w '
1|

W

1

MW

\
I

109
323
THS

 

THE EFFECT 0F AMBiENT AMMONEA
Lav-m (SN {BLOOD AMMGMA AND
AMMQNIA EXCRETEQN as
MEREQW mam (awe; QAERDNERD

Thesis 50? “‘10 Dog?“ o? M. S.
MICHEGAN STATE UNIVERSITY

Janet R. Giliette
1967

 

ABSTRACT

THE EFFECT OF AMBIENT AMMONIA LEVELS ON BLOOD
AMMONIA AND AMMONIA EXCRETION OF RAINBOW
TROUT (SALMO GAIRDNERI)

by Janet R. Gillette

Blood ammonia concentrations of §a1mo gairdneri were
measured using the Conway technique and compared to the
ambient ammonia concentrations to which these fish had been
exposed for 24 hours. Nitrogen excretion rates for the fish
were measured by comparing initial and final ambient concen-
trations of ammonia, measured by the permutit method, with
initial and final total nitrogen concentrations, measured by
the micro-Kjeldahl method. Fish exposed to 0, l, 3, 5, and
8 pg/ml ammonia (NH3 + NH4+) had blood levels of approxi-
mately 38, 42, 51, 59, and 71 pg/ml ammonia, respectively.
Considering pH, these ambient solutions corresponded to ap-
proximately 0 to ling/ml unionized ammonia (NH3) with the
corresponding blood range being 20E: to #1 pg/ml unionized
ammonia. In each case the blood ammonia levels were higher
than the ambient ammonia levels. Total nitrogen excretion
rates decreased from 250 pgN/g body wt./day for fish in O
lug/ml ammonia to lBO‘pgN/g body wt./day for fish in Bing/ml

ammonia; whereas the corresponding ammonia excretion rates

Janet R. Gillette

decreased from l30‘pg/g body wt./day ammonia nitrogen to 50
pg/g body wt./day ammonia nitrogen. These data support the
theory that ammonia is excreted via passive diffusion along
a steep concentration gradient from blood to water across
the gill surface. The decreased nitrogen excretion and in-
creased blood ammonia concentrations of these fish suggest
that the rate of this passive diffusion is retarded when the
gradient is decreased; and accumulation of endogenous am-
monia appears to result. The percent of total nitrogen ex-
creted as ammonia decreased from 51.7%.at 0 pg/ml of ambient
ammonia to about 30% at Bing/ml of ambient ammonia, indi—
cating increased excretion of another nitrogenous compound(s)

in partial compensation for the reduced ammonia excretion.

THE EFFECT OF AMBIENT AMMONIA LEVELS ON BLOOD
AMMONIA AND AMMONIA EXCRETION OF RAINBOW

TROUT (SALMO GAIRDNERI)

BY

Janet R. Gillette

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Physiology

1967

“1572

N

K

ACKNOWLEDGMENTS

The author wishes to express sincere gratitude to
Dr. Paul O. Fromm for his guidance and encouragement which
made this work possible.

Deep-felt appreciation is also expressed to Joe
Abbate, Cliff Hill and the many other graduate students who
helped so much throughout this study.

The author is indebted to Dr. J. R. Hoffert for his
statistical advice.

The author is indebted also to the Federal Water
Pollution Control Administration, Department of Health Edu-
cation and welfare for their financial support of this work.

Funds were provided under grant number W P 00807.

ii

TABLE OF CONTENTS

Page

INTRODUCTION . . . . . . . . . -.- . . . . . . . . . . l

Ammonia Nitrogen Metabolism 1
General Aspects l
Excretory Forms of waste Nitrogen 5
Production of Ammonia by Excretory Organs 7

Excretion of Ammonia 10
Production of Urea 12
Toxicity of Ammonia 15
Toxicity of Ammonia to the Respiratory
Surface 15
Toxicity of Ammonia to Fish 16
Experimentation l9
METHODS AND mTERIALS O O O O O O O O O O O O O O O O 21
Experimental Animals 21
Experimental Design 22
Blood Ammonia-~Conway Method 23
Water Ammonia--Permutit Method 27
Total Nitrogen--Folin Farmer Micro-Kjeldahl
Method 27
pH Measurements 28
RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . 29
Results 29
Blood Ammonia 29
Nitrogen Excretion 29
Discussion 33
Blood Ammonia 33
Nitrogen Excretion 38
SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . 41
APPENDIX . . . . . . . . . . . . . . . . . . . . . . . 43
BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . 46

iii

LI ST OF FIGURES

Figure Page

1. Unionized ammonia concentration in the blood
as a function of unionized ammonia concen-
tration in the water . . . . . . . . . . . . 30

2. Ammonia concentration in the blood as a
function of ammonia concentration in the
water . . . . . . . . . . . . . . . . . . . 31

3. Total nitrogen and ammonia nitrogen ex—

cretion rates as a function of initial
water ammonia . . . . . . . . . . . . . . . 32

iv

INTRODUCTION AND LITERATURE REVIEW

Nitrogen forms a part of important structural and
metabolic components of all living organisms. There is a
constant turnover of this element during many catabolic and
anabolic processes. When substances such as proteins and
nucleic acids are synthesized, nitrogen is incorporated,
however, as these substances are broken down the nitrogen
may be transferred to other compounds or eventually released
in the form of ammonia. Unlike plants, animals have little
capacity for the storage of ammonia nitrogen. Since there
are large quantities of ammonia released during digestion of
foods, in addition to the small amounts released during
other metabolic activities, and since this substance is
quite toxic, excess quantities of it must be eliminated from
the animal. Animals have adapted to this situation by de-
veloping excretory mechanisms. When, however, a deficiency
in these excretory mechanisms occurs, serious complications

result.
Ammonia Nitrogen Metabolism

General Aspects

The actions of bacterial deaminases (Warren and

Newton, 1959) and bacterial ureases (Fishbein et al., 1966)

on ingested foods contribute markedly to the production of
ammonia in the gastrointestinal tract. This ammonia is then
absorbed from the alkaline intestinal fluids into the portal
circulation. Most of the absorbed ammonia (in the guinea
pig) is then absorbed by the liver before entering the peri-
pheral circulation (warren and Newton, 1959).

The absorbed ammonia has three main fates:

1. Participation in metabolic processes whereby the
following compounds are formed: amino acids,
mucleotide bases, choline and acetylcholine,
bases and phosphagens (creatine and creatine
phosphate, arginine and arginine phosphate), and
betaines.

2. Storage as glutamine.

3. Excretion as NH urea, uric acid, or trimethyl-

. . 3’
amine ox1de.

The chart which follows lists important metabolic re-
actions for incorporation, transfer and release of ammonia—
nitrogen.

Therfixation of ammonia nitrogen occurs mainly via
glutamate, glutamine and asparagine formation. The nitrogen
may then be transferred to other useful metabolic compounds
mostly via transamination reactions. In this way the am-
monia nitrogen becomes a part of such important compounds as
amino acids and nucleotides.

Ammonia is released by several deaminating enzymes
and by some physiological processes, the exact nature of

which cannot at present be explained biochemically. For ex-

ample, nervous activity, according to WGil-Malherbe (1962),

 

AMMONIA NITROGEN INCORPORATION

 

 

 

 

 

 

 

 

Site Requirement
L-glutamic acid dehydrogenase: hepatic NADH
NH3 +.x-ketoglutarate yield mitochondria
glutamate (may occur in 2 steps)
Glutamine and asparagine synthesis:
aspartate + NH3 yield asparagine ATP
glutamate + NH3 yield glutamine ATP
Synthesis of carbamyl phosphate: ATP
NH3 + C02 + ATP yield carbamyl
phosphate
NITROGEN TRANSFER REACTIONS
Site Requirement
Transamination: kidney pyridoxal
transfer of an amino group heart phosphate
for a keto group liver
Transimination kidney magnesium
transfer of amino group liver ATP
for a hydroxyl group
Transamidination:
transfer of the aminine group
of arginine to other compounds
DEAMINATION REACTIONS
Site Requirement
Hydrolytic deamination:
hydrolysis of amino acid with intestinal
production of ammonia bacteria
Oxidative deamination: kidney,
oxidation with molecular oxygen liver
to yield the keto acid and
ammonia
L-glutamic acid dehydrogenase:
glutamate yields ammonia and ADP
*keto glutarate NAD

is accompanied by a liberation of ammonia and the rate of

its release has been correlated with the intensity of

nervous activity. As early as 1922, Tashiro was able to
demonstrate production of ammonia by an electrically stimu-
lated excised frog nerve. Takashi,_§:_§l. (1961) found an
increase in ammonia content of the brain as a result of con-
vulsive activity. They suggest that this increase is a meta-
bolic consequence of functional excitation of nerve tissue.
In a recent review of the literature concerning ammonia
metabolism in the brain, weil-Malherbe (1962) reported that
the ammonia level of the brain increases as a result of pain-
ful stimulation of the extremities, and during certain con-
ditioned reflexes. These responses were noted in the ab-
sence of convulsions. Coversely, he described decreased
brain ammonia levels produced during anesthesia and during
sleep.

Weil-Malherbe postulates that oxidative phosphory—
lation is necessary for ammonia formation in the brain since
evidence indicates the importance of intact mitochondria.

In support of this view he presented data (weil-Malherbe and
Green, 1955) which showed that in the absence of oxygen or
in the presence of 2:4 dinitrophenol ammonia formation in
brain slices is reduced to the level found in homogenates
where mitochondria are no longer intact. He further sug—

gests that formation of ammonia in nervous tissue in vivo

and.l£.!£££9 is due to reactions involving proteins and
nucleoproteins.

Ammonia is produced during muscle contraction by
some poorly understood processes. Schwartz §£_§l. (1958)
have demonstrated in humans the elevation of peripheral
blood levels following convulsions due to Metrozol and after
voluntary exercise. They suggest deamination within the
muscle itself as the source of the released ammonia.
Feinberg and Alma (1961) after studying ammonia production
in isolated rabbit heart, suggest that the ammonia results
from AMP deamination. Barnes §E_§1. (1964) demonstrated a
rise in blood ammonia in rats after muscular contraction and
they showed that the rate of removal of the blood ammonia
following exercise could be increased by continued daily
exercise or daily feedings of ammonium carbonate. They sug-
gest that both of these processes "cause adaptation of the

enzymes of the ornithine cycle."

Excretory Forms of Waste Nitrogen

Waste nitrogen may be eliminated in several forms.
A few invertebrates excrete excess nitrogen in the form of
amino acids. Most aquatic invertebrates excrete their waste
nitrogen as ammonia, while terrestrial invertebrates excrete
mainly urea or uric acid. Other nitrogenous excretory pro-
ducts such as trimethylamine oxide, creatine, creatinine,

purines and pyrimidines may also be excreted. Nearly all

animals excrete nitrogen in several of these forms, but
usually one form predominates.

Baldwin (1952) postulates that the availability of
water determines whether an animal will be ammonotelic or
ureotelic. Aquatic vertebrates are usually ammonotelic.
They rid the toxic ammonia via diffusion across the gills
and by production of urine. The lung—fish is ammonotelic,
but during the dry season it estivates as a ureotelic organ—
ism. Amphibia have adapted to the terrestrial environment
by becoming ureotelic; by producing urea the animal can
store waste nitrogen in a relatively non-toxic form until
water is abundant enough that nitrogen can be eliminated via
a highly concentrated urine. The frog is ammonotelic until‘
metamorphosis and then becomes ureotelic. Some species re-
tain the ability to revert back to ammonotelism when placed
in water.

In those animals which do not have the advantage of
an aquatic environment to dilute excreted ammonia, the pro-
duction of ammonia may be deleterious if this substance is
not eliminated by diffusion fast enough to prevent its ac-
cumulation within the animal. Uric acid and urea are less
toxic than ammonia. Needham (1937) suggested that the nature
of an animal's embryonic development determines whether an
animal will be ureotelic or uricotelic. Those terrestrial
animals which hatch from eggs maintained in a moist or wet

environment produce urea which probably escapes into the

water of the immediate surroundings at a rate which is fast
enough to prevent inhibition of ontogenesis. This is the
case with some reptiles--they are ureotelic. The situation
is quite similar with mammals for the embryo is bathed in
fluid and nourished via the placenta which carries away the
relatively non-toxic urea to be excreted via the maternal
kidneys. The eggs of some animals e.g., birds and some
reptiles, however, are supplied with only enough water to
carry them through the embryonic development. These animals
must deposit their nitrogenous wastes in the form of uric
acid, an insoluble non-toxic substance that may accumulate

within the egg without harm to the embryo.

Production of Ammonia by Excretory Organs

Kidney. Ammonia production in the kidney results
mainly from breakdown of glutamate and glutamine (Van Slyke
§£_§l. 1943). Amino acids probably transfer their amino
group to ai-ketoglutarate, forming glutamate which then
undergoes deamination, resulting in release of ammonia in

the mammalian kidney (Richterich and Goldstein, 1958):

' transaminase
amino acid'+o{-ketoglutarateé——? glutamate + a(-keto acid

glutamic dehydrogenase

glutamate + NAD+ <———>d\-ketoglutarate + NH and NADH + H+

3

Pilkington (1965) demonstrated that excess ammonia and NADH
shift the glutamic dehydrogenase reaction to the left.

In dogs, the extraction of glutamine from the blood
by the kidney has been observed by Van Slyke (1943).

glutaminase
H20 + glutamine —% glutamate + NH3

 

Ammonia is released from glutamine by the action of glutamin-
ase in an essentially irreversible reaction. In the mam-
malian kidney glutaminase activity accounts for most of the
ammonia produced (Makarewicz and Zydo, 1962). Goldstein
(1966) has shown that in rats the glutaminase reaction is in—
hibited by glutamate, and Pilkington (1965) has Observed
inhibition of the reaction by ammonia as well. An addition-
al source of renal ammonia may be AMP. Makarewicz and Zydowo
(1962) observed activity of AMPbaminohydrolase and adenosine
aminohydrolase in the kidneys of all animals tested (fish,
frogs, chickens and mammals) except the tortise which has no

adenosine aminohydrolase.

‘gill. Both glutaminase and glutamic acid dehydrogen-
ase activity have been observed in the fish gill by Goldstein
and Forster (1961). Pequin and Serfaty (1966) found that
glutamate dehydrogenase may not contribute to ammonia for-

mation in carp (Cyprinus) gills. Makarewicz and Zydowo

 

(1962) suggest that the activity of AMP-aminohydrolase is

far more important than glutaminase for ammonia production

in the gill. They further suggest that the glutaminase be-

comes important for ammonia production only in higher

P .
€;\\\>X//’”9 NH3
Adenylsuccinate 7—) AMP

fumarate

animals.

Aspartate IM

.Liygr. Although some ammonia is prOduced by the
gill (Goldstein and Forster, 1961), most of the ammonia ex-
creted by ammonotelic animals such as the fish is formed by
the liver (Pequin and Serfaty, 1963). McBean §£_§1. (1966)
have shown that in the liver the amino group of alanine is
transaminated to dertoglutarate thus forming glutamate
which is then acted upon by glutamic acid dehydrogenase to
produce ammonia and :K—ketoglutarate. Their suggestion that
other amino acids follow the same reaction scheme for liber-
ation of ammonia is in accord with the suggestion of
Richerich and Goldstein (1958) concerning ammonia formation
in the kidney.

McBean g£_al. (1966) further suggest that alanine
and leucine are the only important "amine carriers" in fish
because in Pequin and Serfaty's liver perfusion tests (1963)
these two compounds were the only amino acids (other than

glutamine) which gave rise to ammonia. They discount

10

glutamine as a "carrier" since wu (1963) found that gluta-

mine synthetase is present in fish only in brain tissue.

Excretion of Ammonia

Via the_kidney. Although excretion of ammonia by
the kidney allows elimination of waste nitrogen, the main
function of ammonia in the kidney is its role in regulation
of acid—base balance. It is generally considered that the
movement of ammonia from the tubular cells to the urine is a
passive process. RObinson and Owen (1965) have proposed
that a diffusion equilibrium exists between the concen-
tration of ammonia in the fluid of the loop of Henle, vasa
recta blood and collecting duct fluid. Since only the non—
ionic moiety is diffusable the diffusion of ammonia is pH

dependent.
NH3 + H+-—-> NH4+

According to Pitts (1963), ammonia excretion depends
on four factors in the mammalian kidney: (1) Urine pH,
(2) NH3 production rate, (3) Urine flow, and (4) Tubular
cell pH. Sullivan and McVaugh (1963) postulate that blood
pH is also important.

Production of ammonia is increased during acidosis
(Pitts, 1963) and when the acidity of an animal's blood is
increased much H+ is excreted into the urine thus lowering

the urine pH. Ammonia diffuses into the urine and is

ll

trapped as the non-permeable ammonium ion. (Some blood am-
monia is trapped in the same way by the acids of the stomach
as shown by Fleshler and Gabuzda, 1965). More importantly,
however, the H+ is also trapped. As a result sodium may be
reabsorbed in exchange for H+.

Goldstein (1966) has suggested that the observed
drop in glutamate concentration which accompanies acidosis
may permit an increase in ammonia production by glutaminase
activity. The cause of this decreased glutamate level is
not clear, but Goldstein has raised the possibility that a
direct link may exist between renal gluconeogenesis and am-

monia excretion.

Via the Gill. Goldstein, §t_§l. (1961, 1964) have
shown that 60% of the ammonia excreted at the gill surface
of a marine teleost (Myoxocephalus scorpius) results from a
removal of blood ammonia by the gills. They suggest that al-
though their findings may represent a carrier mediated trans-
port, the simplest mechanism of excretion would be a non-
ionic diffusion of unionized ammonia down a steep concen-
tration gradient.

Maetz, §£_gl. (1964) gives evidence to support the
hypothesis that sodium is exchanged for ammonium across the
gill surface of a fresh—water fish, Carassius auratus. They
suggest that oxidative deamination by enzymes in the gill

cells causes production of ammonia in molecular form (NH3).

12

This ammonia is then converted to ammonium by addition of
protons liberated from dissociation of carbonic acid. They
propose an exchange of the ammonium ions for sodium ions and
suggest, alternatively, that ammonia and the hydrogen ions
originating from carbonic acid diffuse independently through
the membrane in exchange for sodium ions with the protons
then being captured by ammonia on the outside. They suggest,
without conclusive data, that the influx of sodium ions de—
creases when external water ammonium is increased and that
when the outside ammonium concentration exceeds the cellular
ammonium concentration an efflux of sodium iOns occurs.

Of related interest are the experiments of Ryberg
(1948) which support the idea that NH4+ and H+ are excreted
in the kidney by exchange.with cations from the tubular

fluid. He postulates an active exchange of NH4+ for sodium.

Production of Urea

The liver is the main site of urea synthesis (White
.§£_§l., 1964) and in ureotelic animals contains all the en-
zymes of the ornithine cycle. This cycle requires 4 ATP per
turn. A key enzyme is arginase which allows the irreversible
formation of urea and ornithine from arginine. Since all
cells contain the amino acid arginine, any cell which has
arginase as well can produce urea. In 1959, Sporn et a1.
noticed urea synthesis in rat brain tissue. Later, Buniatian

and Dautian (1966) observed pronounced arginase activity in

l3

rat, frog, and chicken brain tissue. This tissue also con-
tains argininosuccinate synthetase and arginosuccinase--
enzymes necessary for synthesis of arginine from citrulline
and aspartate, but the activity of the latter two enzymes in'
frog and chicken brain is very low, and the activity of
arginase in the chicken brain decreases during embryonic de-
velopment. The complete ornithine cycle in the brain tissue
of these three animals is not present, however, because in
each case the enzyme system synthesizing citrulline from C02,
NH3 and ornithine is absent.

The frog Rana catesbeiana represents an interesting
intermediate between an ammonotelic animal like the fish and
a ureotelic animal such as the rat. During metamorphosis
this animal changes from an ammonotelic to a ureotelic organ—
ism. Brown gt_§l. (1959) has observed a marked enhancement
in the activities of all the liver enzymes of the urea cycle
at the onset of metamorphosis in this species. They suggest
induction of the urea cycle caused an increased_g§.§gyg
synthesis of the necessary enzymes. Prior to metamorphosis,
the frog, like other ammonotelic animals, does not contain
measurable or detectable quantities of carbamyl phosphate
synthetase (Brown and Cohen, 1960). During metamorphosis,
however, the level of this enzyme increases and shows activi-
ty (Cohen and Sallach, 1961, Florkin~and Mason, 1964). White
.§E_§l° (1964) suggests that non-ureotelic animals contain

carbamyl kinase which allows a reversible synthesis of

l4

   
  

CO2 + NH3
ATP ATP
M(
CARBAMYL CARBAMYL
IMOHPHMTE’ KMMSE
SYN TI-IE TASE - .
ADP C—OH
p + 2ADP NH
¢H2
9H2
? EH2 COOH
' = H —NH NH -
HZN C OPOB COOH2' 2 CH2
COOH
CARBAMYL PHOSPHATE CITRULLINE
ASPARTATE

 

 

/ ORNITHINE
NH 4 NH
O? CYCLE fl cam
¢H c -NH—cH
TH
Hg-NHZ CH2 OOH
OOH C32
¢H2
ORNITHINE HC-NHZ
COOH
ARGINOSUCCINATE
2’.
“9%
‘ \fip OOH
2 t?“ 2H
'NH2 C-NH M1
lé=o NH H
N32 ¢H H?
CH COOH
UREA
Hc-NH2 FUMARATE
:COOH

ARGININE

l4

  
  

   

 

2 3
ATP ATP
CARBAMYE CARBAMYL.
PHOSPHATE KMMSE
SYNTWE7ASE NH'
ADP C—OH
P + 2ADP NH
9H2
9H2
O EH2 COOH
' = H -NH NH -
HZN C 0P03 COOHZ 2 CH2
COOH
CARBAMYL PHOSPHATE CITRULLINE
/ O R N I T H I N E
NH NH
$3 CYCLE H cum
9H c -NH-cH
9H NH CH2
Hg-NHZ CH2 cOOH
OOH CH2
¢H2
ORNITHINE Hc-NH2
2
=0
.2
UREA
HC-NHZ FUMARATE
COOH

ARGININE

15

carbamyl phosphate thus permitting return of the N to the
NH3 pool.

Increased g§_ngyg synthesis has also been found for
the enzymes arginosuccinate synthetase and arginosuccinase.
These enzymes catalyze rate-limiting Steps of the ornithine

cycle.
Toxicity of Ammonia

Toxicity of Ammonia to the
Respiratory Surface

Toxic effects of ammonia are related both to concen—
tration and length of exposure. When 6 men were exposed to
50 ppm ammonia gas for several hours, they suffered con—
siderable discomfort manifest by excessive nasal and lacrimal
secretion and coughing (Anderson_g£_§l., 1964). High concen-
trations of ammonia may lead to intense congestion and swell—
ing of the upper respiratory passages. Death may result
from spasm or edema of the larynx (Henderson §E_al., 1943).

When chickens were exposed to 200 ppm ammonia gas
they exhibited edema, congestion, dilation of veins and some
hemorrhage (Anderson_gt_§l., 1964). When chickens were ex—
posed to 20 ppm ammonia these workers could not find visible
effects due to ammonia until after six weeks of exposure.
Deleterious effects due to ammonia were indicated, however,

when chickens exposed to 20 ppm ammonia for 72 hours were

16

also exposed to Newcastle disease virus. The infection rate
for this disease was increased.

Burrows (1964) found similar responses in fish ex-
posed to sublethal levels of dissolved ambient ammonia in
their water. These fish exhibited a decrease in stamina and
an increased incidence of bacterial gill disease. Histologi-
cal examination revealed proliferation and consolidation of
the gill lamellae. An increase in mucus secretion was also
noted.

The above effects may result from external irri-
tation of ammonia at the respiratory surface or from internal
effects due to a rise in ammonia concentration within the
animal. Ammonia is soluble in water and could possibly dis-
solve in the moist respiratory surface. If this happens it
is possible that the ammonia could diffuse inward and be dis-

tributed within the animal via the blood stream.

Toxicity of Ammonia to Fish

Ammonia, the main nitrogenous excretory product of
fish_(Denis, 1913—14; Smith, 1929 and Fromm, 1963), and one
of the more common pollutants discharged into streams (Klein,
1959 and Lloyd, 1961) is very toxic to fish. Numerous obser-
vations have been made concerning the toxicity of ammonia,

but the mechanism of its action is not yet understood.

General aspects of ammonia toxicity. It is the

unionized form (NH3) ambient ammonia which appears to be

l7

toxic. Thus, the concentration of toxic ammonia is an aque-
ous medium depends on both the pH (Ellis, 1937; Grindley,
1946; wuhrmann, 1948 and Lloyd, 1961b) and temperature
(wuhrmann, 1953). Hydrogen ion concentrations between pH
5.0 and 9.0 are not lethal to fish (DOudoroff and Katz,
1950).

The severity of ammonia toxicity is also increased by
raising the temperature of an ammonia solution, and in turn
that of the fish, thereby causing increased metabolic rate.
Increased metabolism causes increased respiratory flow which,
according to Lloyd and Herbert (1960), may increase the rate
at which the toxic ammonia is absorbed. At the same time a
rise in temperature stimulates a greater production of am—
monia by the animal. This temperature effect must be con-
sidered when fish are to be safely transported in sealed
containers (Summerfelt, 1967; McFarland, 1960 and Gebhards,
1965). Temperature has a direct effect on an ammonia solu-
tion by decreasing the pKa value and thus, influencing the
concentration of unionized ammonia.

Carbon dioxide and oxygen content may significantly
alter the toxicity of a given concentration of unionized am-
monia. Downing and Merkens (1955, 1957), Downing (1957),
Lloyd (1961a, 1961b), Burrows (1964) and Gebhards (1965) all
have shown that decreased oxygen availability greatly en-
hances ammonia toxicity. Moreover, carbon dioxide accumu-

lation will increase the toxicity (Alabaster, 1957; Downing

l8

and Merkens, 1957; Lloyd and Herbert, 1960 and Gebhards,
1965), but its effect can be counteracted by raising the
levels of oxygen in the water (Gebhards, 1955).

Warren and Shenker (1962) have found that pH changes
caused by 002 do not produce the same effects as do pH
changes resulting from addition of a fixed acid or base.

They attribute this difference to the ability of CO2 to free—

ly diffuse across the membrane of mammalian cells. This

diffusable CO lowers the pH on both sides of the cell mem-

2
brane. Warren and Schenker also assume that only the union-
ized ammonia is freely diffusable. Thus, when the cells are
bathed by an ammonia solution the concentration gradient of
the unionized ammonia determines what movement of ammonia
will occur. Since, reduction of pH in the bathing solution
by addition of CO2 may be concurrent with a similar reduction
of pH within the cells, the concentration of unionized am-
monia is decreased on both sides. When fixed acid is intro-
duced into the bathing media, however, the concentration of
ammonia is decreased only on the outside. An efflux of am-
monia results. These workers suggest that similar processes
effect the transfer of ammonia across the gill surface when
CO2 is added to aquarium water containing ammonia.

If unionized ammonia effects the oxygen transport
system of the blood as suggested by Brockway (1950), in-

creased carbon dioxide and/or decreased oxygen could certain-

ly amplify the toxic effects of this substance as would

l9

temperature increases. Wuhrmann (1948) and Gebhards (1965)
infer that ammonia increases fragility of erythrocytes.
These suggestions are in accord with the hypothesis of
Brockway.

Other workers have observed an effect on the nervous
system due to the increased levels of ammonia. wuhrmann
(1948) reports that the first noticeable effect of ammonia
poisoning is on the nervous system causing "ammonia cramps"
and then convulsions. These observations are in accord with
a previous report by McCay and Vars (1930) who relate that
fish exposed to toxic levels of ammonia suddenly swim madly

about the aquarium and may even leap from the water.

Experimentation

If, as suggested, the unionized ammonia affects both
the oxygen carrying capacity of the bloOd and the nervous
system, it appears probable that blood levels of unionized
ammonia are increased when fish are placed in ammonia solu-
tions. This rise in blood ammonia could be accounted for by
an inhibition of the excretory process whereby ammonia is
diffused outward at the gill, and/or by an inward diffusion
of ammonia. Accumulation of ammonia within the fish may be
of greater importance than the external effects of ammonia
mentioned above.

The purpose of the work described here was to demon-

strate what changes occur in blood ammonia levels and in

20

ammonia (or ammonium) excretory rate when fish are exposed

to increased concentrations of ammonia. Both total ammonia
(NH3 and NH4+) and unionized ammonia levels in the blood and
in ambient water solutions are compared to show whether or
not blood levels increase in the fish exposed to the high
ammonia concentrations. Data for 24 hour-excretion of total
nitrogen and ammonia nitrogen were obtained in order to de-
termine if any change in rate or form of nitrogenous waste
excretion occurred in fish exposed to various concentrations

of ambient ammonia.

METHODS AND MATERIALS
Experimental Animals

The animals used were rainbow trout (Salmo gaidneri),
obtained from the Michigan Conservation Department in
Grayling, Michigan. They were transported to Michigan State
University in a large galvanized metal tank Which was coated
inside with non—toxic paint. The water was agitated to pro-
vide continual aeration.

The fish were kept in rooms maintained at 12-130C.
The lights were held on a daily cycle which provided light
from 7 AM to 9 PM.

The fish were stored in 300 liter fiberglass covered
wooden tanks. Tap water was charcoal filtered to remove
chlorine and excess iron, and run into the tanks via a flow-
through system which allowed continuous exchange of the
water. Aeration was provided via air lines held at the
bottom of the tank. Small charcoal filters were inserted
into each air line to trap oil.

Twice a week the fish were fed commercial trout
pellets. Before a group of fish was used in an experiment
the fish were transferred to small (100 liter) tanks held

under the same water, light and temperature conditions.

21

22

These fish were starved for 7 to 10 days to allow a leveling
off of the precipitous drop in both ammonia nitrogen and
total nitrogen excretion which occurs during the first 6 days
of starvation (Fromm, 1963). Thus, the nutritional status

of the fish was controlled and at the same time fecal depo—

sition which might interfere with the experiment was reduced.
Experimentallpesign

Starved fish were placed in small plastic tanks con-
taining 3.5 liters of aerated tap water containing the followe

ing concentrations of ammonia:

O‘Pg/ml
l Pg/ml
3 }Ig/ml or 5 pg/ml
8 pg/ml

Eight fish were used at a time with 2 fish at each concen-
tration. The fish tanks were covered, and a curtain was
drawn in front of the fish to prevent excitation of the fish
when people entered the room. This room was maintained with
the same temperature and light conditions prevailing as were
in the fish storage rooms.

After 24 hours the fish were anesthesized with tri-
cainmethane sulfonate (MS 222; Sandoz Pharmaceuticals,
Hanover, N.J.). Blood samples were taken by syringe via

puncture of the dorsal aortic arch and were placed immediately

23

into Conway diffusion dishes for analysis of blood amonia
nitrogen. This analysis was carried out in the same room

at 10-12°C. Water samples were collected for ammonia nitro-
gen, total nitrogen and pH determinations. Since nitrogen
excretion varies with body weight,the weight of each fish

was recorded.
Blood Ammonia--Conw§y Method

The method for determining blood ammonia was based
on the work of Conway (1933, 1939 and 1963) and of Tashiro
(1922). The modified Conway units were designed by Obrink
(1955). These units were made of polypropylene, a chemically

inert plastic.

 

middle ring

/éé%y center well

outer ring

 

Conway Unit

Reagents:
Tashiro's Reagent. Two hundred ml. of a 0.1% alco-

holic solution were prepared. To this were added 50 m1. of

24

a 0.1% alcoholic solution of methylene blue. The solution
was then transferred into a brown bottle where it may be

kept indefinitely.

Standard HCl with_indicator. Five ml. of Tashiro's
reagent were run into a 500 m1. flask. .One hundred ml. of
absolute alcohol were added. The flask was filled up to
about 3/4 the volume with ammonia free distilled water con-
taining 0.005% Tergitol nonionic NPX detergent (Union Car-
bide Chemicals Co.). The indicator was brought to a neutral
point by dropping in a little dilute alkali until the red
color was gone. To this were added 33.3 ml. of 0.1N stock
solution of HCL. More ammonia free water was then added un-
til the final solution was brought up to the 500 ml. mark.
(Ammonia free water was prepared by adding permutit .to the

distilled water carboys and then siphoning water off the top.)

Stock barium hydroxide. A 0.1 N barium hydroxide
solution was prepared by dissolving 1.5775 g of Ba(OH)2
(formula weight of Ba(OH)2°8H20 = 315.5) in 100 ml. of am-
monia free water. Precipitates (barium carbonate) which

formed were immediately filtered out.

0.014 N. barium hydroxide. Into a 250 ml. volumetric
flask were poured 35.7 ml of the 0.1 N barium hydroxide stock
solution. Ammonia free water was then added to the mark.
This solution was poured into a burette reservior, and a

sodalime trap was placed over the top opening to prevent CO2

25

from entering. The solution was then titrated against a
0.001 N HCl solution containing phenolphthalein. If the
barium hydroxide solution was not exactly 0.014 N. it was
discarded. A fresh solution was prepared approximately every

two days.

50% KOH. A 50% KOH solution was prepared by dis-
solving 50 grams of KOH in water and bringing the volume up

to 100 m1.

Procedure. The Conway units were prepared by adding

5

 

1 ml. of a 0.0067 N HCl solution containing Tashiro's indi-
cator into the center well via a microburette. One ml. of
50% KOH was added to both the middle ring and the outer ring.
The cover was placed ajar on the unit so that it did not
rest in the outer ring of KOH. One half m1. of blood sample
was then placed across from the KOH in the middle ring im-
mediately after withdrawal of the sample from the fish. The
cover was placed in position and twisted until a seal of KOH
was obtained. The unit was then rotated carefully to allow
mixing of the KOH and blood sample.

The strong base allows conversion of all the blood
ammonia to the NH3 form. This NH3 is volatile and diffuses
into the center well of acid which traps the NH3 as NH4+o
The NH3 cannot escape from the unit because of the KOH seal.

After exactly twenty minutes the unit was opened and

placed on a magnetic stirrer. A tiny magnet was placed in

26

the center well. Barium hydroxide was titrated via a micro-
burette into the pink acid until a green color was visible.
The amount of barium hydroxide used was recorded.

Before each series of blood samples were run the
barium hydroxide solution was titrated against a 0.001 N.
HCl solution containing phenolphthalein to check the
normality of the base.

The ammonia that is trapped in the center well binds
free H+ thereby increasing the pH. Therefore, the acid from
the units containing the higher concentrations of ammonia,
required less barium hydroxide to reach the end point (green
color). The volume of barium hydroxide used to titrate the
blank (Olpg/ml) was subtracted from the volume used for each
of the standards and each of the samples. A standard curve
was obtained using NH4C1 standards of 0, 20, 40, and 60’pg

+
NH4 /m1.

Cleaning. The conway units were washed several
times with tap water and soaked for a minimum of 30 minutes
in a 0.001 N HCl solution with Tashiro's indicator. Pro-
viding the indicator did not change from pink to green, the
dishes were then rinsed twice with distilled water and then
two more times with ammonia-free distilled water. It was

imperative that the dishes be clean in a "pH sense."

27
Water Ammonia--Permutit Method

Water ammonia was measured using the permutit method
as described in Oser (1965) with sample sizes of 40 or 80 ml.
depending on the approximate concentration to be meaSured.
Ten ml. of Nessler's reagent were added and the samples to
be read were brought up to 50 ml. with ammonia free water be-
fore being measured on a Coleman spectrophotometer at 480 or
410 mp.

NOte: Initially 480 mp was the wavelength used as
suggested by Oser (1965). Later, however an absorption
spectrum was run on a sample using the Beckman Model D B-G
grating recording spectrophotometer. This instrument re-
vealed a definite peak absorbance at 400 mp. Since the Cole-
man spectrophotometer the blank could not be set at 100%
transmission at a wavelength of 400 mp, a wavelength of 410
was used instead.

A standard curve was established for both of the
wavelengths used. The standards ranged from 0 to lolpg/ml

+

NH4

Total Nitrogen——Folin Farmer
Micro-Kjeldahl Method
Water samples were digested with concentrated sul-
furic acid and hydrogen peroxide. Then, after steam distil-
lation of these samples the total nitrogen was measured by

direct Nesslerization. The procedure is outlined in Oser

28

(1965). A standard curve was prepared for 0 to 20‘pg/ml.
nitrogen.
Ammonia nitrogen produced as a result of bacterial

deamination of any excreta present in the water samples has

been found to be insignicant (Brown, 1958) and therefore,

was not considered during these experiments.
,pH Measurements

A Beckman Model G pH meter was used for all pH

measurements. The pH of the blood was assumed to be 7.6 as

found by Hoffert and Fromm (1966).

RESULTS AND DISCUSSION
Results

Blood Ammonia

When rainbow trout were subjected to increasing
concentrations of unionized ammonia the levels of unionized
ammonia in the blood increased. This relationship is linear.
In each case the concentration of blood ammonia was higher
than ammonia levels in the water from which the fish were
taken. (Figure I).

Although pH variations in the water samples collected

resulted in differing prOportions of NH and NH +, the NH

3 3
concentration in a sample of higher total ammonia was nearly
always greater than the NH3 concentration in‘a sample of
lower total ammonia. For this reason the total ammonia (NH3

and NH4+) values for blood and water demonstrate the same

relationships as for the unionized ammonia (Figure 2).

Nitrogen Excretion

As the water ammonia levels increased the total
nitrogen excretion rates decreased (Figure 3). The ammonia
excretion values decreased also but at a faster rate indi-

cating that the reduction in ammonia excretion is to some

29

30

 

 

 

 

Y = 0.630 + 675 X

S = 0.175

 

in»
‘vi
0
3
w»
c
-H
m 'é'
-H
a}
§5l
'O
m
N
-H
c
o
-H
c
D
0.24
0.0
Figure l.

 

0.2 0.11 CE OS

Unionized Ammonia in water (NH3)
)19/ ml

Unionized ammonia concentration in the blOOd as a
function of unionized ammonia concentration in
the water. The solid line was drawn by the
method of least squares. Ninety five percent
confidence limits for any point on this line are
indicated by the dotted lines. The unionized am-
monia values of the water were obtained by taking
an average of the initial unionized ammonia level
and the final level (after twenty four hours).

31

 

 

 

 

100
'Egn 80
4.
gm
”fig 60
(U
'8}
E 2.
3 40
O
H
a:
20‘ 38.85 + 4.025 x

 

2 4 6 e 10 i2 14
water Ammonia (NH3 + NH4+)
.Mg/ml

Figure 2. Ammonia concentration in the blood as a function
of ammonia concentration in the water. The solid
line was drawn by the method of least—squares.
Ninety-five percent confidence limits for any
point on this line are indicated by the dotted
lines. The ammonia values of the water were ob-
tained by taking an average of the initial ammonia
level and the final level (after twenty four
hours).

32

E Ammonia Nitrogen

E’E Total Nitrogen

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

200 {15: I
:0 _l
r. .
3:5: : .3 .,
‘ ' 0". ' '0 ..° ‘ .'
3 - '5 (51%) :5": : t i
‘0 >1 ’ . . .0 .3, ...'.. 0. o
‘3‘ r3 :2”? (55 ' 5%) {'1': “33:2
8 \ . m- 1"": “'1‘... {0.3.27
"3 E“ .'.' 33:. T 1": :. 22';
m z 3.1.; .:; (55 . 0%) 1:;
H 5‘ °: . a "-3. 5 °°° {“3-
U a ".o... '0‘: . C... ‘0...
a 3...... . :z: o" E {03 .:
100 3:: ~12: a..- :3,
33';- :if -
:ss' :5 '.'~'.~‘ :14. (29 . 4%)
1 .0 .q :00. .2: o :0
3n 0". '. 0.. 1 o ... .
:0: . I'm: '0‘... :' :3 ..
3.2-: 2::
.5 Q.: :0. :: 20:0. :3::0
w ;.":« 4:: 5::
".32 :5': 15:;
2:3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

O
(I)

1' 2‘ 3 4: 5 a 7+
Initial Water Ammonia (NH3 + NH4 )
Jig/ml

Figure 3. Total nitrogen and ammonia nitrogen excretion
rates as a function of initial water ammonia.
The percent of total nitrogen excreted is ammonia

(NH3 + NH4+) is given in parenthesis.

33

extent being compensated for by an increased excretion of

some other nitrogenous compound.

Units for expression of nitrogen excretion. Nitro-
gen excretion values were obtained by measuring water nitro-
gen in units of_ngNH4+/ml. The initial values were sub-
tracted from the final values after 24 hours had elapsed to
give pgNH4+/ml/day. This value was then multiplied by the
volume of water in the tank and divided by the weight of the
fish, give units of ugNH4+/g body weight/day. To convert
into units of ugN/g body weight/day, all values were multi-

plied by 0.78. (NH4+ molecular weight = 18, N molecular

weight = 14; = .78.)

Fri
~Je

We

Blood Ammonia

When the water ammonia concentration was increased,
a rise in the blood levels of ammonia was observed and yet
the blood levels of ammonia were consistently higher than the
water levels of ammonia. These findings indicate that union—
ized ammonia may be excreted via a passive diffusion along a
steep concentration gradient from blood to water across the
gill surface as postulated by Goldstein (1964). Reduction
of this gradient by increasing ambient ammonia concentration,
which led to a decrease in rate of ammonia excretion also

supports the idea that ammonia is excreted via a passive

34

diffusion. Ammonia is apparently produced at a faster rate
than it can be eliminated under these conditions and blood
levels of ammonia rise. The factors controlling gill ex-
cretion of ammonia seem to parallel those factors which
Pitts (1963) suggests control ammonia excretion in the A
kidney.

Since some of the blood ammonia results from 7'
deamination of glutamate in a reaction which yields hydrogen
ions in addition to NH3, the hydrogen ions produced must be
handled by the organism to prevent significant alteration of
pH. Removal of the hydrogen ions might be accomplished by
the action of buffers, by utilization of the ion in synthetic
reactions or by excretion of the ion——perhaps by the mechan-
isms which Maetz (1964) described.for the gill.

Toxic responses due to increased blood levels of
ammonia in fish have not yet been investigated; however,
this aspect of ammonia toxicity has been widely studied in
several mammals. Rosado,_gg_§l. (1962) found that when blood
levels of ammonia are raised by injection of NH4C1 into the
blood stream of dogs, much of the ammonia is rapidly removed
by muscle (75%), liver (2.5%) and brain (0.5%) tissues.

They suggest that the uptake of ammonia by muscle is an
active process involving the exchange of NH4+ for KI ions.
Glutamine synthesis was noted in the brain and liver cells,
but muscle did not appear to be capable of synthesizing this

compound. The trapped or bound ammonia is then released by

35

muscle cells and brain cells at a rate which is compatible
with ornithine cycle activity for urea synthesis in the
liver.

The ammonia concentration of brain tissue depends on
the metabolic processes in the brain Which produce ammonia,
the regulatory nature of the blood brain barrier through
which blood ammonia must pass if it is to enter brain
tissues, and the detoxification mechanisms of the brain
which bind ammonia into non-toxic compounds (Navazio, gt_al.,
1961). Distribution of ammonia across the blood-cerebrospinal
fluid barrier in humans follows the nonionic diffusion theory
(Moore, §E_al., 1963). According to Navazio, gtyal, (1961)
the pH of rat brain tissue is lower than that of rat blood,
and therefore the concentration of ammonia (NH3 and NH4+) in
is

the brain tends to be higher than in the blood because NH3

trapped as NH + at the low pH.

4
Some of the ammonia detoxification mechanisms which
have been Observed in mammalian brain tissues are summarized
by Weil-Malherbe (1962). The main "binding mechanism" is a
combination of cK-ketoglutarate with one molecule of ammonia
to form glutamate and then with another molecule of ammonia
in an endergonic process to form glutamine, a compound which
more easily traverses the blood brain barrier than does
glutamate. This compound may represent an electrically

neutral transport form of ammonia. Ammonia may also give

rise to aspartate and alanine as a result of transamination

36

reactions involving oxaloacetate and pyruvate. These de-
toxification reactions may lead to decreased ATP levels in
the brain and interference with the citric acid cycle.

Under these conditions respiration and glycolysis are stimu-
lated, but with decreased function of the Krebs' cycle, ac-
cumulation of pyruvate and lactate results. According to
wail-Malherbe the toxicity of ammonia is so severe that
"intracellular accumulation of ammonia must be prevented
even though a high price may have to be paid for it" and in
many cases the toxic effects associated with increased blood
levels of ammonia are in fact due to changes brought about
by functioning of the detoxification mechanisms.

McKhann and Tower (1961), however, propose. that a
primary toxic effect of ammonia in the brain may be direct
interference with the oxidative decarboxylation of pyruvic
acid and tX-ketoglutarate. These two reactions are quite
similar and involve the same cofactors, such as ATP and thia—
mine pyrophosphate. These workers relate an interesting com-
munication with Dr. K. S. warren in which an increased poten-
cy of ammonia toxicity in thiamine deficient rats is de-
scribed. The glutamic dehydrogenase and glutamine synthetase
reactions are protective mechanisms, according to McKhann
and Tower (1961), which help maintain low levels of free
ammonia and preserve oxidative metabolism. Moreover they
suggest that an alternative pathway between the ai-keto-

glutarate to succinate stage (present only in the central

37

nervous system gray areas) of the citric acid cycle may inde-
pendently support oxidative metabolism. This would provide

a "safety-valve" mechanism for maintenance of metabolism
arOund any block of «drketoglutarate oxidative de—
carboxylation.

Schenker and Mendelson (1964) report normal ATP
levels in rats with ammonia induced coma, suggesting no
general defect in cerebral ATP synthesis nor widespread uti-
lization of this compound. ATP’may, however, be replenished
by substances such as creatine phosphate, but these compounds
were not measured. McKhann and Tower (1961) found no inhi-
bition of the electron transport chain due to ammonia. As
mentioned by Schenker and Mendelson (1964), there might be
some compartmentalization for ammonia metabolism, and thus
changes in ATP concentration within localized areas may
occur.

When ammonia is not detoxified within mammalian sys-
tems, physiological changes may result. These changes are
manifest by increased respiration (Schwartz, gt_al., 1958 and
Rosado, et al., 1962), spastic muscular contractions (Green-
stein, 1956), occasional tetanus (Navazio,_e£_al., 1961),
and coma (Navazio, g§_al., 1961). Warren and Schenker (1960)
state that there may be a synergistic effect of hypoxia on
ammonia toxicity. They correlate with this the observation

that the neurological syndrome that most closely resembles

38

cOma, whiCh results frOm increased ammonia due to liver
disease, is chronic pulmonary insufficiency.

Although it is well established that ammonia affects
the nervous system of mammals little is known concerning the
effect of ammonia on the nervous systems of lower animals.
However, in yitrg a depolarizing effect of ammonium ions on
frog neurons (Lorente de No', §£_al., 1967) and squid giant
axons (Tasaki, g£_al., 1965, 1966) has been noted.

Perhaps in fish also, the toxic effect of increased
blood ammonia levels is manifest by changes in the nervous
system. In the present work and in previous work (wuhrmann,
1948 and McCay and Vars, 1930) fish placed in high concen-
trations of ammonia were seen to be hyperexcitable.

Other than to show that blood levels of ammonia rise,
the present study on fish does not provide evidence to sup-
port or to disclaim the hypothesis of Brockway (1950) that
ammonia affects the oxygen carrying capacity of the blood.
The observed enhancing effects of increased carbon dioxide
(Alabaster, 1957; Downing and Merkens, 1957; Lloyd and
Herbert, 1960 and Gebhards, 1965) and/or decreased oxygen
(Downing and Merkens, 1955, 1957; Downing 1957; Lloyd, 1961a,
1961b; Burrows, 1964; and Gebhards, 1965) on the toxicity of
ammonia may be due to the direct effect of ammonia on oxygen
carrying mechanisms as Brockway (1950) suggests, or to a
synergistic effect which varies concentrations of carbon

dioxide, oxygen and ammonia exert on the nervous system.

39
Nitrogen Excretion

Data presented lend support to the idea that ammonia
is excreted via passive diffusion along a concentration
gradient across the gill from blood to water. When fish are
placed in ammonia solutions the gradient is reduced and am—
monia cannot be eliminated fast enough to remove excess am-
monia produced by the animal. This hypothesis, based on the
occurrence of high blood ammonia values, is supported by the
nitrogen excretion data which showed decreased nitrogen ex-
cretion rates in fish which were exposed to ammonia solutions.

In addition, the data indicate increased excretion
of some other nitrogenous compound in partial compensation
for the reduced excretion of ammonia. This substance is
probably urea or amino acids. The production of urea in
fish has been well established, however, all experimental
evidence for the occurrence of the ornithine cycle in tele-
osts is negative (Brown and Cohen, 1960). The levels of
carbamyl phosphate synthetase and ornithine transcarbamylase
have been below the level of detection in all teleosts
tested, including Salmo trutta, the brown trout. During
metamorphosis of a tadpole, enzyme induction results in
ornithine cycle activity and subsequently increased pro-
duction of urea. It is conceivable that fish with impaired
excretion of ammonia could somehow induce production of the

ornithine cycle enzymes. A simpler way to increase urea

I

40

production in the rainbow trout, however, would probably be
to increase arginase activity and production of arginine.
Thus, without the complete ornithine cycle which feeds free
ammonia into urea production, ammonia could first be incor-
porated into glutamate and then via transamination become
part of arginine which is subsequently cleaved by arginase
to yield urea and ornithine.

Before these hypotheses are tested, preliminary ex-
periments should be conducted to show that increased pro-

duction and excretion of urea does occur and not merely in-

creased excretion of amino acids. Since increased production

of amino acids is one of the detoxification mechanisms for
ammonia found in mammals, it is quite possible that in fish
with high blood ammonia values, production and excretion of

amino acids is increased.

SUMMARY AND CONCLUSIONS

Rainbow trout were placed in solutions of 0, l, 3, 5,
and 8 pg/ml of ammonia; the respective blood ammonia
concentrations for these fish were approximately 38, 42,
51, 59, and 7l_pg ammonia/ml. Thus, there is a corre-
lation between ambient ammonia and blood ammonia concen—
trations. (These same data may be expressed in terms of
unionized ammonia. :Fish exposed to a level of unionized
ammonia between 0 and l Pg/ml had blood ammonia values
between g and Q )Ig/mlb

In each case the blood level of ammonia was higher than
the water level of ammonia.

Ammonia seems to be excreted via passive diffusion
across the gill surface.

The rate of this diffusion appears to depend on the gra-
dient between blood ammonia and water ammonia. When
this gradient was reduced, excretion of ammonia was re-
tarded. Ammonia excretion dropped from 130 pg/g body
wt./day for fish in 0 pg/ml ammonia to 50 Pg/g body wt./
day for fish in 8 pg/ml ammonia.

The increased blood ammonia is probably due to accumu-

lation of endogenous ammonia rather than to a pronounced

41

42

influx of ambient ammonia, although measurements of the
latter were not made.

The percent of total nitrogen excreted as ammonia de—
creased from 51.7% for fish exposed to O‘pg/ml ammonia
to about 30% for fish exposed to Bing/ml ammonia. This
suggests that increased excretion of other nitrogenous
compounds occurred when ammonia excretion was inhibited

by reduction of the ammonia gradient.

APPENDIX

44
Comment on the Blood Ammonia Analysis

Although contamination of the whole blood samples
may have resulted from deamination of blood proteins, TCA
(trichloroacetic acid) was not used to precipitate blood
proteins as suggested by Nathan and Rodkey (who used a
slightly different assay technique) because this compound

interfered with the Conway assay for blood ammonia.

‘With TCA. When TCA was used the 3 ml blood samples
were added to 1 m1. of chilled 20% TCA and centrifuged for
ten minutes. One half ml. of the supernatant was then intro-
duced to the Conway unit for ammonia analysis. The TCA
caused an increased diffusion time, apparently, and 4 hours
had to be allotted before reproducible results could be ob-
tained with standards. This meant that at least 4% hours

passed before the sample could be analyzed.

Without TCA. Better results were obtained by placing

 

the whole blood sample directly into the Conway unit im-
mediately after withdrawal of the blood from the fish. By
having the unit prepared in advance the total time between
removal of the blood from the animal and reading of the
sample was reduced to about 25 minutes. This analysis was
also conducted at 10-12°C to prevent increased deamination
due to a 010 effect. Accuracy was improved also because the

samples were not diluted. Under these conditions, without

45

TCA, lower blood ammonia values were obtained than before
when TCA was used.

All the data presented here were obtained without
use of TCA. Some deamination may have occurred, but this
probably does not effect the significance of the relative
changes in blood ammonia values observed for fish exposed to

various concentrations Of ammonia.

BIBLIOGRAPHY

Alabaster, J. S., D. W. M. Herbert and J. Hemens. (1957)
The survival of rainbow trout (Salmo qairdenrii
Richardson) and perch (Perca fluvatilis L.) at vari—
ous concentrations of dissolved oxygen and carbon
dioxide. Ann. Appl. Biol. 45:1, 177-188.

Anderson, D. P., C. w. Beard, and R. P. Hanson. (1964) The
adverse effects of ammonia on chickens including re—
sistance to infection with Newcastle disease virus.
Avian Diseases, 3:

Baldwin, E. (1952) Dynamic Aspects of Biochemistry. Cam-
bridge University Press.

Baldwin, E. (1964) An Introduction to Comparative Bio-
chemistry. Cambridge University Press.

Barnes, R. H., B. A. Labadan, B. Siyamoglu and R. B. Brad-
field. (1964) Effects of exercise and administration
of aspartic acid on blood ammonia in the rat. Amer.
J. Physiol. 207:1242-1246.

Brockway, D. R. (1950) Metabolic products and their ef-
fects. New York State Conservation Department. 12,
127-129.

Brown, G. w. and P. P. Cohen. (1959) Comparative biochemis—
try of urea synthesis I. Methods for the quanti-
tative assay of urea cycle enzymes in liver. J.
Biol. Chem. 235:1760—1774.

Brown, G. w. and P. P. Cohen. (1960) Comparative biochemise
try of urea synthesis 3. Activities of urea-cycle in
various higher and lower vertebrates. Biochem. J.
75:82.

Brown, G. w. Jr., W. R. Brown and P. P. Cohen. (1959) Com-
parative biochemistry of urea synthesis II. Levels
of urea cycle enzymes in metamorphosing Rana cates-
beiana tadpoles. J. Biol. Chem. 234:1775—1780.

Buniatian, H. CH. and M. A. Dautian. (1966) Urea synthesis
in Brain. J. Neurochem. 13:743-753.

46

47

Burrows, R. E. (1964) Effects of accumulated excretory
products on hatchery-reared salmonids. U.S. Dept.
of the Interior. Fish and Wildlife Services.
Bureau of Sport Fisheries and Wildlife. Research
Report 66, 1-11. ‘

Cohen, P. P. and H. J. Sallach. (1961) Nitrogen metabolism
of amino acids. In Metabolic Pathways. 2:1-78.

Conway, E. J. and A. Byrne. (1933) KXI. An absorption ap-
paratus for the micro-determination of certain vola-
tile substances. I. _The micro-determination of am-
monia. Biochem. J. 27:419-429.

Conway, E. J. (1963) Microdiffusion Analysis and Volumetric
Error. Chemical Publishing Co.

Denis, W. (1913-14) Metabolism studies on cold-blooded
animals. II. Blood and urine of fish. J. Biol.
Chem., 16, 389.

Doudoroff, P. and M. Katz. (1950) Critical review of litera-
ture on the toxicity of industrial wastes and their
components to fish. I. Alkalies, acids and in-
organic gases. Sewage and Industrial wastes, 22:11,
1432-1459.

Downing, K. M. and J. C. Merkens. (1955) The influence of
dissolved-oxygen concentration on the toxicity of un-
ionized ammonia to rainbow trout (gglmoqgairdnerii
Richardson). Ann. Appl. Biol. 43:2,243-246.

Downing, K. M. and J. C. Merkens. (1957) The influence of
temperature on the survival of several species of
fish in low tension of dissolved oxygen. Ann. Appl.
Biol. 45:2, 261-267.

Elliott, K. A. C., I. H. Page and J. H. Quastel, eds. (1962)
Neurochemistry--The Chemistry of Brain and Nerve.
Charles C. Thomas Publisher, Illinois.

Ellis, M. M. (1937) Detection and measurement of stream
pollution. Bull. U.S. Bur. Fishl, 48: 365-437.

Fishbein, W. N., P. P. Carbone and H. D. Hochstein. (1966)
Urease inhibition by acetohydroxamate in ammonia in-
toxication. Gastronenterology, 50:907-908.

Feinberg, H. and Sister Mary Alma. (1961) Ammonia production
in the isolated working rabbit heart. Amer. J.
Physiol., 200:238—242.

48

Fleshler, B. and G. J. Gabuzda. (1965) Effect of ammonium
levels and acidity of gastric juice. Gut 6:349-357.

Fromm, P. O. (1963) Studies on renal and extra-renal ex-
cretion in a fresh water teleost, Salmo gairdnerii.
Comp. Biochem. Physiol., 10, 121-128.

Gebhards, Stacy V. (1965) Transport of juvenile trout in
sealed containers. The Progressive Fish Culturist,

Goldstein, L. (1966) Relation of glutamate to ammonia pro-
duction in the rat kidney. Amer. J. Physiol., 210:
661-666.

Goldstein, L. and R. P. Forster. (1961). Source of ammonia
excreted by the gills of the marine teleost Myoxo—
cephalus scorpius. Amer. J. Physiol., 200:1116-1118.

Goldstein, L. and R. P. Forster. (1962) Further studies on
the source of ammonia excreted by the gills of marine
fishes. Bull. Mt. Desert Island Biol. Lab., 5:6.
(Issued 1965)

Goldstein, L., R. P. ForSter and G. M. Fanelli, Jr. (1964)
Gill blood flow and ammonia excretion in the marine
teleost, Myoxocephalus scorpius. Comp. Biochem.
Physiol., 12, 489-499.

Greenstein, J. P., M. Wintz, P. Gullino, S. Birnbaum and C.
Otey. (1956) Studies on the metabolism of amino
acids and related compounds in vivo. III. Pre-
vention of ammonia toxicity by arginine and related
compounds. Arch. Biochem. Biophys., 64:352-354.

Grindley, J. (1946) Toxicity to rainbow trout and minnows
of some substances known to be present in waste
water discharged to rivers. Ann._App1. Biol., 33:1,
103-111.

Henderson, Y. and H. w. Haggard. (1943) NOxious gases and
their principles of action. Reinhold Publishing Co.

Hoffert, J. R. and P. O. Fromm. (1966) Effect of carbonic
anhydrase inhibition on aqueous humor and blood bi-
carbonate ion in the teleost (Salvelinus namaycush).
Comp. Biochem. Physiol., 18:333-340.

Klein, L. (1959) River pollution I. Chemical analysis.
London Butterworths. 113-114.

49

Lloyd, R. (1961) Effect of dissolved oxygen concentrations
on the toxicity of several poisons to rainbow trout
(Salmo gairdneriiDRichardson). J. Exp. Biol., 38,
447—455.

Lloyd, R. (1961b) The toxicity of ammonia to rainbow trout
(Salmo gairdnerii Richardson). The Water & Waste
Treatment Journal, 8, 287-289.

Lloyd, R. and D. W. M. Herbert. (1960) The influence of
carbon dioxide on the toxicity of unionized ammonia
to rainbow trout (Salmo gairdnerii Richardson). Ann.
Appl. Biol., 48:2, 399-404.

Lorente de No', R., F. Vidal and L. M. H. Larramendi. (1967)
Restoration of sodium-deficient frog nerve fibers by
onium ions. Nature, 179:737-378.

Maetz, J. and F. G. Romeu. (1964) The mechanism of sodium
chloride uptake by the gills of a fresh-w ter fish,
Carassius auratus. II. Evidence for NH4 /Na+ and
HCO3 7C1 exchanges. J. Gen. Physiol., 47:1209-1227.

Makarewicz, w. and M. Zydowo. (1962) Comparative studies on
some ammonia-producing enzymes in the excretory
organs of vertebrates. Comp. Biochem. Physiol., 6:
269-275. -

McBean, R. L., M. J. Neppel and L. Goldstein. (1966) Gluta-
mate dehydrogenase and ammonia production in the eel
(Anguilla, rostrata). Comp. Biochem. Physiol., 18:
909-920. ‘

McCay, C. M. and H. M. Vars. (1930) XII Studies upon fish
blood and its relation to water pollution. .in A
Biological Survey of the St. Lawrence Watershed,
Suppl. to Twentieth Ann. Rep., New York State Cons.
Dept., 230-231.

McFarland, w. (1960) The use of anesthetics for the
handling and the transport of fishes. Calf. Fish
and Game, 46:4, 407-431.

McKhann, G. M. and D. B. Tower. (1961) Ammonia toxicity
and cerebral oxidative metabolism. Am. J. Physiol.,
200:420-424.

Merkens, J. C. and K. M Downing. (1957) The effect of
tension of dissolved oxygen on the toxicity of union-
ized ammonia to several species of fish. Ann. Appl.
Biol., 45:3, 521-527.

50

Moore, E. w., G. W. Strohmeyer, and T. C. Chalmers. (1963)
Distribution of ammonia across blood-cerebrospinal
fluid barrier in patients with hepatic failure.
Amer. J. Med., 35:350.

Nathan, D. G. and R. L. Rodkey. (1957) A colorimetric pro-
cedure for the determination of blood ammonia. J.
Lab. Clin. Med., 49:779-785.

Navazio, F., T. Gerritsen and G. J. Wright. (1961) Re-
lationship of ammonia intoxication to convulsions and
coma in rats. J. Neurochem., 8:146-151.

Needham, J. and D. E. Green, eds. (1937) Perspectives in
Biochemistry. Cambridge University Press.

Obrink, K. J. (1955) Modified Conway unit for microdif-
fusion analysis. Biochem. J., 59:134.

Oser, B. L., ed. (1965) HaWk's Physiological Chemistry.
Fourteenth Edition. McGraw-Hill.

Pequin, L. and A. Serfaty. (1963) L'excretion ammoniacale
chez un teleosteen dulcicole: Cyprinus carpio. L.
Comp. Biochem. Physiol., 10:315-324.

Pequin, L. and A. Serfaty. (1966) Acide glutamique et ex-
cretion azotee chez la carpe commune, Cyprinus carpio.
’L. Comp. Biochem. Physiol., 18:141-149.

Pilkington, L., J. Welch and R. F. Pitts. (1965) Relation-
ship of pNH3of tubular cells to renal production of
ammonia. Amer. J. Physiol., 208:1100.

Pitts, R. F. (1963) Physiology of the Kidney and Body
Fluids. Year Book Medical Publishers Incorporated,
Chicago.

Richterich, R. W. and L. Goldstein. (1958) Distribution of
glutamine metabolizing enzymes and production of
urinary ammonia in the mammalian kidney. Amer. J.
Physiol., 195:316-320.

Rdbinson, R. R. and E. E. Owen. (1965) Intrarenal distri-
bution of ammonia during diuresis and anti-diuresis.
Amer. J. Physiol., 208:1129-1134.

Rosado, A., G. Flores, J. Mora and G. Soberon. (1962)
Distribution of an ammonia load in the normal rat.
Amer. J. Physiol., 203:37-42.

51

Ryberg, C. (1948) Acta Physiol. Scand., 15:161.

Schenker, S. and J. H. Mendelson. (1964) Cerebral adenosine
triphosphate in rats with ammonia-induced coma. Amer.
J. Physiol., 206:1173-1176. -

Schwartz, A. E., W. Lawrence, Jr., and K. E. ROberts. (1958)
Elevation of peripheral blood ammonia following
muscular exercise. Soc. Exptl. Biol. and Med., 98:
548-550.

Smith, Homer w. (1929) The excretion of ammonia and urea
by the gills of fish. J. Biol. Chem., 81, 727-742.

Sporn, M. B., W. Dingman, A. Defalco and R. K. Davies. (1959)
The synthesis of urea in the living rat brain. J.
Neurochem., 5:62-67.

Sullivan, L. P. and M. McVaugh. (1963) Blood and urine pH
and NH4+ excretion. Amer. J. Physiol., 204:1083.

Summerfelt, R. C. (1967) Changes in certain blood com-
ponents of goldfish maintained in sealed containers.
Progressive Fish Culturist., 29:1, 3-12.

Tasaki, I., I. Singer and A. Watanabe. (1965) Excitation
of internally perfused squid giant axons in socium—
free media. Proc. Nat. Acad. Sci., 54:763-769.

Tasaki, I., I. Singer and A. Watanabe. (1966) Excitation
of squid giant axons in sodium-free external media.
Am. J. Physiol., 211:746-754.

Takahashi, R., T. Nasu, T. Tamura and T. Kariya. (1961) Re-
lationship of ammonia and acetylcholine levels to
brain excitability. J. Neurochem., 7:103-112.

Tashiro, S. (1922) Studies on alkaligenesis in tissues. I.
Ammonia production in the nerve fiber during exci-
tation. Amer. J. Physiol., 60:519-543.

Thornburn, C. C. anth. J. Matty. (1963) The effect of thy-
roxine on some aspects of nitrogen metabolism in the
goldfish (Carassius auratus) and the trout (salmo
trutta). ”Comp. Biochem. Physiol., 8:1-12.

Van Slyke, D. D. (1943) Glutamine as source material of
urinary ammOnia. J. Biol. Chem., 150:481.

52

warren, K. S. and w. L. Newton. (1959) Portal and peri-
pheral blood ammonia concentrations in germ-free and
conventional guinea pigs. Amer. J. Physiol., 197:
717-720.

warren, K. S. and S. Schenker. (1960) Hypoxia and ammonia
toxicity. Amer. J. Physiol., 199:1105.

warren, K. S. and S. Schenker. (1962) Differential effect
of fixed acid and carbon dioxide on ammonia toxicity.
Amer. J. Physiol., 203:5, 903-906.

Weil—Malherbe, H. (1962) Ammonia metabolism in the brain.
In Neurochemistry. Elliot, Page and Quastel, eds.
321—329.

White, A., P. Handler and E. L. Smith. (1964) Principles
of Biochemistry. McGraw-Hill, New York.

woOd, J. D. (1958) Nitrogen excretion in some marine tele-
osts. Can. J. Biochem. Physiol., 36:1237-1242.

wu, C. (1963) Glutamine synthetase--I. A comparative study
of its distribution in animals and its inhibition of
D L-allo-é-hydroxylysine. Comp. Biochem. Physiol.,
8:335—351.

wuhrmann, Von K. and H. WORer. (1948) Beitrage zur toxi-
cologie der fische. II. Expermentelle untersuchungen
uber die ammoniak-und blausaurevergiftung. Schwei-
zerische Zeitschrift fur Hydrologie., 11, 210-244.

Wuhrmann, Von K. and H. woker. (1953) Beitrage zur toxi-
kologie der fische. VIII. Uber die giftwirkungen
von ammoniak—und temperatur auf fische. Schweitzer-
ische Zeitschrift fur Hydrologie., 15, 235-260.