-------
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