ENZYMIC DESULPHATION OF PORPHYRAN
Vol. 80
453
have been hampered by the fact that the loss of phosphate gel followed by elution with sodium
activity in the presence of EDTA rapidly becomes acetate.
irreversible on standing. It is inadvisable to draw
2. A turbidimetric assay method has been
further conclusions on the basis of the available developed.
evidence, since the enzyme is still in a highly
3. The effect on the enzyme of pH and of various
impure state. It is possible that the observed in- activators and inhibitors has been studied. It is
hibition by cyanide (Table 2) is due to cyanide dependent on the presence of a bi- or ter-valent
binding of the essential cation, although other cation, which is not Mg2+, and is markedly actimechanisms are possible (Dixon & Webb, 1958).
vated by borate.
Since porphyran is itself a polyelectrolyte, it is
I thank Professor Stanley Peat, F.R.S., and Dr J. R.
conceivable that certain of the activators influence
for their interest and encouragement, and D.S.I.R.
the reaction by interaction with the substrate Turvey
for the award of a Fellowship.
rather than the enzyme. It is known that salts
present in solution with polysaccharide polyanions
can alter the configuration of the polymer, and it is
REFERENCES
possible that some of the cation activators (Table 2)
operate by increasing the time spent by the sub- Dixon, M. & Webb, E. C. (1958). Enzymes, p. 373. London:
Longmans, Green and Co. Ltd.
strate in a configuration favourable for reaction.
The observed activation by borate might be in- Dodgson, K. S. (1961). Biochem. J. 78, 312.
K. S. & Spencer, B. (1953). Biochem. J. 55,436.
duced by an analogous mechanism, the borate Dodgson,
Keilin, D. & Hartree, E. F. (1938). Proc. Roy. Soc. B, i24,
forming a complex with the hydroxyl groups of the
397.
polymer, thus altering the molecular charge and Peat, S. & Rees, D. A. (1961). Biochem. J. 79, 7.
configuration.
Peat, S., Turvey, J. R. & Rees, D. A. (1961). J. chem. Soc.
SUMMARY
1. The enzyme responsible for the desulphation
of porphyran by extracts of Porphyra umbilicali8
has been purified 22-fold by adsorption on calcium
p. 1590.
Rees, D. A. (1961). Biochem. J. 78, 25P.
Snell, F. D. & Snell, C. T. (1948). Colorimetric Methods of
Analysis, 3rd ed., vol. 2, p. 769. New York: Van
Nostrand Co. Inc.
Biochem. J. (1961) 80, 453
Loss of Liver Glycogen after Administration of Protein or Amino Acids
BY H. N. MUNRO, CATHERINE M. CLARK AND G. A. J. GOODLAD
Department of Biochemistry, University of Glasgow, and Department of Biochemistry,
St Salvator's College, University of St Andrews
(Received 27 January 1961)
In a previous report (Clark, Goodlad, Chisholm
& Munro, 1960), it was shown that the feeding of
protein can cause rapid changes in the amount of
adenosine triphosphate in the liver. Rats were
given a carbohydrate-rich meal in order to produce
a high concentration of adenosine triphosphate in
the liver over a period of several hours. When
casein was then fed to such animals, the adenosine
triphosphate concentration fell rapidly. Further
investigations showed that this fall was accompanied by a considerable decrease in the glycogen
content of the liver.
In the present paper, we have explored the
changes in the glycogen content of the liver caused
by protein administration. It has been shown that
there is also a rapid disappearance of glycogen
from the liver after feeding several amino acids
individually. Along with the decrease in glycogen
content, there is a fall in the total carbohydrate
content of the liver and a decrease in the bloodsugar concentration, but no change in muscleglycogen concentration. Loss of liver glycogen
occurs after administration of protein or amino
acids to adrenodemedullated rats or to alloxandiabetic rats. The nature of the mechanism underlying this action of dietary protein is discussed.
EXPERIMENTAL
Animals. Each experiment was carried out on a group
of female albino rats of closely similar weight. In different
experiments mean weight varied from 125 to 180 g. The
animals were housed in individual cages.
Feeding procedure. On the day of killing, the rats were
fed in the morning with a carbohydrate-rich meal, which
454
4H. N. MUNRO, C. M. CLARK AND G. A. J. GOODLAD
caused deposition of glycogen in their livers; some hours
later, a meal of protein or a single amino acid was given
and the effect of this on liver-glycogen concentration was
measured. In order to train the rats to consume these
meals promptly, they were fed for about a week beforehand on a synthetic diet given as two meals (Goodlad &
Munro, 1959). At 9 a.m. they received 5 g. of carbohydrate
together with vitamins and minerals, and at 5 p.m. they
ate 4-2 g. of a protein-carbohydrate-fat mixture. On the
final day, they received the carbohydrate-rich meal in the
morning as usual, but the evening meal at 5 p.m. was
replaced by the test protein or amino acid fed alone, and
the animals were killed at various times up to 2 hr. later.
Control rats, which received no food at 5 p.m., were killed
at the same time-intervals and the glycogen content of the
livers of fed and control groups were compared.
The casein used as the test protein was commercially
purchased (unextracted grade, Glaxo Ltd., Greenford,
Middlesex). The amino acids fed were obtained from
British Drug Houses Ltd., Poole, Dorset. Glutamic acid
was neutralized with NaHCO3 before it was given.
Analytical procedure8. When only the glycogen concentration in liver was studied, the rats were killed by a blow
on the head and the livers were excised immediately. When
blood-sugar and muscle-glycogen concentrations were also
measured, the animals were first anaesthetized with
Nembutal. Blood was then withdrawn by syringe from the
inferior vena cava and specimens of muscle were obtained
from the fronts of both thighs.
The samples of liver and muscle (about 2-5 g.) were
rapidly weighed and transferred to 6 ml. of hot 30 % (w/v)
KOH1. After heating in a water bath at 1000 for 15 min.,
glycogen was precipitated by addition of 1-2 vol. of
absolute ethanol (Good, Kramer & Somogyi, 1933). After
standing overnight at room temperature, the precipitated
glycogen was separated by centrifuging and dissolved in
water. The amount of glycogen was then estimated by the
anthrone procedure as described by Trevelyan & Harrison
(1952), and the results were expressed as g. of glycogen/
100 g. wet weight of liver.
The rapidity of transfer of the liver and muscle specimens to KOH was found to influence their glycogen content. Removal of both tissues, weighing and immersion in
hot KOH could be accomplished within 1 min. If the liver
and muscle specimens were kept for a further 4 min. before
being plunged into KOH, their glycogen contents were
11 and 19% lower respectively (mean of 3 experiments).
From these findings it was concluded that important losses
of glycogen were unlikely to occur during the minute
needed for rapid removal of liver and muscle specimens.
In one experiment, the total carbohydrate content of the
liver was estimated in place of glycogen. Specimens of
liver (4 g.) were dissolved in 12 ml. of hot 30% KOH and
made up to 50 ml. with water. Estimations of total carbohydrate were carried out by the anthrone procedure; the
presence of protein in the dissolved liver preparation does
not seriously affect the anthrone determination (Seifter,
Dayton, Novic & Muntwyler, 1950).
The reducing-sugar content of the blood specimens taken
from the inferior vena cava was estimated by the Hagedorn
& Jensen (1923) procedure and in a few instanoes by
glucose oxidase (Huggett & Nixon, 1957), with reagents
supplied by C. F. Boehringer und Soehne GmbH., Mannheim, Germany.
1961
Adrenodemedullation. This was carried out by the technique of Evans (1936). The cortex was allowed to regenerate
for 1-2 months before commencing the dietary experiment.
The completeness of removal of the medulla was confirmed
by histological examination at the time of killing. Animals
not completely demedullated were rejected.
Alloxan-diabetic rat8. Diabetes were produced by subcutaneous injection of 200 mg. of alloxan/kg. body wt. The
alloxan was dissolved in citrate-phosphate buffer, pH 4-0,
in order to stabilize it (Klebanoff & Greenbaum, 1954). The
occurrence of diabetes was recognized by the appearance of
glucose in the urine obtained 3 days after injection, and
was confirmed by measurements of blood-sugar concentration at the time of killing, which was generally the tenth
day after injection with alloxan; only animals with bloodsugar concentrations above 200 mg./100 ml. were accepted
as diabetic.
RESULTS
A number of workers (e.g. Barnes, 1953;
Andrews, 1960) have commented on the difficulty
of obtaining reproducible concentrations of glycogen in the livers of mice and rats. In the present
series of experiments this variability was reduced
by using rats that had been trained to consume
food promptly. On the day of the experiment, they
were fed a meal of carbohydrate which was eaten
immediately and they were killed 9-10 hr. later.
When no intervening food was given, this resulted
in a liver-glycogen concentration of 2-3 % at the
time of death.
Changes in liver-glycogen concentration produced
by feeding different amounts of casein. When rats
prepared in this way are fed casein 2 hr. before
being killed, there is a sharp decrease in the amount
of glycogen found in the liver at death (Clark et al.
1960). Table 1 shows that the extent of this effect
is related to the amount of casein fed. When 2-5 g.
of casein was given, the glycogen concentration
fell to one-tenth of that in the livers of the control
Table 1. Changes in liver-glycogen concentration
produced by feeding different amounts of ca8ein to
rat8
A standard carbohydrate-rich meal was fed at 9 a.m.
At 5 p.m. the rats received the amounts of casein indicated
below, and were killed 2 hr. later. Each entry is the mean
(± s.E.) of observations on three animals.
Mean
Liver-glycogen
Amount of
concn.
casein fed
body wt.
(g./100 g. of liver)
(g.)
(g.)
2-36±0-42
144
Nil
1.15±0.11*
147
0-5
0.84±0-31*
142
1-0
0-20±0-03t
142
2-5
* Value differs significantly from the control value
(P < 0-05 by t test).
t Value differs significantly from the control value
(P < 0-01 by t test).
Vol. 80
DIETARY PROTEIN AND LIVER GLYCOGEN CONTENT4455
Table 2. Effect of feeding single amino acids or olive oil on liver-glycogen concentration
The standard carbohydrate-rich meal was fed at 9 a.m. At 5 p.m. the rats were given 1 g. of the amino acid or
2 ml. of olive oil and were killed 2 hr. later. Control groups received no food at 5 p.m. Each entry is the mean
glycogen concentration (±s.xi.) in the liver. In one experiment with glycine, total carbohydrate concentration
was measured in place of glycogen. The data from the fed and control groups were compared statistically by the
t test. Probabilities less than 0-05 were taken to indicate a significant effect of the substance fed.
Probability
Liver glycogen
level for
(g./100 g. of liver)
No. of
difference
rats-nree
between groups
Control group
Fed group
Substance fed
gIroup
< 0-01
0-06+0-01
4
2-06±0-38
Glycine
< 0-01
*2-01 ± 037
*0.37+0.09
5
Glycine
<0-02
2-29+0-38
0-72+0-22
4
DL-Alanine
< 0-05
1-04+0-21
2-25±0-36
4
L-Glutamic acid
> 0-05
2-34+0-18
3
2-36±0-42
Olive oil
* Total carbohydrate measured in place of glycogen.
Table 3. Effect of casein or glycine administration
on glycogen concentration in the livers of adrenodemedullated rats
The rats were fed the standard carbohydrate-rich meal
at 9 a.m. and the casein or glycine at 5 p.m. and were killed
2 hr. later. Control animals received no food at 5 p.m.
Each entry is the mean (±s.z.) of observations on three
rats (Expt. 1) or five rats (Expt. 2).
Liver-glycogen
conen.
Substance
no.
administered
(g./100 g. of liver)
1
1-85+0-16
None
1.15+0.17*
Casein (2-5 g.)
1-32+0-35
2
None
0-07+0Olt
Glycine (1 g.)
* Significantly different from the control value (P < 0 05
by the t test).
t Significantly different from the control value (P < 0 01
by the t test).
Expt.
series. When smaller amounts of casein were
administered, the effect was correspondingly less.
The occurrence of a dose-response relationship of
this magnitude is rather surprising, since most of
the casein was still in the stomach when the
animals were killed 2 hr. after feeding.
Effect of feeding amino acids and fat. The action
of casein could be reproduced by feeding individual
amino acids, but not by feeding fat (Table 2). Only
three amino acids, glycine, alanine and glutamic
acid, proved palatable enough to be consumed
quickly when they were fed individually to the rats
in amounts of 1 g. Each caused a significant decrease in glycogen concentration, glycine being the
most effective. The feeding of olive oil produced no
alteration in liver-glycogen concentration.
In one of the experiments recorded in Table 2, the
total carbohydrate concentration in the liver was
measured in place of glycogen. After feeding
glycine, there was a large decrease in total carbohydrate concentration, comparable in magnitude
with the change observed in liver-glycogen content.
This is consonant with the observation of Seifter
et al. (1950) that total carbohydrate concentration
in the liver is essentially the same as glycogen concentration under a variety of circumstances.
Action of casein and glycine on the liver glycogen of
adrenodemedullated rats. Several investigators
have shown that injection of glycine and other
amino acids into rabbits produces a hyperglycaemia that can be abolished by adrenalectomy or
by ergotamine administration (Pollak, 1922; Nord,
1926; Lundsgaard, 1930; Basiliou & Zell, 1931).
These authors concluded that the injected amino
acids caused release of adrenaline and thus mobilized liver glycogen. A similar mechanism could be
responsible for the loss of liver glycogen observed
by us after feeding protein or individual amino
acids. An experiment was therefore carried out in
which protein or an amino acid was fed to adrenodemedullated rats. Table 3 shows that both casein
and glycine caused significant decreases in liverglycogen concentration in these animals. The
picture in the adrenodemedullated rat is thus
similar to that observed in animals with intact
adrenal glands.
Effect of casein and glycine administration on
muscle-glycogen and blood-sugar concentration. In
order to obtain more information about changes in
carbohydrate metabolism after administration of
protein or amino acids, blood-sugar and muscleglycogen concentrations were measured at various
times after feeding casein (Fig. 1) or glycine
(Table 4). Within an hour of feeding casein, the
concentration of glycogen in the liver had already
fallen considerably (Fig. 1). This coincided with
a significant decrease in blood-sugar concentration,
which persisted throughout the 2 hr. period after
feeding. On the other hand, glycogen concentration in muscle did not undergo any measurable
alterations. A similar picture was obtained in
animals fed with glycine 2 hr. before being killed
H. N. MUNRO, C. M. CLARK AND G. A. J. GOODLAD
456
1961
(Table 4). The decrease in blood-sugar concentration after glycine administration was confirmed in
0
a separate experiment in which the more specific
2C
procedure was used in place of the
glucose-oxidase
Ph =
Hagedorn & Jensen (1923) method (Table 4).
~0
The blood-sugar concentrations of the unfed
0-1
1 0
control rats used in these experiments were appreciably higher than the concentrations normally
observed in starving rats. This is presumably due to
0
the carbohydrate meal that was fed 8-10 hr.
06
before killing in order to cause glycogen deposition
-4
in the liver. Thus the control animals shown in
C)
Fig. 1 had a mean blood-sugar concentration of
oSb---04 109 + 3 mg./100 ml., whereas a group of rats
C)
0o
killed at similar times of day but not given the
carbohydrate meal in the morning had a mean
02
blood-sugar concentration of 88 ± 4 mg./100 ml.
These animals had only traces of glycogen in their
livers.
0
Effect of glycine administration on the liver
0.ti
120
glycogen of alloxan-diabetic rats. Since a decrease in
blood-sugar concentration was observed after
0--feeding protein or amino acids, it was possible that
100
secretion of insulin might occur after giving these
@
C)
nutrients. There is some published evidence to
05)0
support this possibility. In certain human sub80
jects, administration of leucine causes a fall in
blood-sugar concentration through release of
bo
insulin (Butterfield, Whichelow, Wright & Woolf,
60
1960). Insulin so released might cause the disappearance of glycogen from the liver; Levin &
2-0
1-5
1.0
0
05
25
Weinhouse (1958) have observed that administraTime after feeding casein (hr.)
tion of insulin to rats in the post-absorptive state
Fig. 1. Rats were given a meal of carbohydrate at 9 a.m. results in a loss of glycogen from the liver.
At 5 p.m. some of the animals (0) were fed with 2-5 g. of
In order to test the possibility that insulin may
casein; control animals (0) received no food at this time. participate in the action of protein and amino acids
Rats from each group were killed thereafter at the times on carbohydrate metabolism, rats were made
indicated. Each point is the mean of observations on diabetic with alloxan and were then fed with
five animals. Statistical analysis shows that liver-glycogen
and blood-sugar concentrations were significantly lowered glycine. A control group of diabetic rats were fed
by casein administration (P < 0 01), but there were no the same carbohydrate-rich meal in the morning,
significant alterations in muscle-glycogen concentration but received no glycine at 5 p.m.; the glycogen
concentrations in the livers of these control
(P > 0-05).
30
,
.,
5)
1.
,
.
.
o
0
5()
0
-40
0
0
1-
Table 4. Glycogen content of liver and muscle and the blood-&ugar concentration after feeding glycine
The rats were given a standard carbohydrate-rich meal at 9 a.m. At 5 p.m. they were fed 1 g. of glycine and
killed 2 hr. later. Control rats received no food at 5 p.m. Each entry is the mean (± s.E.) of observations on
four rats except for the determinations of blood-sugar concentration by the glucose-oxidase procedure, which were
carried out on a separate series of animals with eight rats in the control and glycine-fed groups.
Blood sugar
(mg./100 ml. of blood)
were
GlucoseLiver
Muscle
Hagedornoxidase
Jensen
Substance
glycogen
glycogen
method
method
administered
(g./100 g. of liver) (g./100 g. of muscle)
123±5
107±3
None
2-13±0-39
0-45:0-04
0-61±0 42t
0-48±0-12*
Glycine
83±6t
99±6t
* Not significantly different from the control value (P > 0.05 by the t test).
t Significantly different from the control value (P < 0 05 by the t test).
i Significantly different from the control value (P < 0 01 by the t test).
Vol. 80
DIETARY PROTEIN AND LIVER GLYCOGEN CONTENT
Table 5. Effect of glycine administration on glycogen
concentration in the livers of alloxan-diabetic rats
The rats were fed the standard carbohydrate-rich meal
at 9 a.m. and 1 g. of glycine at 5 p.m. and were killed
2 hr. later. Control animals received no food at 5 p.m.
Each entry is the mean (±s.F.) of observations on five
rats. Glycine administration caused a significant decrease
in liver-glycogen concentration (P < 0-05 by the t test).
Blood-sugar
Liver-glycogen
conen. (mg./
conen.
Substance
administered
(g./100 g. of liver) 100 ml. of blood)
285±21
1-87i0-51
None
250±23
0-31 ±0-17
Glycine
animals (Table 5) were similar to those observed in
non-diabetic control rats fed in the same way
(Tables 1-4). This confirms previous observations
that fed diabetic rats have a normal concentration
of glycogen in their livers (Renold, Teng, Nesbett &
Hastings, 1953). When glycine was administered to
the diabetic rats, there was a decrease in liverglycogen concentration similar in magnitude to the
effect observed in non-diabetic animals, indicating
that an intact insulin-secreting mechanism is not
essential for the action of glycine. Blood-sugar
concentrations are also quoted in Table 5, to show
that the animals studied were diabetic; wide
variations between individual animals make it
impossible to decide whether glycine administration lowered the blood-sugar concentration significantly.
DISCUSSION
When rats with high concentrations of
liver
glycogen are fed with protein or with single amino
acids, there is a rapid and considerable fall in the
amount of glycogen and of total carbohydrate in
their livers (Tables 1 and 2). Previous experiments
(Clark et al. 1960) show that a decrease in adenosine
triphosphate in the liver accompanies this loss of
glycogen.
Our data appear to exclude the possibility that
administration of protein or amino acids causes
glycogenolysis as the result of adrenaline release,
since loss of liver glycogen occurs after feeding
casein or glycine to adrenodemedullated animals
(Table 3). Release of adrenaline is also incompatible with our finding (Fig. 1 and Table 4) that
the disappearance of glycogen from the liver after
protein or glycine administration is not accompanied by a rise in blood-sugar concentration or by
a fall in muscle glycogen, both characteristic
features of the action of adrenaline (Sokal &
Sarcione, 1959). Release of glucagon, though not
excluded as part of the mechanism, would appear
to be an unlikely reason for the loss of glycogen
from the liver in view of the absence of a rise in
457
blood-sugar concentration (Foa, Galansino &
Pozza, 1957).
Insulin can induce glycogenolysis in the liver
under some circumstances. Levin & Weinhouse
(1958) produced glycogen deposition in the livers of
rats by feeding them with glucose. When insulin
was then injected, the fall in blood-sugar concentration was accompanied by a rapid loss of glycogen
from the liver, but there was no alteration in
glycogen concentration in muscle. The picture we
have obtained after feeding casein (Fig. 1) or
glycine (Table 4) is identical with this. Nevertheless, secretion of insulin does not appear to be part
of the mechanism by which protein and amino
acids cause loss of liver glycogen, since this effect
still occurs in alloxan-diabetic animals (Table 5).
Whatever may initiate the mechanism of glycogen breakdown in the liver, the utilization of the
products of breakdown is of interest in view of the'
specific dynamic action of absorbed amino acids.
The amino acids found by us to cause loss of liver
glycogen are all effective in inducing a specific
dynamic action irrespective of the route of administration (Weiss & Rapport, 1924; Nord &
Deuel, 1928; Lundsgaard, 1931; Wilhelmj, Bollman
& Mann, 1931). This specific dynamic action occurs
very soon after feeding the amino acids and would
thus coincide with the loss of liver glycogen observed in our experiments. A causal relationship is
suggested by experiments in which Wilhelmj &
Mann (1930) injected glycine or alanine into dogs
that had been previously fed on different types of
diet. When the animals had been without food for
some days, injection of the amino acids caused a
specific dynamic action that was unaccompanied
by any change in R.Q. However, when the dogs had
been on a standard mixed diet and notably when
they had been receiving a diet rich in carbohydrate,
administration of the amino acids caused a rise in
R.Q., indicating preferential combustion of stored
carbohydrate during the period of specific dynamic
action.
The amount of glycogen lost from the liver of the
rat after glycine administration can be compared
quantitatively with the additional energy released
by the specific dynamic action of this amino acid.
The data in Table 4 show that, 2 hr. after feeding
1 g. of glycine, liver-glycogen concentration was
lowered by 1-5 g./100 g. of liver below that observed in the control group. Since female rats of
the size used (170 g.) have livers weighing about
5 g. (Goodlad & Munro, 1959), the total quantity of
glycogen lost from the liver thus amounts to some
75 mg., and on combustion this would yield 0- 3 kcal.
The specific dynamic action of glycine has not
been measured on rats under conditions identical
with those of our experiments, but Lewis & Luck
(1933) fed approximately 1 g. of glycine to fasting
458
H. N. MUNRO, C. M. CLARK AND G. A. J. GOODLAD
rats weighing 200 g. and observed an elevation of
energy output lasting about 2-5 hr. and amounting
to 0-32 kcal. Thus the amount of energy dissipated
by the specific dynamic action of glycine could
account for all the glycogen lost from the liver.
This explanation is compatible with the decrease
in adenosine triphosphate concentration which
occurs at the same time as the loss of liver glycogen
after casein administration (Clark et al. 1960). The
coincident decrease in blood-sugar concentration
(Fig. 1 and Table 4) is presumably caused by the
depletion of liver glycogen, since our data show
that loss of liver glycogen as the result of a prolonged lack of food is also accompanied by a fall in
blood-sugar concentration.
SUM:DARY
1. Rats were given a meal of carbohydrate
which caused deposition of glycogen in their livers,
and were then fed on casein, glycine, alanine,
glutamic acid or olive oil.
2. The casein and the amino acids each caused a
large loss of glycogen from the liver. This action
was already considerable within 1 hr. of administration. There was no change in glycogen concentration 2 hr. after feeding olive oil.
3. Coincident with the loss of liver glycogen
after protein or amino acid administration, there
was a fall in blood-sugar concentration. There was
no change in muscle-glycogen concentration.
4. The feeding of protein or amino acids to
adrenodemedullated rats or to alloxan-diabetic
rats resulted in a loss of glycogen from their livers.
It was concluded that secretion of adrenaline or of
insulin is not an essential part of the mechanism
responsible for loss of liver glycogen after protein
administration.
5. It is suggested that the loss of liver glycogen
after administration of protein or amino acids is
associated with the extra energy expended during
the specific dynamic action of the absorbed amino
acids.
It is a pleasure to acknowledge the kindness of Dr R. S.
Crossland in supplying us with adrenodemedullated rats.
Histological examination of the glands to confirm the
1961
completeness of the removal of the medulla was carried out
by Mr F. Weston. Miss M. Mathieson gave skilled assistance
with many of the glycogen and blood-sugar determinations.
The Medical Research Council generously provided a grant
to one of us (G. A. J. G.) for scientific assistance and for
materials used in the course of the investigations.
REFERENCES
Andrews, M. M. (1960). Biochem. J. 75, 298.
Barnes, J. M. (1953). Biochem. J. 54, 148.
Basiliou, B. & Zell, F. (1931). Biochem. Z. 238, 418.
Butterfield, W. J. H., Whichelow, M. J., Wright, P. H. &
Woolf, L. I. (1960). Nature, Lond., 188, 70.
Clark, C. M., Goodlad, G. A. J., Chisholm, J. & Munro, H. N.
(1960). Nature, Lond., 186, 719.
Evans, G. (1936). Amer. J. Phy8iol. 144, 297.
Foa, P. P., Galansino, G. & Pozza, G. (1957). Recent Progr.
Hormone Res. 13, 473.
Good, C. A., Kramer, H. & Somogyi, M. (1933). J. biol.
Chem. 100, 485.
Goodlad, G. A. J. & Munro, H. N. (1959). Biochem. J. 73,
343.
Hagedorn, H. C. & Jensen, B. N. (1923). Biochem. Z. 135,
46.
Huggett, A. St G. & Nixon, D. A. (1957). Lancet, ii, 368.
Klebanoff, S. J. & Greenbaum, A. L. (1954). J. Endocrin.
11, 314.
Levin, HI. W. & Weinhouse, S. (1958). J. biol. Chem. 232,
749.
Lewis, H. G. & Luck, J. M. (1933). J. biol. Chem. 103,
227.
Lundsgaard, E. (1930). Biochem. Z. 217, 125.
Lundsgaard, E. (1931). Skand. Arch. Physiol. 62, 223,
243.
Nord, F. (1926). Acta med. scand. 65, 1, 61.
Nord, F. & Deuel, H. J. (1928). J. biol. Chem. 80, 115.
Pollak, L. (1922). Biochem. Z. 127, 120.
Renold, A. E., Teng, C. T., Nesbett, R. F. & Hastings,
A. B. (1953). J. biol. Chem. 204, 533.
Seifter, S., Dayton, S., Novic, B. & Muntwyler, E. (1950).
Arch. Biochem. 25, 191.
Sokal, J. E. & Sarcione, E. J. (1959). Amer. J. Physiol. 196,
1253.
Trevelyan, W. E. & Harrison, J. S. (1952). Biochem. J. 50,
298.
Weiss, R. & Rapport, D. (1924). J. biol. Chem. 60, 513.
Wilhelmj, C. M., Bollman, J. L. & Mann, F. C. (1931).
Amer. J. Physiol. 98, 1.
Wilhelmj, C. M. & Mann, F. C. (1930). Amer. J. Physiol.
93, 69, 258.