ne-n1tr1te mixture (Neale
and Solt. 1981). Ascorbic add appeared to reduce nltrosamlne formation In
the stomach rather than to Inhibit the toxldty of nltrosamlne after
formation (Cardesa et al., 1974; Neale and Solt, 1981), possibly because
ascorbic add provided an alternate substrate for the nltrosatlng anlon
(Kamm et al., 1975). Similarly, a-tocopherol has been shown to block
hepatotoxldty of n1tr1te-am1nopyr1ne mixtures In rats by providing an
0159d
6-15
07/18/89
-------
alternate substrate for the nltrosatlng anlon (Kamm et al., 1917). This
Information Is potentially useful because many orally administered thera-
peutic agents are nHrosatable amines, and simultaneous administration of
ascorbic acid or a-tocopherol may block nHrosatlon (Kamm et al., 1977).
6.2. CARCINOGENICITY
6.2.1. Inhalation. Pertinent data regarding cancer associated with
Inhalation exposure to nitrite were not located 1n the available literature
cited In Appendix A.
6.2.2. Oral. In most of the studies with nitrite, animals exposed to
nitrite served as controls In experiments designed to test the carclnogen-
Iclty of simultaneous administration of nitrite with a nltrosatable ami no
compound. Generally, these studies do not suggest a carcinogenic effect
from exposure to nitrite as the only test substance, but the studies are
Inadequate to confirm that nitrite 1s noncardnogenlc 1n animals. In some
studies, untreated control groups were not maintained; In others the
duration of exposure was Insufficient to reveal Increased risk of late-
developing tumors. Hlstopathologlc examination was often limited to a few
major organs presumed to be targets for the N-nHroso compound expected to
be formed. In most studies. It appeared the HTD had not been reached and
group sizes were too small to provide sufficient statistical power to detect
a small Increase In tumor Incidence. Representative studies using rats,
mice, guinea pigs and hamsters are summarized below. Similar studies have
been reviewed 1n U.S. EPA (1985) and NRC (1981).
Lljlnsky et al. (1973b) briefly described an experiment 1n which 30 rats
were provided with drinking water containing 2000 ppm sodium nitrite 5 days/
week. The treatment period for rats exposed to nitrite alone was not clear;
rats were observed until they died spontaneously. An untreated control
0159d
6-16
07/18/89
-------
group was not maintained, and U 1s not clear 1f evaluation was limited to
gross examination or Included hlstopathologlc examination. Although
Incidence data were not provided for tumors In the nitrite-exposed control
group, the authors did not attribute a carcinogenic response to exposure to
nitrite alone.
In what appears to be a later, and more thorough, experiment, these
researchers (Taylor and Lljlnsky, 1975; Lljlnsky and Taylor, 1977) provided
drinking water containing 2000 ppm sodium nitrite to groups of 27 male and
30 female Sprague-Dawley rats 5 days/week for 104 weeks. The rats were fed
Purina Laboratory Chow* ad libitum. Untreated controls were not main-
tained, but complete gross and microscopic pathologic examination was
performed. Based on tumor Incidence data for the nitrite-exposed control
group, Lljlnsky and Taylor (1977) concluded that the Identity and Incidence
of tumors observed were those expected 1n aged rats of this strain.
Anderson et al. (1979) provided drinking water containing 14 mM sodium
nitrite (966 ppm) 4 days/week to 20 female and 17 male weanling Swiss mice
for 6 months. Another group of 16 female and 21 male mice served as
untreated controls. The mice were fed Purina Mouse Chow* and were
observed until spontaneous death, at which time they were subjected to
necropsy and hlstopathologlcal examination of all gross lesions. There was
no evidence of a carcinogenic effect of exposure to nitrite.
In a similar experiment, Pearson et al. (1980) administered drinking
water containing 0 or 1000 ppm sodium nitrite to groups of 10 weanling mice
(strain and sex not specified), maintained on a complete laboratory chow
diet for 12 months. Four rats from each group were sacrificed after 8 weeks
and examined for the presence of tumors. Survivors were killed at 12 months
and subjected to gross examination and hlstopathologlc examination of the
0159d
6-17
07/18/89
-------
liver, lungs, heart, spleen and stomach. This experiment was repeated 3
times over the next 3 years, except that the Interim sacrifice at 8 weeks
was omitted, so that a total of 36 mice/group was examined. The tumor
Incidence In the nitrite-treated group was not significantly different from
that In the nonexposed group.
Sen et al. (1975) maintained groups of 20 male English short-hair guinea
pigs on Purina guinea pig chow* supplemented with fresh lettuce for 30
months. Drinking water containing 0 or 800 ppm sodium nitrite was also
provided. Although all guinea pigs were subjected to gross examination and
hlstopathologlc examination of nine major organs, only liver tumors were
reported. There were no liver tumors In the unexposed or nitrite-exposed
guinea pigs.
Groups of -15 male and 15 female 8-week-old Syrian hamsters were fed
powdered diets containing added sodium nitrite at 0 or 2000 ppm 5 days/week
for their lifetimes (Ernst et al., 1987). These groups served as controls
In an experiment to test the carclnogenlcHy of nitrite combined with
powdered areca nut, which Is chewed recreatlonally and contains alkaloids
that form N-nltroso compounds. At death, the hamsters were subjected to a
comprehensive gross and hlstopathologlc examination. There were no
significant differences between untreated and nitrite-exposed groups In the
Incidence of tumors.
In another study using Syrian hamsters, groups of 10 consisting of both
sexes were fed laboratory chow and provided with drinking water containing 0
or 1000 ppm sodium nitrite (Bergman and Wahlln, 1981). Necropsy and
hlstopathologlc examination of liver, gall bladder, lungs and spleen were
performed after 20 weeks. Cholanglocarclnomas were located In 5/10
nonexposed and 1n none of the nitrite-exposed hamsters.
0159d
6-18
07/18/89
-------
An Indication that nitrite may Induce tumors of the lymphoretlcular
system was reported by Shank and Newberne (1976). A group of 96 Sprague-
Dawley rats of both sexes were fed an agar-gel semi synthetic diet to which
was added sodium nitrite that resulted In consumption of 50 mg/kg/day. A
control group of 156 rats received the diet without added sodium nitrite.
Treatment began by feeding dams of the test animals at the time they were
mated. Tumors of the lymphoretlcular system occurred In 21% of nitrate-
exposed rats, compared with 6% of nonexposed controls. Tumors In organs
other than the Hver were reported In 61% of treated and 18% of control rats.
The study by Shank and Newberne (1976) prompted a much larger
FDA-sponsored Investigation of the cardnogenlclty of nitrite administered
1n a variety of modes (Newberne, 1978, 1979). The study Involved 18 groups
of 68 male and 68 female Sprague-Dawley rats (except groups 15 and 16)
treated as specified In Table 6-2. Interim sacrifices of unspecified
numbers were performed at 6, 12, 18 and 24 months. The experiment was
terminated at 26 months. Survival was generally high In all groups. The
tumor Incidences presented In Table 6-2, taken from Newberne (1979),
differed slightly from those presented 1n the earlier report (Newberne,
1978) because of a revaluation of the hlstologlcal slides. Lesions of
Interest occurred In the spleen, lymph nodes and other components of the
lymphoretlcular system. All malignant tumors of the lymphatic system were
reported as malignant lymphomas. Immunoblastlc cell proliferation, observed
In the spleen of all groups except urethane-exposed positive controls, 1s
considered by some to be a preneoplastlc lesion. The Incidence of malignant
lymphomas was higher 1n nitrite-exposed groups than their respective
controls. When data for all control groups were combined and data for all
nitrite-exposed groups were combined, the Incidences were calculated to be
5.4 and 10.2%, respectively, which was statistically significant.
0159d
6-19
07/18/89
-------
TABLE 6-2
Summary of Results for Rats Fed N1tr1tea
Proportion0 (%) of Rats with
Treatment
Semlpurlfled diet
(agar gel)d
In water (agar diet)
Positive control, 2000
ppm urethane (agar diet)
Commercial lab chow
(Purina)
Positive control, 2000
ppm urethane (chow diet)
Casein diet6
Agar diet (mothers of
groups 1 and 4)
Nitrite exposure after
weaning (agar diet)
Group
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
Sodium
Nitrite
Dose
0
250
500
1000
2000
1000
2000
NS
0
1000
2000
NS
0
1000
0
1000
0
1000
Malignant
Lymphomasc
5/136
10/136
11/136
11/136
14/136
16/136
14/136
37/135
9/132
14/134
12/132
19/136
10/136
18/136
1/33
6/34
6/136
16/131
(3
(7
(8
(8
(10
(11
(10
(27
(6
(10
(9
(14
(7
(13
(3
(17
(4
(12
.7)
.3)
• 1)
.1)
.3)
.8)
.3)
.4)
.8)
.4)
.0)
.0)
•4)
.2)
• 0)
.6)
.4)
.2)
Immunoblastlc Cell
Proliferation
10/136
9/136
23/136
14/136
23/136
17/136
18/134
0/135
5/132
1 2/1 34
11/132
0/136
10/136
11/136
8/33
8/34
21/136
18/131
(7.
(6.
(16.
(10.
(16.
(12.
(13.
3)
6)
9)
3)
9)
5)
4)
(0)
(3.
(8.
(8.
8)
9)
3)
(0)
(7.
(8.
(24.
(23.
(15.
(13.
4)
0)
2)
5)
4)
7)
aSource: Newberne, 1979
^Number of rats bearing tumors/number of rats started [except groups 8-11
In which rats that died early (not at risk) were not Included]
CA11 malignant tumors of the hypophatlc system
^Groups 1-5 subjected to Intrantervlne exposure; dams were fed these diets
5 days before parturition, offspring continued at weaning
eDr1ed agar-gel diet
NS = Not stated
0159d
6-20
07/18/89
-------
The author speculated that nitrite may have acted as an Indirect carcinogen
or promoter by stimulating overactlvlty of the Immune system.
A U.S. Government Interagency Working Group on Nitrite Research reviewed
a sample of the h1stolog1cal slides from the Newberne study (1978, 1979) and
recommended a more complete review. This review was conducted by a Joint
Committee of Experts established by The Universities Associated for Research
and Education In Pathology (UAREP), a pathology consortium of 15 universi-
ties. Their reports concluded that many of the lesions originally diagnosed
as lymphomas were extramedullary hematopolesls, plasmacytosls or h1st1ocyt1c
sarcoma (PDA, 1980a,b). The Incidence of confirmed lymphoma was -1% In
control and treated groups, comparable with the rate associated with aged
Sprague-Oawley rats In other studies. The UAREP also Interpreted the
splenic lesions originally diagnosed as Immunoblastlc cell proliferation as
extra-medullary hematopolesls, plasmacytosls or lymphold hyperplasla.
It 1s appropriate to mention animal cancer studies with nitrate, because
Ingested nitrate Is converted to nitrite 1n the human GI tract (U.S. EPA,
1985). Three animal studies were located: a lung tumor assay In strain A
mice given drinking water containing 12,300 ppm sodium nitrate (Greenblatt
and Mlrvlsh, 1973); a 2-year study using rats given sodium nitrate at 238
mg/kg/day In drinking water (Lljlnsky et al., 1973a); and a lifetime study
using mice fed diets containing 0, 25,000 and 50,000 ppm sodium nitrate
(Suglyama et al., 1979). No statistically significant elevation 1n tumor
Incidence was reported 1n any of these studies; however, an Increased
Incidence of pituitary tumors was observed In treated female rats (11/15)
compared with controls (3/15) that the Investigators described as unexpected
and difficult to explain (Lljlnsky et al., 1973a). The small numbers of
rats (15/sex/group) In this study and the lack of presented details
seriously reduced the statistical power of the experiment.
0159d
6-21
07/18/89
-------
o
A large body of literature reviewed by NRC (1981) equivocally linked
human exposure to nitrate with Increased risk of cancer of the stomach,
esophagus, nasopharynx and bladder. In all cases, however, the Increased
risk was attributed to the presence or formation of N-n1troso compounds
rather than to nitrate.
Stomach cancer has been attributed to 1ngest1on of large quantities of
salted dried fish, which contains high levels of nttrosatable secondary
amines (Singer and Lljlnsky, 1976). This may explain why Japan has the
highest reported Incidence of stomach cancer 1n the world (American Cancer
Society, 1980). Conditions such as gastric achlorhydrla, pernicious anemia,
and treatment for gastric ulcer are associated with Increased pH of the
stomach that permits the development of a resident population of micro-
organisms (Ruddell et al., 1978; Fraser et al., 1980; Forman et a!., 1985;
Cayglll et al., 1986). These mlcrooganlsms readily reduce Ingested nitrate
to nitrite, resulting 1n nitrite concentrations In gastric fluid as high as
50-100 times normal (Fraser et al., 1980). Unusually high concentrations of
nitrite 1n the stomach may result In a more efficient nltrosatlon of low
levels of nltrosatable compounds naturally present In the diet. Increasing
the risk of gastric cancer.
A high level of esophageal cancer has been reported in certain regions
In China (LI et al., 1980; Yang, 1980). Compared with populations 1n
low-Incidence regions, populations 1r high-Incidence regions consumed more
pickled vegetables, which are high In nitrite. Populations 1n high-
Incidence regions also consumed fungus-Infested corn bread. Treating this
bread with nitrite produced nltrosamlnes, suggesting that nltrosatable
substrates were produced by the fungus.
0159d 6-22 04/05/89
-------
Increased risk of nasopharyngeal cancer has been reported In parts of
China, Including Singapore and Hong Kong (Huang et al., 1978a). Low levels
of n1trosod1methylam1ne were measured In salted fish from Hong Kong.
Nasopharyngeal tumors developed In 4/20 rats fed the same fish meal (Huang
et al., 1978b).
An epldemlologlc study revealed that 6X of men wHh cancer of the
urinary bladder had a history of cystHls. This percentage Is significantly
higher than that of cystitis 1n men without bladder cancer (Wynder et al.,
1963). High levels of nHrosodlmethylamlne have been measured 1n the urine
of patients suffering from cystitis (Radomskl et al., 1978). It was hypoth-
esized that the microorganisms In the bladder during cystitis converted
urinary nitrate to nitrite, which resulted In the formation of nHrosodl-
methylamlne from dlmethylamlne that 1s ordinarily 1n urine.
More recently, however, Forman et al. (1988) relnvestlgated populations
In the United Kingdom at high and low risk of gastric cancer and found that
cancer risk correlated Inversely with both drinking water nitrate concentra-
tion and total dally nitrate Ingestlon. An epldemlologlcal examination of
>1300 men employed In the manufacture of nitrate fertilizer revealed no
evidence of Increased mortality from any cause Including gastric, esophageal
or bladder cancer. A subgroup of 10 heavily exposed workers had an ~50X
Increase In urinary excretion of five N-n1troso compounds, compared with 10
unexposed controls. The authors concluded that exposure to nitrate Is not
the determining factor In cancer risk to humans.
6.2.3. Other Relevant Information. As was mentioned In Section 6.2.2.,
many cancer studies In animals have been performed with nitrite In combina-
tion with amines. Many of these studies, reviewed by NRC (1981) and U.S.
EPA (1985), yielded positive results, presumably because of the formation of
0159d 6-23 04/05/89
-------
N-nHroso compounds from the Interaction of nitrite with amlne (Swann, 1975;
Lljlnsky, 1976). A common factor In the positive studies was the simulta-
neous administration of large nitrite doses and large doses of a nltrosat-
able ami no compound. These studies are more appropriately considered Inves-
tigations of the carclnogenlclty of the N-nltroso compounds thus formed,
rather than of the carclnogenlclty, cocarclnogenldty or cancer-promoting
ability of nitrite. Therefore, these studies are not reviewed herein.
6.3. HUTAGENICITY
Data regarding the genotoxldty of nitrite are summarized In Table 6-3.
In many of these studies, nitrite was a control In Investigations of geno-
toxlclty of nitrite combined with nltrosatable amlno compounds. Results In
bacterial tests were largely positive. Negative results were reported 1n
host-mediated assays 1n mice with S. typhlmuMum strain G46 (Couch and
Friedman, 1975; Hhong et al., 1979), but 1t appeared likely that the sodium
nitrite may never have reached the site where the test organisms were
located (NRC, 1981).
Studies for clastogenlclty In mammalian systems were clearly positive.
However, negative results were reported In dominant lethal tests 1n ratr.
treated with potassium nitrite (Jorgenson and Rushbrook, 1979) and 1n mice
treated with sodium nitrite (Teramoto et al., 1978).
In a review, Zlmmermann (1977) speculated that nitrite as nitrous acid
may exert Us genotoxldty by deamlnatlng the DNA bases, by Inducing Intra-
or Interstrand cross-links between puMne residues, or by combining with
amlno compounds to form N-nltroso compounds.
A vast body of literature reviewed by NRC (1981) unequivocally demon-
strates the mutagenlcHy of many N-nltroso compounds. It Is beyond the
scope of this document to review those data.
0159d
6-24
04/05/89
-------
4)
«j
e
rt —
W-l —
s *
O
1
m«;q
§
41
fc
Mb>
s
-
41
|
S
Activating
Systen
o
Concentrat
or Dose
C
o
i
"s
e tf
3 *-
% *
*•
4* I/I
* —
u e
«0
>»
«c
H
•o en r- r—
"* * 1 * »• * "* «
"* £ 2 JS i^§ "5 § •oS § S o S
4p 4) T3 ^^
»• i/> « e i
t** e B us vj v c
COta5«t B 41 41
O H ' Q, -O ! O «- — •* >
L. e 4i »u i. o a.ite < •< > a.
«>a>e AI 41 o ^ o •o — 41 " i"
<^ 3 S s 4lU O. C C O1 C
4iO— >"4.O-'!S>--5I O k«
t^i^U l^^u^*^^ ^"2 "* **
o e ff"*- o e g>«- S < o «i« o o 41 e
o— ioo— • o a. i- c i i- a z i- —
^* °? i iii
C/> C/) (^ (^ C^ (/) ^
* * * * *l * *
4) * vT
li 41 i^ U ••*
•2 ^. a. ^ ^^g,
5 - - * U*
? o a 55 oc a>«wi»t
vi vi vi 5 * vl i-i— ^
41
U
§s B c e -a •
e o & £ tt
iv iv *o «v ifl ^ iv
u u u i. w -o
a i a • i a |s
41 t. 4IU 4IU 41 L. 4lt-T>V
*O •- O -MO «-• O »J k.
^e 'S.J 'i.f at "a- 'a.- J^
1 1 II III
* * •* * m£ m£ m £
MC «^6 we e we w c ^ c
! 1 S 1 1 S i 1
^ _ . u> — — .in — ^
ill ;i :ili :il :il ;i :i
41 0 4*0 41 0 4) 0 410 41 O 41 §
11 li si li li si si
6-25
r» r^
C 6 'V
• ^
O «^ 3»C
e
5±
^ b.
S B.C
01 •? U
»- ^ — Z
,
• *
>i
||
e 3
^3. S
I
i«
-I
1^ S
1 1
• * S
~ l» U
w> ^ 'c
i i
^i +*
C/9IO
4) i^ u «~
^ C — 1-
c o a —
3 — 0 —
»^we
>«— 41
*c ? "* 5
o o e>
Positive
aerobic c
suggest In
reduction
*
«
„
;
!
4)
e
u
oc
04/05/
-------
Reference
e
Response
Activating
System
S
Concentrati
or Dose
llcatlon
I
>.
^ >»
c «•<
3 —
O i-
II
Indicator
Organ ISM
VI
VI
CO/4
p~ -O «
. on e •
* T» *" ' 3 * " ' " • ' r
** — .Q i I-" J «> tt 41 4) tl 41 —
E co vi «. vi — ^> p» uco u CD u OB u CD u co k CB «• > •* ia <« >a e e e o IQ vt
a> 4i o — o — e £ A .c e o C ^ x A
> •< L. a ••* c -^ c ex a. a. t- & n 19
'vix •— — o. l a. 3 Id so tl~>s^
OO)i«O l-i. -9 G t tl tt
O. <- C £41 £41 -fl- -*U ***22U<~
>> c i- to 3e4>ae4i'eZ -o 1 ^ < — 'S -o 3 ^
i— «- i o> eo^oecu L. k4>«'Q4)O'~
2_, o o u ^4<4)<-4>«l^e 4^0 ^J^^Slio
„ *
? . I . . . . . . . .
4, - 4,
•x j -^Is ^ ^g >«o
g> '01 ^ >• >. ^ c v. ^ >, *. 5 o
• • • a v -—aivi •A'v'xu-
at o ac i~ o o >D>o-e>e • 01 o ivivac
X VI m A Vl VI I— M u- i— .M •- P> ^» J< «. 1— WB X
e 4i ti ti 41
— L. L. L. U
X «* 3 3 3 9
fQ vi ** «-• •• •*
34-133 3 OIO O O13
M UUU>|> >>|U
ae £ -2 'w 'Z 1 c| e c cl 1
Z O>BvUW U— |— — — | U
i iiliii. ill
s is is s is is ss is is is s
B viBviee vie vie vie vie v*e vie e
CO
en
»(V * VI —
f\J O. ' 1— VI 1— »« ••
CLS*^O'IM r-»v< ^ ^ Vt VI 41
I i— V) CX ' r- 4) <— w — VI VI u
— .<" 2 < — ^ 52 u I! — — vi S "5 >•
— 1_ X X i— r- O >ti— U 3 £«»(/) (/> OO
" f? <5'S ° °-3 C^ Jji— i— •* •—•»-« — U ^U II
k 01 'ige ^ c ige 5 c 9B a e
— 5> IS i o le le> i o go -a
e C2 3 1* J» 3*
vi<— vi^4l iQ Q'V Q«Q Qi^ QiV Q*V fi'V kgc
13 $3 < I Is Is Is is Is is ^?l
6-26 °* "" "* "' "04/05
/89