TED, ROLE OF HORMONES IN HEPATIC CARBOHYDRATE METABOLISM
A thesis submitted for the degree of
Doctor of Philosophy
in the
University of London
by
Patricia Denise Williams (nee Whitton)
Department of Biochemistry Imperial College of Science and Technology London, S.W.7 2AZ., U.K.
September, 1975
2
abstract
Net rates of glycogen accumulation (followed in sequential
liver samples) were measured in the perfused liver of 48h- starved
rats in the presence of glucose and gluconeogenic precursors. The
role of glucose as a carbon source of glycogen was assessed and found
to be minimal, gluconeogenic precursors beingthe major substrate of
glycogenesis. The activities of glycogen synthetase and phosphorylase
were assessed; their activities were related to the rate of glycogen
synthesis, and to the nature of the circulating substrates.
The achievement of normal net rates of glycogen accumulation,
in vitro, permitted the investigation of the role of hormones in glycogen
synthesis. Hormones of the posterior-pituitary gland,at "physiological"
concentrations,were found to cause glycogenolysis, prevent glycogen
synthesis and increase gluconeogenesis in the perfused liver. Although
the full mechanism of vasopressin action was not elucidated, increased
hepatic phosphorylase activity was found.
The role of insulin in hepatic carbohydrate metabolism was
assessed. An impairment in glycogen accumulation and response of glycogen
synthetase to substrates in the perfused liver of starved diabetic rats
was observed, which was restored by insulin, or glucose and fructose
treatment in vivo. Insulin did not, however, have any effect in vitro.
Glycogen accumulation and the response of glycogen synthetase and,
phosphorylase to substrates were impaired during perfusion of livers
from starved adrenalectomised rats. Restoration of net rates of
glycogenesis were observed after treatment in vivo with hydrocortisone,
or fructose, glucose and insulin, prior to perfusion, confirming previous
suggestions that the action of glucocorticoids is mediated by insulin.
However, no direct hepatic effect of insulin was obtained.
3
In conclusion, it would appear that glucose is not the major
precursor of hepatic glycogen (perhaps even in.the fed state), and
that adrenal corticosteroids and insulin exert crucial regulatory
effects on glycogen metabolism)which are not direct actions on the
liver. There are however, direct hepatic effects of vasopressin, whose
significance is not fully clear.
ACKNOWLEDGMENTS
I wish to express my sincere thanks to Dr. D. A. Hems for
his supervision, encouragement and discussions during the tenure of
the work reported in this thesis. I would also like to thank Professor
Sir Ernst Chain, F.R.S., and Dr. Anne Beloff -Chain for their
encouragement and interest in the project. \
My thanks are due to members of the Department, past and present,
(especially those who have resided in Rooms 5021'205) for many
stimulating discussions, andhelp. In particular I would like to thank,
Dr. I. Das for advice concerning the enzyme assays, and Mr. E. Taylor
and Mr. C. Harmon for excellent technical assistance. I am indebted
to Mr. D. Green for aid in the animal work, in breeding the rats and
help in the preparation of diabetic and adrenalectomised animals. The
amino acid analyses were kindly performed by Mr. C. Dykes.
My thanks are offered to Dr. M. Forsling (Department of Physiology,
Middlesex Hospital, London) who kindly carried out the bioassays of
vasopressin.
I am also indebted to Mrs. R. Sayer for the valuable "sun" time
she forfeited in order to type this thesis which must have seemed "double
-dutch" to her.
Finally, I would like to thank the Medical Research Council, U.K.,
for financing this project.
5
TO MY MOTHER, FATHER, HUSBAND AND MAX.
6
Investigating the mechanism of action of insulin on the glycogen
synthetase system is like peeling an onion. Not only may it bring
tears to the eyes, but after each successful step, one is left with
the layer beneath.
Wolff & Wolff (1964)
CO
Page
Abstract .2
Acknowledgements
Abbreviations and enzymenomenclature•
List of tables 16
List of figures
18
Chapter 1 : Introduction
21
Chapter 2 : Animals, Materials & Methods 47
Chapter 3 : Results • 73
Chapter 4 : Discussion 180
For details of each chapter see next page.
References 213
Page CHAPTER. ONE
INTRODUCTION
1.1 Glycogen metabolism in the liver 22
1.1.1 The characteristics of glycogen 22
1.1.2 The pathways of hepatic glucose metabolism 24
1.2 Hepatic glycogen synthetase and phosphorylase 27
1.2.1 General considerations 27
1.2.2 Hepatic glycogen synthetase 29
1.2.3 Hepatic glycogen phosphorylase 31
1.3 The circulating precursors of hepatic glycogen 33
1.4 The role of hormones in hepatic glycogen
35 metabolism.
1.4.1 Insulin and hepatic glycogen
35
1.4.2 Adrenal cortical steroids and hepatic glycogen 39
1.4.3 "Glycogenolytic" hormones and hepatic glycogen 42
1.5 Scope and aims of the present study
43
1.5.1 General considerations
43
1.5.2 The use of the perfused liver for metabolic
43 studies
CHAPTER TWO
ANIMALS, MATERIALS AND METHODS
2.1 The preparation of animals 49
2.2 Sourcesof materials 54
2.3 The technique of liver perfusion 56
2.3.1 Perfusion apparatus 56
2.3.2 Perfusion medium 59 ,
.2.3.3 Surgical procedure for liver perfusion 61
2.3.4 Sample preparation 63
9
2.4 The techniques used in intact animal
Page,
experiments 64
24.1 Measurement of net glycogen accumulation in vivo 64 '
2.4.2 Measurement of the enzymes of glycogen metabolism in response to hormones 65
2.5 Analytical methods ' 66
2.5.1 Glucose and glucose polymer determination 66
2.5.2 Carbohydrate metabolite determination 67
2.5.3 Determination of nitrogenous compounds 69
2.5.4 Assay for the enzymes glycogen synthetase and phosphorylase
CHAPTER THREE
RESULTS
3.1 The characteristics and control of hepatic looens -EIEMthesisilhe 8h- starvedrat
74
3.1.1 The validation of sequential sampling in the perfusion 74
3.1'.2 The role of glucose and gluconeogenic precursors in glycogen deposition in the perfused liver 77
3.1.3 The role of insulin and fatty acids in hepatic glycogen metabolism 188
3.1.4 Characteristics of glycogen synthetase and phosphorylase in the liver 90
3.1.5 Control of glycogen synthesis in the perfused liver of normal starved rats 97
3.1.6 Hepatic glycogen accumulation in the intact rat 99
3.2 Hepatic carbohydrate and fat metabolism in the "fed" rat 103
3.2.1 Glucose metabolism in the perfused liver of fed rats 103
3.2.2 Glycogen synthesis in the perfused liver of overnight-starved rats 105
10
Page:
3.2.3 Pattyacidsynthesis in the perfused liver of overnight-starved rats 107
.3.3 The actions of the hormones of the posterior-pituitary gland on hepatic glycogen metabolism 109
3.3.1 Stimulation of hepatic glycogen breakdown by (8 -arginine) -vasopressin and oxytocin 109
3.3.2 The role of vasopressin in hepatic glucose metabolism 115
3.3.3 The stimulation of hepatic gluconeogenesis by (8 -arginine) -vasopressin 117
3.3.4 The action of vasopressin and oxytocin on glycogen synthesis in the perfused liver and the intact rat 120
3.3.5 The effects of vasopressin and other glycogenolytic hormones on hepatic glycogen synthetase and phosphorylase in vivo 122
3.4 Hepatic glycogen metabolism in the starved streptozotocin-diabetic rat 128
3.4.1 Glycogen accumulation in the perfused liver from diabetic rats 128
3.4.2 Glycogen synthetase and phosphorylase activities in vivo and in the perfused liver of diabetic rats 133
3.4.3 Influence of glucose and fructose on the, activity of glycogen synthetase and phosphorylase in vivo 143
3.4.4 Hepatic glycogen accumulation in the intact diabetic rat 145
3.5 Hepatic carbohydrate metabolism in the starved adrenalectomised.rat 148
3.5.1 Glycogen accumulation In the perfused liver from adrenalectomised rats 148
.3.5.2 Glycogen synthetase and phosphorylase activities in vivo and in the perfused liver of adrenalectomised rats 153
3.6 Amino acid balance in the perfused liver 168
11
Page 3.6.1 Amino acid metabolism in fed, starved
and diabetic rat liver 168
3.6.2 Urea formation in the perfused liver 173
3.6.3 The role of the liver in amino acid utilisation 177
CHAPTER FOUR
DISCUSSION
4.1 Hepatic glycogen metabolism in the normal rat
4.1.1 The circulating precursors of hepatic glycogen 181
4.1.2 The role of glucokinase in glycogen accumulation 185
4.1.3 Control of hepatic glycogen synthesis
186
4.2 The role of insulin in hepatic carbohydrate metabolism
4.2.1 Insulin and hepatic glycogen metabolism in the normal starved rat 191
4.2.2 Hepatic glycogen accumulation in the starved diabetic rat 193
4.2.3 Properties of glycogen synthetase and phosphorylase in the perfused liver of diabetic rats 197
4.3-- The role of adrenocortical steroids in hepatic glycogen metabolism
4.3.1 Glucocorticoids and hepatic glycogen metabolism in the normal (starved) rat '201
4.3.2 Hepatic glycogen accumulation in the starved adrenalectomised rat 201
4.3.3 The characteristics of glycogen synthetase and phosphorylase in the perfused liver of adrenalectomised rats 205
4.4 The role of the hormones of the posterior- pituitary_gland in hepatic carbohydrate metabolism
12
Page
4.4.1 The posterior-pituitary gland hormones and the metabolism of liver carbohydrate 208
4.4.2 The mechanism of vasopressin action. 210
13
ABBREVIATIONS AND ENZYME NOMENCLATURE
Units and Physical Constants
Those used were as recommended in Biochem. J.
(1975) 145, 1
Chemicals
Abbreviations for amino acids are as in Biochem. J.
(1972) 126, 773
ADP .adenosine 5 -pyrophosphate
AMP adenosine 5
1 -phosphate
ATP adenosine 5' -triphosphate
Butyl PBD
cyclic AMP
5-(4-biphenyly1)-2(4-t-butylphenyl) -1 -oxa -3, 4- diazole
adenosine 3': 51 -cyclic
phosphate
cyclic GMP guanosine 3 : 5'-cyclic phosphate
EDTA ethylenediamine-tetra- acetic acid
K0104
potassium chlorate
KCN potassium cyanide
ICP potassium fluoride
K3Fe(CN)6 potassium ferricyanide
KOH potassium hydroxide
LiBr lithium bromide
MgC12 magnesium chloride
NaC1 sodium chloride
NAD nicotinamide-adenine dinucleotide
14
NADH2 nicotinamide-adenine
dinucleotide, reduced
NADP nicotinamide-adenine dinucleotide phosphate
RALPH.2 nicotinamide-adenine dinucleotide phosphate, reduced
NaF sodium fluoride
Nall2PO4
sodium dihydrogen orthophosphate
Na2BP04
disodium hydrogen orthophosphate
Pi orthophosphate
TCA trichloroacetic acid
Trig 2-amino-2-hydroxymethyl- -propane-1,3-diol
UDPG uridine diphosphoglucose
Other Abbreviations
Adx. adrenalectomised
intragastric
S.C. subcutaneous
15
Enzyme Nomenclature
The names of the enzymes are those recommended by a
I.U.P.A.C., and I.U.B., Commission, 1972. In a few instances,
trivial names are used and the Commission's recommendations are
included in brackets.
E.C. Number
Amyloglucosidase/Lysosomal tC-glucosidase 3.2.1.3. (Eko-1, 4-4-glucosidase)
Glucokinase 2.7.1.2.
Glucose oxidase 1.1.3.4.
Glucose 6-phosphatase 3.1.3.9.
Glucose 6-phosphate dehydrogenase 1.1.1.49.
Glycogen branching enzyme ; 2.4.1.18. Amylo-1, 4 .41, 6 glucosidase
Glycogen debranching enzymes ; Amylo-1, 6-glucosidase 4 oligo-1, 4 >1, 4 glucan transferase (4-oG- Glucanotransferase)
Glycogen phosphorylase
Glycogen phosphorylase kinase
Glycogen phosphorylase phosphatase
Glycogen synthetase (Glycogen synthase)
3.2.1.33. 2..4.1.25.
2.4.1.1.
2.7.1.38.
3.1.3.17.
2.4.1.11.
Glycogen synthetase kinase 2.7.1.37. (Protein kinase)
Glycogen synthetase phosphatase
Hexokinase 2.7.1.1.
Lactate dehydrogenase 1.1.1.27.
Peroxidase 1.11.1.7.
Phosphoglucomutase 2.7.5.1.
UDPG pyrophosphorylase 2.7.7.9. (Glucose 1-phosphate uridylytransferase)
16
LIST OF TABLES
Page
1. The effect of time of day on rat body weight and blood glucose concentration.
2. Seasonal variation in the growth of adrenalectomised rats.
Glycogen content and synthesis in the major lobes of perfused liver from 48h- starved rats.,
Glycogen synthesis and changes. in medium glucose during perfusion of livers from starved rats.
Calculated total synthesis of glucose during perfusion of liver from starved rats.
6. , Incorporation of 14
C from cg-14 C)-glucose (30mM) into glycogen of perfused livers from starved rats.
. 7a. Rates of net glycogen accumulation in perfusions with single substrates.
7b. Rates of net glycogen accumulation in perfusions with substrate conbinations.
8. Influence of added sodium oleate and insulin on glycogen synthesis in the perfused liver from starved rats.
The effect of different assay times on the activities of glycogen synthetase and phosphorylase.
10. The effect of variation in temperature and purification on the activities of hepatic glycogen synthetase and phosphorylase in the fed and starved rat.
11. Concentrations of enzymes and pathway intermediates during glycogen synthesis in the perfused liver of starved rats.
53
75
78
80
85
86
87
89
93
95
• i98
12. Glycogen synthesis in the perfused liver of overnight starved rats. 106
13. Effect of (8-arginine)-vasopressin on hepatic glycogen synthesis in vivo. 123
14. Effect of glycogenolytic hormones on, the activity of glycogen synthetase. 126
15. Glycogen accumulation during perfusion of livers from streptozotocin-diabetic rats. 130
16. Activities of glycogen synthetase and phosphorylase in intact starved normal and starved streptozotocin - diabetic rats. 134
17. Glycogen synthetase activity in the perfused liver of starved streptozotocin-diabetic rats. 135
18. Phosphorylase activity in the perfused liver of starved streptozotocin -diabetic rats. 137
19. Glycogen synthetase activity in the perfused liver of starved streptozotocin-diabetic rats. 141
20. Influence of glucose and fructose on glycogen synthetase and phosphorylase in intact starved normal and streptozotocin -diabetic rats. 144
21. Glycogen accumulation in intact starved normal and starved streptozotocin -diabetic rats. 146
22. Glycogen accumulation in the perfused liver from starved adrenalectomised rats. 150
23. Glycogen synthetase and phosphorylase activities in intact sham adrenalectomised and adrenalectomised rats. 155
24. Glycogen synthetase activity in the perfused liver from starved adrenalectomised rats. 156
25. Glycogen phosphorylase activity in the perfused liver from starved adrenalectomised rats. 158
26. The concentration of amino acids after perfusion in the presence of "normal" levels or no added amino acids. 171
27. Effect of added amino acids on urea production in the perfused liver. 174
28. Nitrogen balance in the perfused liver. 176
18
LIST OF FIGURES
Page
1. The pathway of liver glycogen metabolism. 25
2. Growth curve of streptozotocin and control injected rats. 49
Growth curve of adrenalectomised, sham-operated, normal on 0.9% Nacl and normal rats. 52
4. Apparatus for the perfusion of rat livers. 57
Rates of glycogen synthesis at various glucose concentrations in the perfused liver of starved rats. 79
6. Time course of glycogen synthesis in the perfused liver.. 82
7. Time course of glucose formation in the perfused liver. 83
8. Time courses of glycogen synthetase and phosphorylase 92 assays.
9. Rates of glycogen synthesis in vivo, at various blood glucose concentrations. 100
10. Time course of glycogen synthesis in vivo. 101
11. Glucose metabolism in the perfused liver of fed rats. 104
12. Influence of vasopressin on the time course of glucose output in the perfused liver of fed rats. 110
13. Dependence of the stimulation of hepatic glucose output on vasopressin concentration. 111
14. Effect of vasopressin on glycogen content of the perfused liver.
15. Influence of oxytobin on the time course of glucose output in the perfused liver of fed rats.
16. Effect of vasopressin on glucose metabolism in the perfused liver of fed rats.
17. Influence of vasopressin on gluconeogenesis or net glycogen accumulation in the perfused liver of starved rats.
-113
114
116
118
18. Effect of vasopressin on gluconeogenesis (endogenous and from added substrate)in the perfused liver of starved rats. 119
19
Page
19. Effect of (6 -lysine) -vasopressin and oxytocin on glycogen synthesis in the perfused liver of starved rats. 121
20. The time course of phosphorylase activation by glycogenolytic hormones. 124
21. Time course of net glycogen accumulation in normal and diabetic rats. 129
22. Relationship between net hepatic glycogen accumulation and glycogen synthetase and phosphorylase in diabetic rats. 139
23. Relationship between the changes in glycogen synthetase and phosphorylase during liver perfusion in diabetic rats. 142
24. Time course of restoration of net rates of glycogen deposition in the perfused liver from adrenalectomised rats. 152
25. Change in sensitivity of glycogen synthetase to substrates during net glycogen accumulation in perfused livers from adrenalectomised rats. 160
26. Response of phosphorylase activity during net rates of glycogen accumulation. 162
27. Relationship between the changes in glycogen synthetase and phosphorylase during liver perfusion in adrenalectomised rats. 164
28. Changes in combined response of both glycogen synthetase and phosphorylase during rates of net glycogen accumulation in the perfused liver from adrenalectomised rats. 166
29. Metabolism of amino acids during perfusion at high initial concentrations. 1169
30. The effect of four times "normal" concentrations of amino acids on urea production in the perfused liver. 175 •
CHAPTER ONE
INTRODUCTION
21
CHAPTER 1
INTRODUCTION
1.1 GLYCOGEN METABOLISM IN THE LIVER
1. The characteristics of glycogen
2. The pathways of hepatic glucose metabolism
1.2 HEPATIC GLYCOGEN SYNTHETASE AND GLYCOGEN PHOSPHORYLASE
1. General consi derat ions
2. Hepatic glycogen synthetase
Hepatic glycogen phosphorylase
1.3 !WE CIRCULATING PRECURSORS OF HEPATIC GLYCOGEN
1.4 THE ROLE OF HORMONES IN HEPATIC GLYCOGEN METABOLISM
1. Insulin and hepatic glycogen
2. Adrenal cortical steroids and hepatic glycogen
"Glycogenolytic" hormones and hepatic glycogen
1.5 SCOPE AND AIMS OF THE PRESENT STUDY
1. General considerations
2. The use of the perfused liver for metabolic studies
22
1.1 GLYCOGEN METABOLISM IN Thh LIVER
1.1.1 The characteristics of glycogen
Glycogen has been sometimes called animal starch. It is also
however, found in yeasts, algae and fungi and large amounts are found
in oysters and other shellfish. A similar polysaccharide has also been
found in the golden bantam sweetcorn. In higher animals, glycogen is
deposited in liver and muscle as a carbohydrate storage material,
available as an immediate source of energy.
The level of glycogen varies from tissue to tissue, the liver
being the main site of deposition where there may be up to 409/.4mol
of glycogen -glucose/g wet weight of liver, which can be rapidly depleted
by hormones or starvation. In skeletal muscle the level of glycogen is
only about 251Lmol of glyeogen-glucose/g which is not depleted on
starvation, but is decreased by exercise and hormones, especially
adrenalin. Glycogen is a glycolytic fuel in muscle, providing lactate,
but in the liver, due to, the presence of glucose 6 -phosphatase it is
also degraded to glucose. This glucose is then released into the
circulation to be utilised by other tissues especially skeletal muscle
and the brain. The above considerations show the importance of hepatic
glycogen and why the levels must be tightly controlled by hormones and
the blood glucose level.
Glycogen is a polymer in which glucose residues are cc-1, 4—
linked with0C-1, 6 branchpoints. A number of structures have been
proposed from available chemical and enzymatic data, e.g., the one
proposed by Whelan (1971).
It is widely accepted that the glycogen molecule is a compact
multi-branched structure, with an average chain length ofi2-)4glucose
residues. Due to the high degree of branching, about 10 per cent of
the glucose units are situated at non-reducing termini available for
degradation, enabling rapid mobilisation. The branching also confer
a high degree of solubility, whereas the large molecular weight exerts
a small osmotic pressure in comparison to the same amount of glucose in
a free form.
Several pioneering observations, which have been crucial in
the progress of biochemistry, may be cited for work in the glycogen
field (see Ryman & Whelan, 1971). Thus glycogen was the first polymer
to be synthesised in vitro, and the first for which it was recognised
that nucleoside diphosphate sugar and a primer were necessary for
synthesis. It is one of the few branched polysaccharides whose
molecular structure has been studied extensively and has been used as
a model for the enzymatic determinations of polymer structure..
Glycogen synthetase and glycogen phosphorylase (the enzymes involved
in the metabolism of glycogen) were the first enzymes shown to be
regulated by phosphorylation and dephosphorylation of the protein, a
phenomenon which now appears to be implicated in the control of a number
of enzymes. The adenine nucleotide, cyclic AMP, was discovered when
its role in the control of phosphorylase activity was recognised.
These and other "firsts" illustrate the contribution that the study
of glycogen has made to a number of different aspects of biochemistry.
1.1.2 The pathways of hepatic glucose metabolism
The pathway of glycogen metabolism in the liver is summarised in
Fig.1. Glycogen synthesis from glucose occurs by the successive
action of glucokinase (1), phosphoglucomutase (2), MPG -
pyrophosphorylase (3), and glycogen synthetase and branching enzyme'
(4 and 5). The phosphorylytic degradation of the polysaccharide
involves the co-ordinated operation of phosphorylase and debranching
enzyme (6 and 7); glucose 1-phosphate formed by phosphorolysis,
is converted to glucose by the successive action of phosphoglucomutase
(2) and glucose 6 -phosphates° (8). Glycogen degradation can also
occur by hydrolysis under the action of lysosomaloC-glucosidase
(Jeffrey et al., 1970; Lejeune et al., 1963). The role of this
mechanism in the control of physiological degradation of glycogen is
still uncertain.
The liver contains three enzymes capable of phosphorylating glucose,
namely hexokinase, glucokinase and glucose 6-phosphatase. Hexokinase
has a very high affinity for glucose (Km of rat liver enzyme about
3 x 10 -5M; Vinuela et al., 1963; Walker, 1963) and is thus always
saturated in vivo (glucose concentration 5 to 10mM). This enzyme
cannot therefore, be affected by variations in the blood suger level.
Glucokinase is however, an enzyme which can convert glucose into glucose
6-phosphate at a rapid rate. It displays a Km for glucose of 10-40 mM
(Vinuela et al., 1963; Walker, 1963) and this activity is thus highly
responsive to changes in glucose. Microsomal glucose 6-phosphatase,
in addition to its hydrolytic role, can catalyse phosphotransferase
reactions (see Nordlie, 1968 for review). The Km for glucose is 80 mM
and so it would appear that glucose phosphorylation by this enzyme
LUCOSE 1-P
U D PG
PP i
UD P
(GLUCOSE)
[1, 4--)1,6
(GLUCOS)n+1
GLYCOGEN
UTP
GLUCOSE 6-P
Fig. 1
(1) (2)
(3) (4) (5) (6) (7)
(8)
The pathway of liver glycogen metabolism
Glucokinase
Phosphoglucomutase
UPG — pyrophosphorylase
Glycogen synthetase
Branching enzyme
Phosphorylase
Debranching enzyme
Glucose 6—phosphatase
•
26
would not occur except at exceptionally high glucose concentrations;
this action has been invoked in diabetic animals (Friedmann et al.,
1967) which have a very low level of glucokinase (Vinuela et al.,
1963). As will be seen, the present experiments show there is no
requirement to invoke a role for this enzyme in glucose phosphorylation,
in diabetes or in any other situation.
1.2 HEPATIC GLYCOGEN SYNTHETASE& GLYCOGEN PHOSPHORYLASE
1.2.1 General considerations
Glycogen synthetase and the branching enzyme (amylo
1, 4-41, 6 glucosidase) are involved in glycogen synthesis, and
phosphorylase and the debranching enzymes (olig0-1,4-)1, 4 glucan
transferase and amylo-1, 6-glucosidase) in glycogen breakdown.
Although the branching and debranching enzymes are important in
determining the structure of glycogen they do not appear to be rate
controlling (Birch et al., 1974; Ryman & Whelan, 1971). Glycogen
ynthetase and phosphorylase are however, rate-limiting for synthesis
and degradation respectively.
Glycogen synthetase catalyses the formation of 0C -1,
4-glucosyl bonds on the outer branches of glycogen utilising UDP -
glucose as the glucosyl donor:
DTP glucose + glycogen ------41111° + glycogen
(n glucosyl units) (n + 1 glucosyl units)
Salsas and Larner (1975) have shown that muscle glycogen synthetase
is able to use free glucose as the glucosyl acceptor, although the
presence of very high glucose concentrations were necessary. It may
therefore be unlikely that this occurs in vivo.
28
The discovery of glycogen synthetase in liver by Leloir and Cardini
(1957) ended an era in which it was thought that glycogen synthesis,
as well as degradation,,was catalysed by phosphorylase. The
equilibrium of the above reaction favours the formation of glycogen
and so the reaction is irreversible in vivo. Liver glycogen
synthetase is bound to liver glycogen (Leloir & Goldemberg, 1960)
which is found associated with the smooth endoplasmic reticulum (Pbrter
& Bruni, 1959). The enzyme is located in both the microsomal pellet
and supernatant, although greater activity is found in the former
fraction (Maddaiah & Madsen, 1968). Unlike phosphorylase however,
the distribution of glycogen synthetase does not appear to be influenced
when hepatic glyCogen content is changed by diet (Maddaiah & Madsen,
1968).
Glycogen phosphorylase catalyses the transfer of glucosyl units
from the non-reducing ends of the polysaccharide to inorganic phosphate:
Glycogen + + Glucose 1-P
(n glucosyl units) . glucosyl units)
The equilibrium of the reaction is reached when the ratio of
inorganic phosphate to glucose 1-phosphate is about 3 at pH 7 (Cori
et al., 1940 & 1943). Although this reaction could therefore be
reversible the reaction is primarily degradative in vivo due to the
high intracellular concentration of inorganic phosphate and low level
of glucose 1-phosphate. Subcellular fractionation of rat liver
suggests that phosphorylase is bound to liver glycogen in the well-fed.
state and sediments with the microsomal fraction. However, when
glycogen levels are depleted the enzyme is found in the supernatant
fraction (Maddaiah & Madsen, 19663 Tata, 1964).
29
1.2.2. Hepaticfllycoen synthetase
Two forms of glycogen synthetase have been isolated from
the liver, the "a" or "I" form which is dephosphorylated and the
"b" or "D" form which is phosphorylated (Bishop & Larner, 1969;
Eiznkuri & Larner 1964; Mersmann & Segal 1967). The possibility
of a third form of glycogen synthetase in muscle at least, has
been proposed by Rosell-Perez (Hildalgo & Rosell-Perez, 1971).
The "I" and "D" (i.e., independent and dependent on glucose
6-phosphate for activity) nomenclature is derived from the muscle
enzyme but as will be seen later does not strictly apply to the
liver enzyme. The two forms of the enzyme are interconverted by
a kinase requiring Mg and ATP (D6Uulf & Hers, 1968a) which is
irhibited by high ATP and high Mg2+ (Bishop & Larner, 1967 & 1969)
and activated by cyclic AMP (Bishop & Lamer, 1969; Glinsmann &
Bern, 1969; Jefferson et al., 1968) and a phosphatase (two forms of
this enzyme have been postulated by Bishop, 1970) which is Mg2.+
dependent (Hizukuri & Lamer, 1964), activated by glucose (DeWulf
& Hers, 1967a& 1968b) and inhibited by high concentrations of
glycogen (Hers et al., 1970) physiological concentrations of ATP
(Gilboe & Nuttall, 1973 & 1974; Gold, 1970a) and phosphorylase "a"
(Stalmans et al., 1971 & 1974a).
The two forms of glycogen synthetaSe are independent and
dependent on glucose 6-phosphate for activity (respectively) when
measured at saturating concentrations of MPG (Hornbrook et al.,
1966) and the nomenclature "In and "D" was based on this concept.
30
It is now apparent, however, that synthetase "I" is stimulated
by glucose 6-P when assayed in the presence of physiological
concentrations of DFDPG and is thus not independent Of the sugar
nucleotide for activity (Hornbrook et al., 1966; Mersmann & Segal,
1967) and that under these conditions the "D" form of the enzyme
is virtually insensitive to glucose 6-phosphate in the physiological
range of concentrations. The role of glucose 6-phosphate in the
regulation of liver synthetase has been doubted also by DeVUlf et
al., (1968)2who favour a major regulatory function for Pi. In
mouse liver, a physiological concentration of Pi (5)4mol/g)
stimulates the "I" form and overcomes inhibition due to 3mM ATP,
but inhibits the stimulation of the "D" form by glucose 6-P and
is unable to overcome the nucleotide inhibition of this form.
Therefore, at the concentrations of hepatic UDPG, ATP and Pi observed
in vivo, there is no stimulatory effect of glucose 6-P on the "D"
form but up to 60% stimulation of the "I" form. Since the "I" and
"D" nomenclature as originally defined for muscle does not apply
to the liver enzyme, and "b" will be used henceforth for the
"active" and "less active" forms of glycogen synthetase (see DeWulf
et al., 1968; Hers et al., 1970).
In view of the complexity of allosteric regulation of glycogen
synthetase in vitro, it is difficult to assess the role of
metabolites in modulating synthetase activity in vivo. In general,
it appears that the combined effect of inhibitory metabolites (ATP,
ADP, Pi) and stimulatory effectors (glucose 6-P, Pi, citrate) is to
maintain synthetase "a" activity and inhibit synthetase "b" in vivo,
in conditions conducive to glycogen deposition.
31
The role of hormones in the control of glycogen synthetase
will be discussed later (see Section 1.4; for reviews see Lamer &
Villar-Palasi 1971; Ryman & Whelan, 1971; Soderling & Park, 1974).
1.2.3 Hepatic glycogen phospho lase .
Two forms of phosphorylase exist in the liver, the "a" or
phosphorylated form and the "b" or dephosphorylated form (Sutherland
& Wosilait, 1956). They are interconverted by phosphorylase phosphatase
which is activated by glucose (Stalmans et al.,/19790 4:nd ATP (Merlevede
et al., 1969) and inhibited by AMP (Wosilait & Sutherland 1956), and
phosphorylase kinase which is activated by cyclic AMP (Rail & Sutherland
1958). There is some evidence to suggest that the phosphatase exists
in two forms (Merlevede et al., 1969)as is the case with the muscle
enzyme. The kinase has not been purified from liver. •
In muscle the "a" form is active in the absence of other factors
but the "b" requires AMP. The situation is not as clear in the case of
the liver enzyme for although "a" is intrinsically active, 1mM AMP
increases its activity by about 15 to 40% (dog liver: Sutherland &
Wosilait, 1956; rabbit liver: Wolf et al., 1970). In the absence of
AMP, phosphoi:ylase "b" was thought to be virtually inactive until the
recent work of Tan and. Nuttall (1974) who demonstrated that the rat
liver enzyme was active in the absence of the nucleotide. The
reported values for activity of phosphorylase "b" in the presence of
AMP are variable: 15% of total phosphorylase activity for dog liver
enzyme (Wosilait & Sutherland, 1956); 0.2% for pig and rabbit (Appleman
et al., 1966); up to 25% for mouse liver (DeWulf, 1971) and 80% for
rat liver (Tan & Nuttall, 1974).
The situation, therefore, as regards the activities of phosphorylase
"a" and "b" is not clear, although it is likely that in vivo,
phosphorylase "a" is the "active" form and that "b" is converted to
nan when an increase in phosphorylase activity is required. Such
a change is clearly established in response to hormones or hypoxia
for example; the roles of hormones and cyclic AMP on the enzyme
glycogen phosphorylase will be discussed later (see Section 1.4 and
for review see Fischer et al., 1970; Soderling & Park, 1974).
33
1.3 THE CIRCULATING PRECURSORS OF HEPATIC GLYCOGEN
The circulatory precursors which provide the carbon atoms of
hepatic glycogen have not been fully identified. Synthesis and
de gradation of hepatic glycogen is closely related to the intrinsic
hepatic homeostatic control of blood glucose concentration. It
has been shown (Soskin, 1938 & 1941) that when circulatory glucose
levels are high there is glucose uptake by the liver and that during
hypoglycaemia glucose is released by the liver.
The influence of increasing glucose levels has been widely
studied and in general it has been found that glycogen synthesis is
enhanced in a dose-dependent manner. This has been shown in a variety
of systems: liver slices (Ballard & Oliver, 1964; Cahill et al., 1958),
perfused livers (Buschiazzo et al., 1970; Glinsmann et al., 1970; Haft
1967; Rudman & Herrera, 1968; Sokal et al., 1958) and in vivo
(Madison, 1969). Based on the properties of glucokinase and glycogen
synthetase the following was proposed (Leloir, 1964 & 1967; Steiner
1964): an increase in glucose concentration is expected to accelerate
glucose phosphorylation and lead to an increase in the intrahepatic
concentration of glucose 6-phosphate; stimulated by this metabolite,
glycogen synthetase converts the glucose moiety of UDPG into glycogen.
An increase in glycogen accumulation could also occur because of the
known activation of hepatic glycogen synthetase phosphatase (DeWUlf &
Hers, 1967a & 1968b) and phosphorylase phosphatase (Stalmans et al.,1970
• 1974a)by glucose.
34
However, in the starved-refed state the role of glucose as a
glycogen precursor is likely to be diminished due to the reduced
levels of glucokinase on starvation (Salas et al., 1963; Walker &
Rao, 1964). When material other than glucose is administered after
starvation, hepatic glycogen deposition (Hornbrook et al., 1965 &
1966; Winternitz et al., 1957) is likely to be due to gluconeogenesis
(defined as the synthesis of glucose in a monomer or polymer form).
If, however glucose is ingested after starvation, the accumulation
of hepatic glycogen could be due to continued gluconeogenesis or
hepatic uptake of circulating glucose. From experiments in vivo
with 140-labelled precursors, Olavarria et al.,(1968) suggested that
hepatic glycogen synthesis in starved-refed rats,- even when they
receive glucose, is mainly a result of gluconeogenesis, at least
initially
One of the objectives of this work was to clarify which were
the circulatory precursors of hepatic glycogen in both starved and
fed animals, and determine the role of glucose in glycogen
accumulation.
35
1.4 THE ROLE OF HORMONES IN HEPATIC GLYCOGEN METABOLISM
Perhaps the most outstanding and least clarified aspect of
glycogen metabolism is that of the role of hormones in glycogen
metabolism, where many conflicting data exist, especially with respect
to insulin (for reviews see Nuttall, 1972; Pilkis & Park, 1974) and
adrenal cortical steroids (for review see Landau,l965). This could
be due to the fact that the hormones have been tested in isolated
liver preparations under suboptimal conditions, e.g., when net glycogen
synthesis has not been observed.
1.4.1 Insulin and hepatic glycogen
One experimental approach used to study the role of insulin
in metabolism is to make animals deficient in the hormone (diabetic)
and compare their carbohydrate metabolism with that of matched normal
animals. In diabetic animals there is an impairment of extrahepatic
glucose utilisation. Their low hepatic glycogen concentrations
compared with-those of the normal fed animals (Friedmann et al., 1963),
combined with a high blood glucose (as a result of increased
gluconeogenesis: Exton et al., 1972b& 1973a; Friedmann et al., 1965;
Renold et al., 1953), point to a major impairment in the glycogen-
synthesising mechanism. Loss of control of glycogen synthetase and
phosphorylase by glucose has been observed in the perfused liver of
diabetic rats (Miller et al., 1973) which correlates with the reported
loss of glycogen synthetase - activating system (Bishop, 1970; Gold,
1970b; Nichols & Goldberg, 1972).
36
This loss has not been confirmed by DeWulf (1971) who found that
glycogen synthetase phosphatase was normal in diabetic animals. The
diabetic animal has increased activity of hepatic glycogen synthetase
"b" (Kreutner & Goldberg, 1967; Steiner et al., 1961)'and an increase
in total activity (Steiner et al., 1961).
A number of studies using 14C-labelled substrates e.g., lactate
or alanine, have shown an impaired incorporation of 14C into hepatic
glycogen, although glucose becomes highly labelled (Exton et al.,
1972b& 1973a; Friedmann et al.,111Ar)Only on administration of
insulin in vivo does glycogen become labelled, in preference to
glucose (Exton et al., 1972b& 1973a) and an increase in activity of
glycogen synthetase"a"occur (Gold, 1970b; Kreutner & Goldberg, 1967;
Miller & Larner, 1973; Steiner et al., 1961 & 1964).
Although insulin appears necessary in diabetic animals for the
restoration of net glycogen accumulation (Steiner & King, 1964) and
of the activity of glycogen synthetasej several studies in vivo
suggest that synthesis can occur in insulin-deficient animals, without
administration of the hormone. Hornbrook (1970) and Friedmann et al.,
(1963 & 1967)_have reported normal rates of glycogen accumulation
although the eventual levels of glycogen stored were less than in the
normal animal and increased after insulin administration (Friedmann
et al., 1963 & 1967). They proposed that synthesis was not impaired
in the liver but that it was the capacity for glycogen storage which
was affected in diabetes.
Study of the role of insulin in vivo in normal animals has been
complicated by the role of glucose in hepatic glycogen synthesis
(Buschiazzo et al., 1970), glycogen synthetase (DeWulf & Hers, 1967a
& 1968b) and phosphorylase (StaImans.et al., 1970 & 1974a).
37
It has been proposed that glucose and not insulin, is the important
factor in the activation of glycogen synthesis (DeWulf & Hers, 1967a)
and glycogen synthetase (DeWulf,1971), and inhibition of phosphorylase
activity (StaImans et al., 1974a). This has been shown in perfusion
studies where the increase in glycogen synthetase and decrease in
phosphorylase activities due to glucose, were unaffected by the
presence of insulin (Glinsmann et al., 1970). Miller and Larner (1973)
have however, shown a direct rapid effect of insulin on glycogen
synthetase in perfused livers, and Hostmamk (1973) has shown in liver
perfusion that this effect is not mediated by or dependent on glucose.
The most consistent short-term hepatic action of insulin
observed in vitro has been to antagonise the effects of glucagon
and adrenalin (Exton et al., 1970 for review; Exton et al., 1973a;
Glinsmann & Mortimore, 1968; Hostmark, 1973). Under these conditions
a decrease in elevated cyclic AMP is observed, which is not apparent
when insulin is added alone. It has been suggested that this lack
of fall in cyclic AMP is due to the presence of a small metabolically
active pool upon which insulin acts (for reviews see Pilkis & Park,
1974; Walaas et al., 1974).
The discovery that insulin activated glycogen synthetase (in
muscle) led to the proposal of two rapidly interconvertible forms
of the enzyme (Villar-Palasi & Larner, 1960). Since then assay
systems have been developed which differentiate between these two
forms and have shown clearly that in intact animals, insulin causes
a conversion of the hepatic "b" (phosphorylated) form to the "a" (non-
phosphorylated) form.
38
Such activation could be a direct effect of insulin (if glucagon
is also present) mediated by prevention of both the formation and
action of cyclic AMP (Glinsmann & Mortimore, 1968). Whether there
is a separate "second messenger," different from cyclic AMP,
such as cyclic GMP (proposed by Illiano et al., 1973), or whether all
insulin effects occur via a decrease in cyclic AMP is unclear. It
is also uncertain whether insulin acts via the activation of glycogen
synthetase phosphatase (Bishop et al., 1970 & 1971; Gold, 1970b), or
. inhibition of glycogen synthetase kinase (both in the absence and
presence of glucagon: Miller & Lamer, 1973) and whether it has any
effect on glycogen phosphorylase.
As can be seen from the above there is still controversy over
the role and mechanik of insulin action in hepatic glycogen metabolism.
One group of the experiments reported here was designed to clarify this
issue.
39
1.4.2 Adrenal cort acsteroids a lauatisgia
The role of glucocorticoids in hepatic glycogen metabolism,
like that of insulin, is somewhat unclear. Although it has long
been known that the hypoglycaemia that kills adrenalectomised animals
upon prolonged starvation is due to an ultimate failure of
gluconeogenesis (Long et al., 1940), it is not clear how adrenal
steroids act on gluconeogenesis. It is known that (in intact animals)
they enhance the peripheral breakdown of proteins and stimulate the
uptake of Amino acids by the liver but their influence in isolated
liver preparations on conversion of amino acids to glucose is variable.
Friedmann et al., (1965) found that adrenalectomy did not impair
gluconeogenesis from pyruvate or alanine'inthe starved animal, unlike
Eisenstein et al., (1966) who found an impairment from alanine, which
was corrected by dexamethasome in vivo or in vitro. Exton and co.
workers also found a decrease in gluconeogenesis from lactate in the
starved adrenalectomised (1965) and diabetic-adrenalectomised rat
(1970 & 1973b) which was restored by treatment in vivo or in vitro
(1973b). Adrenalectomy curtails the stimulation of gluconeogenesis by
glucagon in the perfused liver (Exton et al., 1972a;Friedmann et al.,
1967); the glucagon response returns on treatment with glucocorticoids
in vivo. Although addition of steroids alone in vitro has no rapid
effect on gluconeogenesis (in the absence of glucagon), restoration
of the glucagon response is seen with dexamethasome (Friedmann et al.,
1967).
Although hepatic gluconeogenesis appears to be little altered
by steroids in vitro, hepatic glycogen synthesis is low after
adrenalectomy (Friedmann et al., 1965) and is not responsive to hormone
40
added in vitro (diabetic-adrenalectomised rats, Exton et al., 197313);
treatment in vivo for 2-3h being required for glycogen accumulation to be
restored (Hornbrook et al., 1966; Nichols & Goldberg, 1972). However,
it has been reported that hepatic glycogen synthesis is not
significantly impaired in vivo (Friedmann et al., 1967; Kreutner &
Goldberg, 1967) although a two hour lag in synthesis was observed
which was shortened by prior steroid treatment in vivo (Kreutner &
Goldberg, 1967). The rate of glycogenesis wasas rapid as in the normal
animal but "plateaued" at a lower level, indicating that steroids may
be involved in the amount of glycogen stored.
In the fasted•adrenalectomised rat most of hepatic glycogen
synthetase is in the inactive "b" form (Glinsmann et al., 1970;
Homnbrook et al., 1966; Mersmann & Segal, 1969) due to the low
activity of glycogen synthetase activating system; it is restored
after 2-3h hydrocortisone treatment in vivo (Gruhner & Segal, 1970;
Mersmann &.Segal, 1969). There does not however, appear to be such
an impairment in the fed adrenalectomised animal where the conversion
of glycogen synthetase "b" to "a" can be activated by glucose in vitro
(Glinsmann et al., 1970; Miller et al., 1973). The response of glycogen
phosphorylase to glucose is also impaired, in the starved adrenalect-
omised rat (Miller et al., 1973). These differences, between fed and
starved animals, could implicate insulin in steroid action (since
insulin levels are higher in fed animals). Although the lack of adrenal
corticosteroids results in no change in the levels of"activeu hepatic
glycogen phosphorylase, the levels of inactive enzyme are greatly
diminished; restoration of the enzyme activity and hyperglycaemic
effects of adrenalin and cyclic AMC' being obtained with steroid
replacement (Schaeffer et al., 1969).
Studies on the role of steroids in the normal animal have been
extensive (for review see Landau 1965), their administration in
vivo leading to an increase in blood glucose concentration, hepatic
glycogen and glycogen synthetase "a", and decrease in phosphorylase
activity (DeWulf & Hers, 1967b & 1968b; Hornbrook et al., 1966 & 1970).
There does however, seem to be a time lag (shorter than in
adrenalectomised animals) before a response is observed. The effects
of steroids on glycogen accumulation in vitro are not clear.
The restoration of hepatic glycogen accumulation and increase
in glycogen synthetase "a" (associated with an increase in glycogen
synthetase phosphatase) seen in adrenalectomised animals treated with
hydrocortisone (Nichols & Goldberg, 1972) is not observed within 2h
in adrenalectomised-diabetic rats (Exton et al., 1973b; Nichols &
Goldberg, 1972). Restoration of glycogen synthetase "a" activity
is however, seen after 2-4 min of insulin treatment (Nichols &
Goldberg, 1972). It was proposed (Hornbrook, 1970; Kreutner & Goldberg,
1967; Nichols & Goldberg, 1972) that the ability of steroids to induce
glycogen deposition is largely dependent upon insulin release. Insulin
secretion is reduced to 60X, by adxenalectomy OPIalaisse et al., 1967) and it
is thought that steroids increase the sensitivity of insulin secretion
to glucose.
It is not clear therefore, whether adrenal corticosteroids exert
direct effects on hepatic glycogen metabolism or whether their action
(in vivo) is mediated through insulin or glucose. These aspects were
investigated in the experiments to be described.
42
1.4.3 "Gle°n°1-bicithor icirc°ge
an
n
The above two hormones (insulin and glucocorticoids) are
generally considered to be involved in the control of glycogen
accumulation. Glycogen breakdown is however, as important a process,
since it is the balance between the two which governs the actual
amount of glycogen stored. The two main hormones known to cause
glycogenolysis are glucagon and adrenalin (Exton et al., 1970 for
review), although the importance of the latter hormone in hepatic,
glycogen breakdown in rodents is thought to be minimal (Exton & Park,
'1968; Sokal et al., 1964).
Additional hormones can cause glycogenolysis. For example,
early work demonstrated that extracts of the neurohypophysis could
cause hyperglycaemia (Clark, 1928; Claude & Baudouin, 1912) and
prevent the hypoglycaemia due to insulin (Burn, 1923) perhaps as a
result of breakdown of liver glycogen (Clark, 1928; Imrie, 1929; Stehle,
1950). When purer preparations of hormones became available
vasopressin was shown to bring about an increase in blood glucose in
mammals (Bergen et al., 1960; Cash & Kaplan, 1964; Schillinger et al.,
1972), possibly through hepatic glycogenolysis (Bergen et al., 1960;
Heidenreich et al., 1963). Evidence showing that this hyperglycaemia
was due to hepatic glycogenolysis was obtained by Heidenreich et al.,
(1963) in liver slices and Vaisler ( 1965a ) in perfased liver, who
found that oxytocin displayed a similar action, although the response
was smaller and of a shorter duration that vasopressin (Vaisler, 1965b).
The significance of these results has been uncertain, since the
concentrations of the hormones were usually higher than those which may
occur "physiologically". One portion of the investigation reported
here was designed to clarify the roles of vasopressin and oxytocin in
liver carbohydrate metabolism.
43
1.5 SCOPE AND AIMS OF THE PRESENT STUDY
1.5.1 General considerations
From the above, it is obvious that the role of hormones in
particular in hepatic glycogen metabolism is unclear. As has been
stated, most of the work has been carried out in systems which have
not exhibited the rates of glysogen accumulation which are known to
occur in vivo.
The aim of this research was to obtain conditions in the
perfused liver which are suitable for maximal net rates of glycogenesis,
comparable to the in vivo rates, and to study the role of hormones in
this state. In parallel with measurements of glycogen and substrates,
the enzymes of glycogenesis and glycogenolysis have been measured under
conditions of maximal and sub-maximal glycogen accumulation. Glycogen
breakdown is as important in the liver as glycogen synthesis and the
role of hormones in this aspect of glycogen metabolism has also been
investigated, with a similar group of measurements.
1.5.2 The use of the perfused liver for metabolic studies
There are a number of different experimental techniques available
for metabolic studies. They may be divided into two groups: those
using the intact animal and those involving isolated tissue preparations.
The use of the intact animal for metabolic studies of a
particular tissue has a number of disadvantages. It is difficult to
produce a specific change in one tissue without stimulating parallel or
compensatory changes in others, or without initiating nervous or
hormonal discharges which could produce marked metabolic changes in the
tissue under investigation.
Thus metabolic studies in the whole animal do not yield insights
into the specific or primary sites of action of circulating hormones
or substrates, but may be used for obtaining confirmatory evidence
for theories of metabolic control.
There are four main types of isolated tissue preparation:-
(a) homogenates, (b) cells (c) pieces or slices and (d) organ perfusion.
Metabolic studies involving the use of homogenates are primarily
utilised for the assay of various tissue constituents including
enzymes. The use of tissue slices, on the other hand, has been a
very productive technique, a large part of basic metabolism (especially
the elucidation of pathways) being based on data obtained by this
method. Liver slices do however, have a number of disadvantages in
the study of metabolic control. They lose sugar phosphates,
nucleotides, proteins and ions and are subject to variable diffusion
properties. The use of hepatocytes in metabolic investigations has
an advantage compared with liver slices, in that metabolic functions
(e.g., gluconeogenesis) are maintained. However, until recently
their response to hormones did not appear to be as sensitive as the
perfused live.
Metabolic studies have been extensively carried out using the
perfused liver. It has a number of important advantages over the
techniques described above in that it is the most "physiological"
system, it involves the least disruptive procedures during the
preparation, and the perfusate may be easily controlled and monitored.
Per-haps however, the most important aspect of liver perfusion is that
the role of hormones may be evaluated, especially to elucidate whether
there is direct or indirect hepatic action.
45
Although this could be tested in slice or cell systems they have
the disadvantages given above with the additional factor that
hepatocyte studies tend to isolate particular cell types neglecting
the role of non-parenchymal cells, and the circulation within the
liver.
In view of considerations such as these, the liver perfusion
technique was employed in the present study.
CHAPTER TWO
ANIMALS, MATKRIALS & METHODS
47
CHAPTER 2
ANIMALS, MATERIALS AND METHODS
2.1 THE PREPARATION OF ANIMALS
2.2 SOURCES OF MATERIALS
2.3 THE TECHNIQUE OF LIVER PERFUSION.
1. Perfusion apparatus
2. Perfusion medium
3. Surgical procedure for liver perfusion
4. Sample preparation
2.4 ME TECHNIQUES USED IN INTACT ANIMAL EXPERIMENTS
1. Measurement of net glycogen accumulation in vivo.
' .2. Measurement of the enzymes of glycogen metabolism in response to hormones.
2.5 ANALYTICAL METHODS
1. Glucose and glucose polymer determination.
2. Carbohydrate metabolite determination.
3. Determination of nitrogenous compounds.
Assay for the enzymes glycogen synthetase and phosphorylase.
48
2.1 THE PREPARATION OF ANIMALS
Albino, male, Sprague -Dawley rats of CFY strain derived
from Carworth, Europe were bred in the Biochemistry Department of
Imperial College; the male stock were renewed every three months,
ensuring a close genetic relationship to the foundation stock. Animals
weighing between 170 and 220g were allowed free access to a standard
(Thompson?s) cereal diet, and water. They were maintained on a light/
dark cycle of 12h (daylight period 06,00h - 18.00h GMT), in a relative
humidity of 55%, and in a temperature of 19-23 °C.
Some rats were starved from 5p.m. until used at 10a.m. or
2p.m. the following day,but the majority of animals were starved for
'48h from 10a.m.
Diabetes were induced by injecting 75mg/kg streptozotocin
(dissolved in about 0.25m1 0.01M citrate - Na pH 4.5) into a tail vein
of a rat under ether anaesthesia. The initial weight (about 210g) on
injection was noted and the animals weighed on subsequent days at a
comparable time-of day (Fig. 2). Four days after injection the blood
glucose of the animal was determined by sampling from a tail vein, and
only if an elevated blood glucose (>12mM) and loss in body weight was
observed, was the rat used in subsequent experiments. It should be
noted that if the animals were not checked at the same time of day each
day, a false impression of changes could be obtained (Table 1). A
similar weight loss between the morning and afternoon weighings was
observed in both the streptozotocin and citrate injected animals but
there was a significant change in the blood glacose of the diabetic
animal compared with the normal.
1
2
4 DAYS
Fig. 2 Growth curve of streptozotocin and control injected
rats
Streptozotocin (75mg/kg) was injected in citrate buffer 0.0IM p114.5
into a tail vein on day zero (0) or citrate buffer alone was injected (0).
Rats were weighed at a comparable time each subsequent day. Results are
mean values S.E.M. of 8 streptozotocin injected animals and 3 controls.
Table 1. The effect of time of day on rat body weight
and blood glucose concentration.
Streptozotocin (75mekg), in citrate buffer 0.0IM pH4.5., was injected
intravenously, or citrate buffer alone. After 4 days the rats were
weighed and the blood glucose determined at the times shown. Each result
is from one observation.
11.30 a.m. 2.30 p.m.
Streptozotocin
Rat No.
1
2
3
1
2
vb. (g)
185
195
212
236
236
Blood glucose 01
20.4
21.6
21.6
6.0
5.2
(mM wt. (g)
180
193
200
229
231
Blood glucose
14.8
16.0
19.4
6.4
5.3
injected rats.
Citrate injected rats.
This was to be expected as animals rarely eat during the day, and
starving is known to partially overcome the elevated blood glucose in
the diabetic state. The rats were starved in separate cages for 48h
prior to use. The rats were not insulin-maintained and some deaths
did occur, especially during starvation. All animals were used within
seven to eleven days of the injection.
Bilateral adrenalectomy was performed' under ether anaethesia
by means of'two lateral incisions. If both adrenal glands were not
clearly removed, in their entirety, the animal was discarded. After
the operation the rats were kept on 0.9% (W/V) sodium.chloride and
weighed every dayI at the same time of day. Their initial weight was
about 165g. (Fig. 3)$ and adrenalectomised rats gained less weight than
controls. Some seasonal variation in the growth of adrenalectomised
animals was noted (Table 2), the animals gaining more weight in the
winter than in the summer. The animals were starved in separate cages
for 48h prior to use, during which some fatalities did occur. All
animals were used within seven to eleven days of the operation.
205
195,
0 *--‘185
I-
W175
165
2 155
2
DAYS
1 3
Fig. 3 Growth curve of adrenalectomised, sham-operated, normal
on 0.9% Nacl and normal rats.
Bilateral adrenalectomy was performed as described in the text and the
rats weighed at the comparable time of day, each subsequent day. Results
are mean values + S.E.M. of at least 24 adrenalectomised animals (0), 8
sham-operated rats maintained on 0.9% Nacl (0), 3 normal maintained on
Nacl (C) and 3 normal animals W. The first two groups of experiments
were carried out in the summer and the last two in the winter.
53
Table 2. Seasonal variation in the growth of
adrenalectomised rats.
Bilateral adrenalectomy was performed as described in the text,
and the rats weighed at the same time of day each day. Results are
means.+ S.E.M.s
Month
with the number of observations in parentheses.
Weight on Weight after Chan e in operation CO 4 days (g) weldL_ )
May 171+2 1641, - 7 (4)
June 164+2 16219 - 2 (4)
July 162+2 — 164+2 + 2 (5)
September 163+2 162+2 1 (18)
October 174+2. 182±3 8 (12)
November 167+2 1682:3 (5)
January 182+2 186+2 4 (12)
February 1.610 173+4 + 12 (5)
March 165+1 175+2 + 10 (6)
54
202 SOURCES OF MATERIALS
All reagents were Analar grade and obtained from Hopkins and
Williams (Ghadwell Heath, Essex), May and Baker (Dagenham, Essex), BDH
(Poole, -Dorset) or Fisons (Loughborough, Leicestershire) unless otherwise
stated. Enzymes for analytical purposes, nicotinamide-adenine
nucleotides, sugar phosphates, pyruvate (mono-sodium salt), UDPG (disodium
salt) and AMP were obtained from C.F. Boehringer Corp Ltd., (London).
Oleic acid, lactic acid, fructose, amino acids, Trizma base and the
hormones hydrocortisone (-21-sodium succinate), vasopressin (8-arginine
and 8- lysine oxytocin and adrenalin (bitartrate) were bought from
Sigma (Kingston-upon-Thames, Surrey).
(8- Arginine)- vasopressin (Grade VI) obtained in solution form
was prepared from synthetic vasopressin of activity about 360 units/Mg;
the concentration of each batch, checked by assay of its antidiuretic
effect in the ethanol loaded rat (carried out by M. Forsling) was about
80% of the stated activity, this factor was taken into account in
calculating the concentration of vasopressin. Synthetic (8 -lysine) -
vasopressin (Grade IV : essentially oxytocin-free) was a powder of
activity 70-100 units/mg and oxytocin (Grade III : synthetic) was in
aqueous solution and reported to be free of vasopressor activity.
Glycogen (rabbit liver) and some amino acids were from BDH, glycerol and
glucose from Fisons. Radiochemicals were obtained from the Radiochemicals
Centre (Amersham, Bucks). Heparin was from Evans Medical Supplies
(Liverpool) and Nembutal from Abbott (Queenboraugh, Kent). Streptozotocin
was prepared and donated by E. Karunanayake (Imperial College, London)
or bought from Upjohns (Michigan, U.S.A.) and recrystallised in ethanol.
55
Insulin was the highest grade commercial ox preparation from Burroughs
Wellcome (Dartford, Kent) and glucagon (crystalline) was obtained from
Eli Lilly (Indianapolis, U.S.A.). Anti-insulin (and control) serum / Dept.
was prepared in the Biochemistry/at Imperial College from guinea-pigs
(Mhnsford, 1967).
56
2.3 THE TECHNIQUE OF LIVER PERFUSION
2.3.1 Perfusion apparatus
The apparatus was essentially that described by Hems et al.,
(1966) based on the designs of Miller et al., (1951) and of Schimassek
(1963). Perfusions were carried out in a thermostatically controlled
cabinet with sash Perspex windows. The contact thermometer in the cabinet
was maintained at 37°C but due to the positioning of the fan, at the top
of the cabinet, the medium entering the liver was a degree lower at
35-36°C. In order to quickly restore the steady temperature after any
____ loss of heat due to the opening of the window for sampling, an
additional fan was placed in the cabinet which could be independently
controlled. The perfusion medium was constantly mixed by magnetic stirrers
built into the floor of the cabinet and small apertures were made in the
sides of the cabinet for the gas supply. This assembly thus provided
a constant temperature environment for perfusion.
The arrangement of the glassware and tubing was as shown (Fig. 4).
'The perfusion medium was pumped from a collecting vessel by a MERE roller
pump (supplied by Watson Marlow Ltd., Cornwall, U.K.) and: passed through
a plastic mesh filter, taken from a disposable blood transfusion set. It
then passed to an oxygenator which was maintained in a vertical position,
facilitating the even flow of perfusion fluid over the surface. The
bulbous nature of the oxygenator increased the surface area of the medium
available for gas exchange. The gas was saturated with water by
bubbling through a wash bottle fitted with a sintered-glas3distributor,
and entered the oxygenator at the bottom and left through an outlet at
the top. At the base of the oxygenator was a small resevoir of perfusate
which was kept at a constant height by an overflow tube leading to the
Filter
✓ Gas
outlet
Oxygenator
Gas in via
wash bottle
Platform for
animal
Outflow to liver
Outflow from liver
Roller pump
Collecting vessel
Fig. 4
aratus for the erfusion of rat liver. • •
57
58
collecting vessel. The height of this resevoir could be adjusted
to give a hydrostatic pressure for optimum flow rates without the liver
swelling. The height was kept at 18 cm. The input to the liver led
from the bottom of the oxygenator via a length of silicon tubing
supplied with a roller clamp, enabling fine adjustments of flow. Th6
perfusion medium from the liver was then returned to the collecting
vessel, and ,the recirculation continued.
59
2.3.2 Perfusion medium
In all experiments the perfusion medium initially consisted
of 50m1 Krebs-Ringer bicarbonate buffer (Krebs & Henseleit, 1932)
and 10m1 15% (N/V) bovine serum albumin. The albumin (Pentex : Fraction
5 from Miles-Seravac Ltd., Berks) was dialysed from three days against
four changes of gassed Krebs-Ringer bicarbonate before use and then
kept frozen at - 20°C. Normally 28-30mM glucose was present in the
medium.
Red blood cells were obtained from a fed donor rat of more
than 600g (one animal, yielding sufficient blood for one perfusion).
The animal was bled from the aorta under ether anaesthesia and the
blood defibrinated on glass beads (Baron & Roberts, 1963) by a rotary
motion of a siliconised flask. After a period of half-an-hour, allowed
for clot contraction, the supernatant was divided into two portions and
washed twice with 20 volumes gassed Krebs-Ringer bicarbonate containing
5mM glucose, except when gluconeogenesis was being studied. A few
experiments were carried out with whole defibrinated blood. The red cells
were then made up to the original blood volume with bicarbonate buffer.
On addition to the perfusate after the start of perfusion, this gave
a haemoglobin of about 4%. In most cases, glucose was added to give an
initial concentration of 28-30MM. Red cell glycolysis was tested in the
absence of a liver, in the presence of 6mM glucose, and was found to
be negligible. The pH of the medium was tested before and during perfusion
and was 7.3 - 7.5. Haemolysis during perfusion was about 1% /h
determined using Drabkin's reagent : 20/41 of sample was added to 4m1
Drabkints (0.05g KO and 0.2g K3Fe (GN)6 / 1) mixed and allowed to stand•
' for 5 min.
60
The resultant coloured solution was read against water at 540nm;
standard 18% haemoglobin gave an optical density of 0.618.
Substrates or hormones were added as an initial dose to the
perfusion medium about 15min after thebeginningof the perfusion and then
infused with a Delta pump (Watson-Marlow Ltd.) at such a rate as to'main-
tain approximately the concentration. The substrate and hormone
concentrations varied from experiment to experiment and will be stated
where necessary. In a number of experiments a mixture of amino acids
was used. This was kept as a stock solution, at concentrations
approximately one hundred times "physiological". "Physiological" levels
were taken as the following, the values being based on the data of
Scharff and Wool (1966) : arginine 0.1mM, lysine 0.4MM, histidine 0.1mM,
pheayialanine 0.1MM, leucine 0.2mM, isoleucine 0.1MM methionine 0.1mM,
valine 0.3mM cysteine 0.1MM, alanine 0.5mM, glycine 0.4MM, glutamic
acid 0.1MM, serine 0.4mM, asparagine 0.2mM, threonine 0.2mM, glutamine
0.6mM, aspartic acid 0.05mM and tryptophan 0.1mM. Tyrosine was added
separately in powder form, giving a "physiological" concentration of 0.2mM
in the perfusate. The stock amino acid mixture was added to give initial
concentrations of one or four times "physiological" (expressed per vol.
whole blood), and then infused at 3ml/h in some experiments.
In some perfusions a supplemented medium was used (essentially
according to John & Miller, 1969), in which were present initially, in
addition to carbohydrate substrates : insulin (500mU), hydrocortisone 21 -
sodium succinate (lmg) and the above amino acid mixture at about four
times the normal concentrations in fed rats.
61
These constituents were maintained by infusion (1.5m1 or 3m1/h
respectively) of separate solutions (in water) containing insulin,
330mU/ml, plus hydrocortisone, 33014-g/ml, or amino acids (see above.)
2.3.3 Surgical procedure for liver perfusion
The operation was carried out essentially as described by
Hems et al., (1966). In early work the animal was anaesthetised with
an intraperitoneal injection of Nembutal (60mg in 0.1m1/100g body
weight) but most operations were carried out under diethyl ether
anaesthesia.The rat was placedon the operating platform and taped
into position. A beaker containing diethyl ether-soaked cotton wool
was kept over the animals head during the operation procedure. A
horizontal cut was made in the abdomen and the intestines deflected to
the animal's left onto saline-soaked tissue. Heparin (0.2m1 = 200
units) was injected into the inferior vela cava and the injection point
covered with a'tissue. The thin strands of connective tissue between
the right lobe of the liver and the vena cava were cut and a loose
ligature of silk (size3/0) placed around the vena cava above the right
renal vein. A double thread was then tied loosely around the base of
the hepatic portal vein and another single ligature put around the vein,
including the hepatic artery. The portal vein was then cannulated with
a No.17 Frankis -Evans needle (trocar and cannula, Luer fitting), the
needle removed and the double thread tied to hold the cannula in place.
A rapid backflow was usually observed which indicated a good and rapid
operation and that the animal was at the correct stage of anaesthesia.
If there was no backflow the cannula was carefully filled with Krebs-
Ringer bicarbonate so as to exclude air bubbles.
62
The thorax was then quickly opened by a transverse incision
above and along the line of the diaphragm and by two deep longitudinal
cuts towards the head. This flap was then removed above the heart.
A loose ligature was placed around the arterior vena cava and a cannula
of Portex tubing (3.00 x 2.00mm / 2.42 x 1.67mm drawn out to a bevelled
point) inserted through the right atrium into the vena cava and down as
far as the diaphragm. The cannula was tied in place.
The preparation was then connected up to the perfusion apparatus
and the first 20m1. medium discarded. During this time the inferior vena
cava ligature and the single one around the hepatic portal vein were
tied. Care was taken that the ties enclosed the camnulae and did not
impede the flow of medium.
After the discard the platform plus animal was placed on the
collecting vessel, the recirculation of medium started and the washed
red cells added. The whole operation took about 10 minutes but the time
from the insertion of the hepatic portal vein cannula to the connection
to the perfusion medium was less than 2 minutes. An indication of the
success of the operation was the uniform olive-green colour obtained
during the washout and the even red-brown colour observed on addition
of the red cells. An extended interruption of the liver circulation
could cause a patchy liver seen either during wash out or when the red
cells were added.
A cage was put over the liver, covered in tissue soaked in
warm Krebs'-Ringer bicarbonate which was itself prevented from drying by a
polythene sheet on top. The flow rate was then adjusted to give a flow
of 16.8 - 18.4m1 /min; difficulty in obtaining this flow rate gave an
indication of lack of success of the operation. The duodenum was
cannulated to allow free flow of bile. There appeared to be no correlation
between the bile produced during perfusion and the rate of glycogen
63
accumulation although it was usually 0.3 - 0.6m1 after 50min of
perfusion. Five minutes after the start of perfusion the gas was changed
from 02 : CO2 (95 : 5) to air : CO
2 (95 : 5). Early experiments were '
carried out with the 02 : • CO2 mixture only.
2.3.4 Sample preparation
Medium samples were removed from the collecting vessel and
mixed with an equal volume of. egperchloricsacid (usually 2ml-i4ml)
for determination after spinning, of glucose or lactate.
Liver samples were removed by looping a thread around the lobe
to be sampled and tying as tightly as possible without shearing through
the liver. The sample would then be taken with the minimum loss of
medium from the cut surface and enabled at least two biopsies to be
taken from each experiment. The liver sample was promptly frozen in
liquid nitrogen and kept at -20°C for eventual determination of glycogen,
enzymes or metabolic intermediates.
64
2.4 THE TECHNIQUES USED IN INTACT ANIMAL EXPEaIMENTS
2.4.1 Measurement of net glycogen apcumulation in vivo
Infusion experiments in intact rats were carried out
as follows. Rats were anaesthetised with an intraperitoneal injection
of Nembutal (0.1m1/100g). Approximately 0.1m1 of 0.9% (W/V) sodium
chloride was injected into a tail vein of the animal; after a few seconds
the syringe was removed. A smallbackflowof blood was observed if
the needle was in the tail vein. The infusion tubing was then connected
to the needle and the pump (Delta : Watson-Marlow, U.K.) was started.
As soon as possible after the infusion was started a liver biopsy for
glycogen determination was taken by making a small transverse incision
to the left of the animals midline and manipulating a lobe of the liver
(left lateral lobe) to the exterior. The biopsy procedure was as for
the perfusion experiments. A small piece of saline soaked tissue was
then placed in the abdomen and the animal kept warm by the proximity
of a lamp. Sixty minutes after the first sample, the second (median
lobe) was taken, and also some blood for glucose determination,
• 65
2.4.2. Measurement of the enzymes of glycogen
metabolism in reponse to hormones
The experimental system was carefully chosen in order to
reduce stress and thus minimise enzyme changes due to stress. Fed
rats were anaesthetised with Nembutal, the abdomen opened and the
rat kept warm by a nearby lamp for 50min. Preliminary experiments
established that steady basal (lowest) hepatic ca' concentrations were
attained 15-20min after opening the abdomen (Kirk and Hems, 1974) and
thus hepatic metabolism was normal by this time. The hormones were
then injected (in 0.25m1 NaC1) into the hepatic portal vein; glucagon,
8 1.0 g; .adrenalin 1.5 x 10 - mol and (8-arginine)-vasopressin, 10 or
100m units. The control injections were 0.9% NaCl. The needle and
syringe were left in position in the blood vessel to eliminate any loss
of blood and thus hormone. The liver was then removed after various
times, rapidly frozen in liquid nitrogen and kept at -20°C until the
enzymes were assayed.
66
2.5 ANALYTICAL METHODS
2.5.1 Glucose and glucose polymer determination
About 0.5g of frozen liver was ground to a powder and boiled
for a few minutes in 10 volumes of aqueous 30% (W/V) KOH. The sample
(which could be stored at this stage) was then boiled for 30 min , and
the glycogen was precipitated with 3 volumes of ethanol (modification
of the method by Good, et al., 1933) and left overnight at 4°C. The
glycogen was then sedimented by centrifugation at 4°C at 10,000 r.p.m
for 15 mins the supernatant discarded and the pellet taken up in
20m1 water, using a motor driven pestle. A suitably sized sample
(usually 0.15m1) containing 0.05 - 0.31-moles of glycogen-glucose was
then hydrolysed to glucose with 30/A1 amyl oglucosidase
in 25MM sodium acetate 1E4.8 (final volume 1m1) for lh at 37°C (Lee
and Whelan, 1966). There was a 98% recovery of glycogen assayed by
this procedure. When 14C-glycogen was counted, glycogen (15-50/Amoles
of glucose) was washed twice in aqueous 70% (V/V) ethanol, hydrolysed
lml of dilute enzyme overnight and dissolved for liquid scintillation
counting in a toltene-based scintillator fluid (see Section 2.5.4)
Glucose in perchloric extracts of perfusion medium,or blood,
or after the hydrolysis of glycogen,was determined with glucose oxidase
(Krebs et al., 1964). A'lml. sample was prepared containing 0.01 - 0.3
rmoles of glucose and to this was added 2.5m1 of enzyme reagent,
composed of 12.5mg glucose oxidase, 4mg peroxidase and 0.5m1
dianisidine in 95% ethanol, per 100m1. glucose buffer (0.5M Na2HPO
4'
0.5M NaH2 PO4 and 0.1M Tris pH 7.3). This was incubated for lh at
37°C and the resultant brown colouration due to oxidation of diarisidine
read at 440nm against a reagent blank.
67
2•5• • Carbohydrate metabolite determinations
Lactate
Lactate was determined with lactate dehydrogenase (Hohorst,
1963). A stock solution containing o.am hydrazine sulphate 2.0M
glycine and 0.02M EDTA (disodium salt) was stored at 4°C until required.
Each day the pH was adjusted to 9.5 and then made up to 2 volumes with
water. The neutralised sample (usually 50/1) was added to a cuvette
containing 1.5m1 of the above buffer, 0.2ml 15mM (1yo) NAD and 1.25m1
water and the extinction read at 340nm against water. Once the optical
density reading was stable 10)A1 lactate dehydrogenase was added and the
reaction allowed to go to completion, as shown by the attainment of a
steady extinction. Any changes due to reagent blanks were subtracted from
E340 and presuming that the molar extinction coefficient of NADH2 was
6.22 at 25°C in a lcm cell the/i -11101 lactate in the cuvette was
calculated. The validity of the assay was checked with occasional use
of"standard lactate solutions.
Glucose 6-phosphate and uridine diphosphoglucose
Glucose 6-phosphate and UDPG were determined as described by
Hohorst (1963) and Mills and Smith (1963) respectively. The analyses
were carried out in sequence in the cuvette.
Liver samples were homogenised in 6% (0) perchloric acid
(1+4 volumes) (Hems & Brosnan, 1970) centrifuged and the pH of the
supernatant adjusted to 7. After standing at 0°C for lh the KC104
precipitate was removed by centrifugation and the supernatant assayed.
(a) Glucose 6-phosphate
lml of the neutralised extract was added to a cuvette
containing 0.9m1 0.IM- Tris -HC1 pH 7.8, 0.0Pm1 0.5M MEC12,
0.02m1 0.25M cysteine (freshly made each day)
and 0.05m1 NAT2. The extinction was read at 340nm against water
and once the reading was stable 2041.41 glucose 6-phosphate dehydrogenase
was added and the final optical density reading taken after completion
of the reaction. The same procedure was applied to tissue and reagent
blanks. Thelt.mol glucose 6-phosphate in the cuvette, and thus
original sample, was calculated, presuming that the molar extinction
coefficient of NA1PH2 was 6.22 at 25°C in a lcm cell. The validity of
the assay was checked by the use of standard glucose 6-phosphate
.solutions; cysteine did not interfere with the assay.
(b) Uridine diphosphoglucose. (.2m3 irn
0.01m1 phosphoglucomutaseiwas added to the cuvette after
completion of the above reaction. No change in optical density reading
at 340nm was observed. 0.02m1 0.I17 potassium pyrophosphate was then
added and no change in E340
noted. Finally 0.02m1 UDPG pyrophosphorylase
was added to the cuvette and the enzymatic reaction allowed to go to
• completion. The same procedure was applied to reagent and tissue blanks;
the recovery of UDPG was 85%. The UDPG in the cuvette and liver sample
was calculated from the molar extinction coefficient of NADPH2 of 6.22
at 340nm.
69
205.3 Determination of nitrogenous compounds
Amino acids.
Perfusion medium was deproteinised with 20 sulphosalicyclic
acid, centrifuged at 3,000r.p.m for 5 min , and the supernatant diluted
and assayed for amino acids.
The amino acids were analysed by means of automated ion
exchange chromatography (Thomas, 1969), the separation and quantification
of each amino acid being completely automatic. The sample was loaded
onto a cation exchange column (Zeocarb 225) from the sampler via a
chromatography pump. This gave a defined flow rate and column pressure.
After electronically programmed elution of amino acids from the column with
an acid to base buffer gradient, fractions of the eluent, separated into
discrete portions by regularly placed gas bubbles, were mixed with nin-
hydrin. The intensity of the reaction was measured colorimetrically,
monitored by means of a logarithmic recorder and quantitated by means
of an integrator. A standard set of amino acids was run with each batch
of samples and the amounts in each sample calculated on the basis of peak
area.
Urea.
The method used for the determination of urea was based on the
principle that urea reacts directly with diacetyl monoxime under
strongly acidic corwi.tions to give a yellow condensation product.
The reaction can be intensified by the presence of ferric ions and
thiosemicarbazide and will then occur without such concentrated acid
reagents; the red coloured complex so formed is more linear with
concentration (Marsh et al., 1965).
Cell free perfusion medium samples were assayed for urea using
the above colour reaction incorporated into an autoanalyser system.
2.5.4 Assay for enzymes glycogen synthetase
and phosphorylase
These enzymes were determined essentially as described by
Das and Hems (1974).
Glycogen synthetase
This enzymes exists in two forms : "a" and "b", usually
assayed as "a" and "a" plus "b". 200-300mg powdered tissue were
homogenised in a Vortex mixer with 6 volumes of cold buffer containing
50MM Tris-HC1, 10mM EDTA and 50mM KF pH 7.5.. Two methods were
employed for the synthetase "a" assay : (i) Method 1 : the extract
was spun at 9,500 r.p.m at 4°C for 15min and the supernatant used for
assay. The assay mixture contained 6.67mM U1PG, 10mg/m1 glycogen,
50mM Tris-HC1 and 5mM EDTA i 7.8 (final concentrations 2/3 those stated);
(ii) Method 2 : the extract was not. spun and the UDPG was increased
to 18mM and 1501 sodium sulphate was added (final concentrations 2/3 those
stated). The mixture for the total glycogen synthetase ("a" plus "b")
assay was as for "a" (Method 1 or 2) but 7.201 glucose 6-phosphate pH
8.0 (final concentration) was present and no sodium sulphate was added.
14C-labelled MPG was added to each mixture to give 40,000 - 80,000 d.p.m
10.1m1 mixture. 0.1m1 of each assay mixture was incubated with 0.05m1
of enzyme at 30°C for 15 min (Method 1) or 10 min (Method 2), the
reaction was stopped by the addition of lml ice-cold 6% (V/V) TCA
containing lmg/m1 glycogen and 2m ml LiBr. Glycogen was precipitated
by 2 volumes of 95% (V/V) ethanol and left overnight at 4°C. The
sample was filtered on a glass filter (Millipore). The assay tube
washed 4 times with 3m1 66% ethanol in water (V/V) and the paper twice
with 2m1 The filter with precipitate was put in a scintillation vial
• and lml of 0.1M sodium acetate pH 4.8 containing 5)&1 amyloglucosidase
added and the vial was left overnight.
71
16m1 scintillation fluid (2.51 2-methoxyethanol, 2.51 toluene, 15g
butyl PBD and 125g naphthalene) were added and the vials counted in
a Packard liquid scintillation spectrometer. Synthetase rates were
calculated as mol glucose transferred from the specific radioactivity
of the precursor.
Glycogen phosphorylase
50-100mg of powdered tissue were homogenised in a Vortex
mixer with 20 volumes of cold buffer containing 35mM 0:- glycerophosphate,
30mM cysteine, 1mM EDTA and 20MM NaP ; final pH was 6.1. This buffer
was made up each day. In some determinations the extract was spun at
9,500 r.p.m at 4°C for 15min, and the supernatant used for assay. The
assay mixture contained 32mM glucose 1-phosphate, 2% glycogen, 2mM AMP
(final concentrations half those stated) and 14C-labelled glucose
1-phosphate to give 40,000 d.p.m/0.2m1 mixture. AMP was included in
the assay to systematise the contribution of phosphorylase wb" to the
activity recorded.
0.2m1 of the assay mixture was incubated with 0.2m1 of enzyme
extract at 3000-for 10min. The remainder of"the procedure was as for
glycogen synthetase.
CHAPTER THREE
RESULTS
73
3.1' THE CHARACTERISTICS AND CONTROL OF HEPATIC GLYCOGEN
SYNTHESIS IN THE NORMAL 48H- STARVED RAT
1. The validation of sequential liver samplin
in the perfusion.
2. The role of. glucose and gluconeogenic precursors
in glycogen deposition in the perfused liver.
The role of insulin and fatty acids in hepatic
glycogen metabolism.
4. Characteristics of glycogen synthetase and
phosphorylase in the liver.
Control of glycogen synthesis in the perfused
liver of normal starved rats.
6. Hepatic glycogen accumulation in the intact rat.
74
3.1 THE CHARACTERISTICS AND CONTROL OF HEPATIC GLYCOGEN
SYNTHESIS IN THE NORMAL 48H- STARVED RAT.
In order to study the control of glycogen synthesis in the
perfused liver, a pre-requisite was the achievement of rates of net
glycogen synthesis, at rates similar to those observed in vivo. Such
rates had not previously been attained in perfusion experiments (e.g.,
see review by Exton et al., 1970). To facilitate this objective,
net glycogen synthesis was measured in perfused livers from starved
rats, so th6A proportional changes in glycogen would be greater (as the
initial glycogen content of livers is lower after starvation). In
this section, the characteristics and control of net glycogen accumulation
in the perfused liver are described.
3.1.1 The validation of sequential liver sampling
in the perfusion
In order to measure glycogen synthesis in the liver by a
sequential sampling procedure it was necessary that the glycogen contents
of different lobes of the liver could be validly related to each other,
i.e., that measurements in glycogen content in one area of the liver should
resemble those in other areas. Biopsies could not be removed in sequence
from within a single lobe, for technical reasons, and also because once
the first liver biopsy had been taken, the remaining part of that lobe
would have an interrupted blood circulation due to the ligature, and would
not represent what may have occurred had it been left intact.
When the major areas of the liver were sampled simultaneously
in intact rats the glycogen contents (5-601xmol of glucose g of fresh
liver) were usually within 10-20% of each other. This was also true
after a short period of perfusion (Table 3) but differences between
• Table 3. Glycogen content and synthesis in the major lobes of perfused liver from 48h- starved rats.
Livers were perfused with bicarbonate -albumin- saline plus fresh defibrinated whole rat blood, containing
glucose (30mM). After 15min, a single dose of pyruvate, serine and glycerol was added (each initially 5mM). Other
details are given in the text. Results are means + S.E.M. with the number of observations in parenthesis.
Glycogen content (tkmol glucose/g fresh liver)
Lobes sampled simult- Lobes sampled sequentially aneously after 20min during perfusion. perfusion.
Rate of glycogen synthesis calculated from sequential liver samples.
(p.mol glucose/min per g)
20 min. 50 min.
Median lobe : Median lobe : Leftlaterel lobe :
13.2 + 2.0 (3)
22.4 + 4.8 (11) 47.0 + 5.0 (11) 0.82
Left lateral lobe : Median lobe : Caudate lobes :
15.2 + 1.5 (3) 22.4 + 4.8 (11) 39.0 + 5.4 (11) 0.55
Combined naudate lobes : Left lateral lobe : Median lobe :
12.3 ± 3.7 (3) 14.6 3.5 (5) 28.3 ± 4.0 (5) 0.46
76 •
the glycogen content of the lobes was observed after longer perfusion.
The apparent rate of glycogen synthesis was greatest in the left lateral
lobe. Since the initial glycogen content of this lobe could be
reasonably estimated by sampling any other part of the liver, the
standard procedure in subsequent experiments was to sample the median
lobe intially, followed by the left lateral. The apparent differences
in the rates of synthesis between the lobes (Table 3) could be explained
by the handling of lobes during sampling. This would be especially true
when the left lateral was being sampled first, followed by the medians
as the latter lies on top of the former and would have to be deflected
towards the thorax in order to ligature the left lateral lobe. It should
be noted that these perfusions were done with whole defibrinated
rat blood and the rates of glycogen deposition were higher than those
obtained when washed rat red cells were used (cf Tables 3 & 4).
77
3.1.2 The role_of_glucose and gluconeogenic precursors
in glycogen deposition in the perfused liver.
The circulating precursors of hepatic glycogen have not been
fully identified (see Introduction Section 1.3). Since the level of
the enzyme glacokinase is reduced on starvation, glucose may not be, the
major carbon source for hepatic glycogen in the starved animal. Glycogen
accumulation was therefore studied in the presence of glucose and
gluconeogenic substrates and their importance as glycogen precursors
assessed.
When the livers from 48h- starved rats were perfused with 30mM
glucose alone, low rates of glycogen accumulation were obtained (Table
4) and glucose uptake of 0.6tkmol /min per g liver was seen. Higher
rates of glycogen synthesis were observed if gluconeogenic substrates
were added to the perfusion medium in addition to the glucose. A standard
mixture of pyruvate, serine and glycerol (each initially 5mM) was used
due to the known rapid rates of gluconeogenesis obtainable from such
substrate combinations (Ross et al., 1967). This mixture provided a
'substantial quantity of carbon atoms for glycogen synthesis without the
risk of inhibitory concentrations of any one substrate, If this mixture
was added 15min after the start of the perfusion, the rate of glycogen
deposition was 0.68)A-mol/Min per g in the presence of 30mM glucose. In
these perfusions no glucose uptake was seen (Table 4).
The glucose concentration in the perfusion medium was critical
for glycogen synthesis. When the concentration of glucose was varied
between 10mM and 40MM maximal rates of glycogen accumulation were seen '
only when the glucose. was 25-30mM (Fig. 5; Table 4). It can be seen
(Table 5) that by increasing the concentration of glucose added to the
medium, the increase in glucose concentration during perfusion became
Table 4. Glycogen synthesis and changes in medium glucose during perfusion of livers from starved rats
Livers were perfused with bicarbonate-albumin-saline containing washed rat erythrocytes and
glucose at various concentrations (initially 10-40MM). When present, a single dose of gluconeogenic substrates
(pyruvate, serine and glycerol; each 5mM) was added after 15min. Glucose in the medium was detected in
duplicate or triplicate. Other details are given in the text. Results are mean values + S.E.M. with the number
of observations in parenthesis.
Medium Glucose - (mM) Glycogen Content . • (pmol glucose/g fresh liver)
MIN: 20 35 50 20 50 (medium lobe) (left lateral
lobe).
Approx. initial Gluconeogenic concn. of glucose Substrates in medium
(Elm)
Rate of glycogen Synthesis
(1nmol glucose/ min per g)
10 11.0+1.1 13.3+1.1 13.7+0.6 7.1 8.3 0.04+0.02 (3)
20 18.1+1.0 19.3+1.0 20.5+0.9 12.1 18.1 0.20+0.06 (7)
30 27.9+0.6 28.9+1.9 28.5+0.6 21.5 42.1 0.68+0.05 (11)
40 38.2+0.8 41.6+1.0 42.8+1.1 13.6 21.5 0.27+0.05 (6)
30 31.6+1.8 30.2+0.1 28.7+1.7 29.8 -- 35.1 0.17+0.09 (5)
ou
a a a
a
79
a
Q0
ra
a
a
a
a
a a a
a
20 MEDIUM GLUCOSE (mM)
40
Fig. 5 Rates of glycogen synthesis at various glucose concentrations
in the perfused liver of starved rats.
Livers of starved rats were perfused with the standard medium,
containing 040mM glucose. After l5min,gluconeogenic substrates were
added (see Table 4). Glucose concentration was measured after 20min of
perfusion, and glycogen synthesis between 20 and 50min. Each point
represents a single perfusion.
Table 5. Calculated total synthesis of glucose during perfusion of liver from starved rats
The amounts of glucose and liver glycogen formed have been calculated from the results in Table 4 and Fig, 5.
Metabolic changes between. 20 and 50min perfusion
Approx. initial Gluconeogenic Change in Medium Glucose Release Glycogen Synthesis Total Glucose con= of glucose in Substrates Glucose (mM) (y.mol/g) ol of glucose/g Synthesised medium (mM) f liver) /30min per g
0 + 3.5 22.4 0 22.4
10 + 2.7 20.6 1.2 21.8
20 2.4 18.0 6.0 24.0
30 + 0.6 5.6 20.4 26.0
40 + 4.6 35.3 8.1 43.4
30 - -2.9 -23.1 5.1 -18.0
81
less extensive, and an increase in glycogen-glucose occurred. However,
the total glucose synthesised did not appreciably alter, and so it would
appear that the glucose was directing the products of gluconeogenesis to
glycogen rather than to glucose. In other experiments with 40thM glucose
plus gluconeogenic precursors (Table 5; Fig. 5) the total glucose
synthesised was greater than that at lower glucose concentrations, and
yet glycogen synthesis was submaximal. These experiments showed that
glucose in the perfusate did not inhibit gluconeogenesis (in the sense
of total glucose synthesis).
The time course of glycogen deposition was investigated under optimal
conditions with the standard liver sampling technique, i.e., median
followed by left lateral lobe. In one group of perfusions the samples
were taken at 5min and 35min, instead of the normal 20min and 50min.
The rate of glycogen synthesis during the first 35min was not as rapid
as that seen between 20 and 50min (Fig. 6),and was possibly due to the
adjustment of liver function after the operative procedure, and not a
true lag in glycogen synthesis.
Since there was no' net uptake of glucose by the liver when
gluconeogenic precursors were added to the perfusion, the net source
of carbon atoms for the accumulated glycogen must have been the
gluconeogenic substrates. Ratesof gluconeogenesis of 1.4 pmol/min per g
were observed in the absence of added glucose (Fig. 7) which showed. that
gluconeogenesis was sufficient to support glycogen deposition, and that
the total glucose synthesised in 30min was of the same order as that
during glycogen accumulation. The rate.fell off with time, probably due
to the decline in the concentration of the substrates.
When glycogen deposition occurred in optimal conditions, i.e.,
in the presence of gluconeogenic precursors and 30mM glucose, there could
have been formation of free glucose by gluconeogenesis and then uptake of
glucose to form' glycogen, in which case there would be no significant net
alteration in medium glucose. The extentof this process
50
40
82
10 20 30 40 50 TIME (MIN)
Fig. 6 Time course of glycogen synthesis in the perfused liver.
Livers of starved rats were perfused as described in Table 4;
the initial glucose concentration was 30mM. Liver samples were taken
at 20 and 50min (E3; 11 perfusions; gluconeogenic precursors added after
15min) or at 5 and 35min (o; 3 perfusions; gluconeogenic precursors added
after lmin). The bars represent the S.M.
83
10 20 30 40 50 TIME (MIN)
Fig. 7 Time course of glucose formation in the perfused liver.
Livers of starved rats were perfused as described in Table 4, except
that no glucose was added. Gluconeogenic substrates were added after 15min.
The larger part of the median lobe was removed after 20min , as in the
standard procedure for measuring glycogen synthesis; its glycogen content
was less than 5pmol of glucose/g. Results are the mean values ± S.E.M.
of 3 perfusions of livers whose average weight (after liver biopsy at 20min) was 5.5g.
84
was igyestigated using (U-14C) glucose. The incorporation of 14C
into liver glycogen (Table 6) showed that assuming that the specific
radioactivity of the medium glucose did not change appreciably, about
one-third of the total glycogen synthesised was derived from glucose
(which was replaced by gluconeogenesis since its concentration did not
fall).
Glycogen accumulation, at rates which occur in vivo, had
thus been observed in the perfusion when the medium contained 30mM
glucose and glycerol, serine and pyruvate (each 5mM).These precursors
were added to the perfusion with glucose in different combinations and
concentrations to see whether any one precursor was the major substrate
for glycogen formation. It would appear that no single substrate was suff-
icient to support significant glycogenesis (Table 7a). The addition of •
10mM substrate to the perfusion medium produced higher rates of synthesis
than when 5mM were used, except in the case of glycerol when the higher
concentration inhibited.Mhen two substrates were added to the perfusion
medium (with 30mM glucose) the rates of glycogen accumulation were
faster than in the presence of each substrate separately (Table 7b) and
the rates appeared to be additive. This also seemed to be the case when
three or more percuxsors were added to the perfusion (Table 7b). No
single substrate thus supported significant hepatic glycogen accumulation,
perhaps with the exception of 5mM glycerol or 10mM serine, alanine or
pyruvate. It seems likely that the net rate of synthesis obtained with
glycerol, serine, pyruvate and glucose,was due to the sum of rates from
each component separately.
85
Table 6.
Incorporation of 140 from (U-140) glucose (30mM) into
glycogen of perfused liver from starved rats.
Livers were perfused as described in Table 4, (30mM glacosb
and gluconeogenic substrates); (U-140) glucose was added to the medium
at 15min. Results are the mean values of three perfusions, in which
the average initial specific radioactivity of the glucose in the medium
was 6.36 x 103 d.p.m/pmol.
Medium 140-glucose (d..p.p/m1) 193,300
Glucose (d.p.m4amol) 6,360
Glycogen in liver l/g) 20min liver sample 37.0+12.0 50min liver sample 59.1+15.5
Net glycogen synthesis (pmol/g/min)
0.74
d.p.m/pmol glycogen 20min liver sample 62.3
50min liver sample 644.0
pmol 140-glucose incorporated into glycogen/g liver. 20min liver sample
003
50min liver sample 5.4
Rate of 140-glucose incorporated into liver glycogen ol/g/min)
0.17
Table 7a. Rates of net glycogen accumulation in perfusions with single substrates
Livers were perfused as described in the text. 30mM glucose was present in the medium and
substrates were added at 15min: any further additions were made at 25 and 35min. Continuous infusion was at
3-4ml/h. Results are means + S.E.M. with the number of observations in parenthesis.
Type of addition (mM) Rate of glycogen -accumulation (Jumol of glucose / g / min)
Glycerol Lactate Pyruvate Serine Alanine
Initial,5 0.42+0.12(5) - 0.09(2) 0.20+0.10(3)
Three additions, 5 0.48(2) 0.42(2) _
Initial 5, plus 0.15+0.06(5) 0.12+0.09(3) 0.02(1) infusion (conc. indicated)
(0.IM) (0.5M) (o.am)
Initial 5, plus two additions, 2.5
0.27+0.10(4) 0.43+0.08(4) 0.30+0.01(3) 0.49+0.16(5) 0.43+0.07(3)
Initial 10, plus infusion (1.0M)
0.23+0.15(3) 0.56(2)
Table 7b. Rates of net glycogen accumulation in -perfusions with substrate combinations
Details are as in Table 7a.
Group Substrate present (initial mM) Rate of net glycogen accumulation ol of glucose/g/min) Glycerol Lactate Pyruvate Serine
1. - - 5 5 0.32+0.32 . (3)
2. 5 - 5 - 0.43+0.08 (4)
5 5 0.65+0.05 (4) (infusion 0.511) (infusion 0.511)
4. 5 0.55+0.09 (4)
5 5 0.68+0.0 (11)
6. 5 5 5 0.52 (2)
88
3.1.3 The role of insulin and fatty acids in
hepatic glycogen metabolism
Due to the metabolic interest of fatty acids and insulin,
their influence on glycogen synthesis was investigated. In optimal
conditions sodium oleate (1mM initially) did not significantly affect
glycogen accumulation(Table 8) although there was a small decrease in
the rate of synthesis. Insulin also had no effect under these
conditions, although in suboptimal conditions (absence of gluconeogenic
precursors, or if glucose was less than 30mM), there was a moderate
stimulation of glycogen deposition ( cf Tables 8 & 4). This apparent
stimulation could have been due to the inhibition of glycogenolysis by
insulin, rather than an actual increase in the rate of synthesis.
Table 8. Influence of added sodium oleate and insulin on glycogen synthesis in the perfused
liver from starved rats.
lavers were perfused as described in Table 4. When present sodium oleate,was initially
1mM (after 15min of perfusion). Insulin was added at 10min and subsequently at 10min intervals, as 0.1m1 of a
1U/M1 solution. Other details are given in the text. Results are mean values + S.E.M., with the number of
measurements given in parentheses when these are different from the number of perfusions.
Gluconeogenic Other No. of Approx.. Glucose in medium (mM) substrates additions perfusions initial
conc. of glucose () 20min 35min 50min
Glycogen content Oamol of glucose /g fresh liver)
After 50mIn (left lateral lobe)
Rate of glycogen synthesis Oamol of glucose/Min per g of fresh liver) After
20min (median lobe)
3 20 22.2+0.6 21.4(2) 21.8+0.7 11.0 21.3 0.35+0.11
Insulin 5 20 19.1+0.9 19.0+0.9 18.7+1.0 22.7 31.3 0.28+0.11
Insulin 5 20 19.4+0.4 21.3+0.4(3) 21.6+0.7 22.3 39.7 0.58+0.14
Insulin 6 30 27.9+0.9 27.5+1.0 27.5+0.8 39.6 51.6 0.40+0.08
9 30 28.4+0.3(6) 30.1+0.5(4) 30.4+0.5 23.6 43.3 0.66+0.05
Oleate 5 30 29.0+0.4 30.6+0.3 30.2+0.5 26.6 43.0 0.54+0.05
Insulin' 5 30 26.1+0.7 27.4+0.4(4) 27.5+0.6 37.6 58.5 0.69+0.09
Insulin, oleate
3 30 30.2+1.6 29.6(2) 30.8(2) 14.7 29.7 0.50+0.10
0,
90
3.1.4 Characteristics of glycogen synthetase and
phosphorylase in the liver
As a pre-requisite to a study of the hormonal control
of the regulatory enzymes of glycogen metabolism, properties of these
' enzymes in homogenates were assessed.For this study, radiometric
assays of synthetase and phosphorylase were employed, as described in
the Methods chapter (Section 2.5.4).
The time courses of the glycogen synthetase
and phosphorylase assays
The time course of the enzyme assays was studied using
centrifuged and crude liver (fed rat) homagnates. The glycogen synthetase
assay mixture was as in Method 1. The centrifuged homogenates brought
about 14C incorporation into glycogen in a linear fashion (Fig. 8).
When the crude liver homogenate was assayed, the rate of reaction was
faster initially.
On calculation of the activity of the enzymes in txmol/min
per g, for each assay time, there appeared to be little difference in
% "a" form glycogen synthetase between 5, 10 and 15min although a fall
in activity of this form and total enzyme was observed (Table 9). The
% "a" form glycogen synthetase in the crude homogenate was however,
higher after 5min incubation than at 10 or 15min. There was a noticeable
fall in the activity of phosphorylase with time measured in the crude
liver homogenate (Table 9). A small decrease was also seen in the
centrifuged homogenate.
Due to the more consistent results, (albeit lower activity);
in the spun homogenate, the majority of subsequent assays were carried
out using this enzyme preparation.
91
For Fig..8 see over-leaf
15 5
10
ASSAY TIME MIN
I I
0 5 10 15 0 6 10
7000
A
PHOSPHORYLASE SYNTHETASE "a"
TOTAL
2400 SYNTHETASE
1600
800
1-:=1
n 5000
O
m 0
2 O
3000
0 0 m
1000
Table 9. The effect of different assay times on the activities of hepatic glycogen synthetase and
phosphorylase.
Assay details are in the text. Results are from a single fed rat liver, except for the glycogen
synthetase measured at 5 and 15min which are means of 2 animals. Assays were done in duplicate at 30oC.
Glycogen synthetase activity -method 1) pimol/min per g)
Assay time
Iran
5min. Total OA Half
lOmin. Total % "a"
Crude homogenate Centrifuged honte
0.31 0.10
0.69 0.21
45 48
0.26 0.11
0.81 0.21
33 50
"a"
0.18 0.08
15min. 15min. Total 0.53 0.16
Yo "a"
34 48
Glycogen
3min. 14.70
5.75 phosphor-lase activity
6min. 13.73
5.27
per g)
10min. 11.68
5.21
94
The effect of centrifugation on enzyme activity
Glycogen synthetase was assayed using the mixture defined as
Method 1 and incubation was for 15min for glycogen synthetase and 10min for
phosphorylase at 30°C, or 2min (both enzymes) at 37°C. When the assay was
carried out at 30°C, centrifugation or the homogenate caused loss of activity
of both glycogen synthetase and phosphorylase (Tables 9 and 10). The total
glycogen synthetase of fed rat liver however, decreased more, resulting in
an apparent increase in % "a" form glycogen synthetase. A similar fall in
activity was observed when the assay was done at 37°C (Table l0),although
there was not such a large difference between the spun and crude homogenate
(compared to that at 30°C) when the glycogen synthetase of the fed rat liver
was assayed.
It appears therefore that centrifugation of the liver (from a
.starved rat) homogenate halves the activity of glycogen synthetase and
phosphorylase when assayed at 30°C and 37°C. In the fed liver homogenate
the situation is not as clear, as spinning caused a greater loss of glycogen
activity when assayed at 30°C than at 37°C. This apparent loss of activity
due to centrifugation has been reported especially with respect to
phosphorylase,which sediments with the microsomal fraction in the fed state
but not in the starved (Maddaiah & Madsen, 1968; Tata, 1964). No such
correlation has been observed with glycogen synthetase, although a large
proportion of activity is associated with the microsomal fraction in the fed
state (Maddaiah & Madsen, 1968).
Table 10. The effect of variation in tem•erause and •unification on the activities of he•atic
glycogen synthetase and phosphorylase in the fed and starved rat
Assay details are in the text. The assay mixture for glycogen synthetase was that defined as
Method 1, and incubation was for 15min for glycogen synthetase and 10min for phosphorylase at 30°C, or for 2min
(both enzymes) at 37°C. Results are means of 3 rats + S.E.M.; mean percentages of synthetase "a" were
calculated from % "a" values in individual samples. Assays were done in duplicate.
Fed rat
48h- starved rat
Assay Temperature
30°C
37°C
Liver homogenate
Crude homogenate
Spun homogenate
Crude homogenate
Spun homogenate
Glycogen synthetase (12 mol/min per g)
Glycogen phosPhorylase. Ou mol/min per g)
14.61
+1.47
5.20
+0.93
25.73
+2.24
7.57 +0.91
Glycogen synthetase
mol/min per g)
Glycogen '.phosphorylase ()a mol/min per g)
9.20
+0.95
5.06
±0.94
16.42
+0.90
8.73
+1.04
"a"
0.16
+0.02
0.09
+0.01
o.56
+0.10
0.50
±0.04
Total
0.50
+0.04
0.15
+0.02
1.64
+0.13
1.05
+0.24
% "a"
32
±2
49 ±1
36
+7
51
i)
"a"
0.18
+0.04
0.08
+0.01
0.85
+0.09
0.47
+0.04
Total
0.39
+0.07
0.19
±0.02
1.35
+0.16
0.83
+0.19,
% "a"
45 ±3
43 ±1
63
±1
63
+13
96
The effect of assay temperature on enzyme activity
When crude liver homogenates were assayed at 37°C a 3-4 fold
increase in activity was observed (Table 10) for glycogen synthetase
and a 2 fold increase for phosphorylase,compared to values at 30°C.
The glycogen synthetase activity of the spun homogenate at 37°C was
increased 6-7 fold, and phosphorylase 2 fold.The assay temperature ,the
therefore had a greater effect on/spun homogenate than on the crude
homogenate. This response could imply that the crude extract activity
approximates the ttruel activity of the enzyme and thus increasing the
assay temperature has a less dramatic effect.
97
3.1.5 Control of glycogen synthesis in the perfused
liver of normal starved rats.
In order to assess the role of hormones in the control of
the regulatory enzymes of glycogen metabolism, the properties of these
enzymes were studied in the perfusion. Glycogen synthetase and
phosphorylase were measured in the perfused liver of normal starved rats
under optimal and sub-optimal glycogen accumulation conditions and any
correlation evaluated.
In these conditions, the hepatic concentrations of UDPG and
glucose 6-phosphate were measured (Table 11); the latter value was of
the same order as previously reported for perfused liver of starved
rats (Ross et al., 1967). When maximal rates of glycogenesis were
observed, the concentration of MPG was decreased, compared to that when
lower rates of glycogen accumulation were found (Table 11), suggesting
that the glycogen synthetase reaction was accelerated. This inference
was confirmed, since under these conditions an increased proportion
(60-80) of glycogen synthetase was present in the "a" form (an active
form in vivo). This was true despite the lack of a pattern in total
synthetase activity or actual activity of the "a" form, and suggests
that the % of synthetase in the "a" form may relect its capacity to
bring about net glycogen deposition.There was decreased activity of
glycogen phosphorylase in perfusions with glucose (Table 11) but this
was not sufficient in itself for high rates of glycogen accumulation
to be obtained, unlike the situation when C3-substrates or fructose
were added to perfusate in addition to the glucose. In these latter
perfusions glycogen phosphorylase was decreased in parallel with the
increase in % glycogen synthetase "a".
Additions to perfusion medium
No. of Enzyme assays (amolAin/g of perfusions Glycogen synthetase
(Method 1)
Han Total
% Hatt
None 3
0.14 + 0.01 0.38 + 0.04 38 + 3
Glucose 3 0.18 + 0.03 0.37 + 0.05 48 + 3
03-substrates 4
0.17 + 0.02 0.42 + 0.11 46 + 8
Glucose plus 6
0.20 + 0.01 0.34 + 0.02 61 + 3 03-substrates
Glucose plus 0.16 + 0.02 0.21 + 0.03 80 ± 4 fructose
Glucose plus* 4 0.17 + 0.06 0.20 + 0.05 83 + 7 03-substrates
* Supplemented medium, including insulin, hydrocortisone and amino acids.
t Net rate of glaconeogenesis 1.22 + 0.10 (5)
fresh liver) Glycogen phosphorylase
(centrifuged homogenate)
5.6 ± 1.0
3.6 + 0.3
7.3 ± 1.1
3.4 ± 0.5
1.4 + 0.4
2.3 + 0.5
Table 11. Concentrations of enz es and •athw intermediates durin 1 co:-n s thesis in the .erfused
liver of starved rats.
Livers from 48h-starved rats were perfused with bicarbonate-albmin-saline containing washed rat erythrocytes. Substrates where present were added after 15min. Glucose when added was initially 28mM, and the 07- substrates, lactate, glycerol and pyruvate, were initially 5mM, 3.3mM and 1.7mM (respectively) and then ilifused (3m1/h) in a mixture containing 0.5M-sodium lactate, 0.33M-glycerol and 0.17M-sodium pyruvate. Fructose was initially 5mM, and then infused (0.5M, 3m1/11). Perfusion was for 50min except those with glucose alone : 60min. Enzymes, glycogen and metabolites were assayed in frozen samples of the left lateral lobe. An initial sample (median lobe) was removed, after 20min to measure glycogen. Results are means + S.E.M of the no. of observations indicated; mean percentages of synthetase "a" were calculated from % "a" values in individual samples.
•
Metabolite content Rate of net (nmol/g) glycogen accumulation
UDP- Glucose 6- (p010 1 glucose/min/g) .glucose phosphate
28 ± 7(3) 20(3) 0
96 ± 12(4) 48 ± 8(4) 0.21 + 0.08
89 ± 13(5) 51 ± 6(5) 0
52 ± 11(10) 71 ± 11(10 0.75 + 0.10
0.84 + 0.07
0.54 + 0.19
99
3.1.6 Hepatic glycogen accumulation in the intact
rat.
In order to assess the physiological significance of the
properties of hepatic glycogen accumulation observed in the perfusion
(Section 3.1.2), glycogen deposition was measured in the intact
anaesthetised 48h- starved rat, by infusion of substrates via a tail
vein. The loft lateral lobe was removed first, followed by the median lobe,
as this was technically the easier order of sampling. Separate experiments
(not shown) revealed that in vivo (unlike perfusion), either order of
sampling gave the same rate of net synthesis. Unlike the perfusion
studies, glycogen accumulation was obtained when glucose alone was
infused. However, this result is not incompatible with those obtained
by perfusion as gluconeogenic percursors would have been released by
other tissues (e.g., muscle) in the intact rat. The rate of glycogen
deposition was dependent upon the final blood glucose concentration
(Fig. 9) in a similar way to that observed in perfusion experiments. As
might be expected the time lag observed in glycogen synthesis in the
perfused liver was not as pronounced when liver biopsies were measured
at various. times during in vivo infusions (Fig.10).
100
0.8
0 0
0
.E 0.6 E
0 E
a, _c C
0.4 — c o.) 0 t.) a) 0
0
0.2--
0
0 0
1 I I I 20 40
Blood glucose (mM
Fig. 9 Rates of glycogen synthesis in vivo, at various blood
glucose concentrations.
Glucose was infused intravenously, from 5 to 90min
after anaesthesia. Liver samples were taken after 30 and 90min, and
blood after 90min. Other details are given in the text. Each point
represents a single experiment.
. 0
0 0
0 0 0
0 0
0
60
40 w - N Z 0 u - 7
0)
LIJ g- o 0 •
O -6
>L2" X20 0
101
1
0 20 40 60 80 100
TIME(MIN)
Fig. 10 Time course of clycotIen synthesis in viva.
Glucose was infused intravenously into anaesthetised rats.
In three groups of experiments, each with six rats, and each denoted
by a different symbol, two sequential liver samples were taken : 0,
at 5 and 35min after the start of the infusion, or A , after 5 and
65min, or 0 after 30 and 90min. The blood glucose concentration
at the time of the second sample was 23-36mM.
102
302. HEPATIC CARBOHYDRATE AND FAT METABOLISM IN
THE "FED" RAT
1. Glucose metabolism in the perfused liver of fed rats.
2. Glycogen synthesis in the perfused liver of overnight-starved rats.
3. Fatty acid synthesis in the perfused liver of overnight-starved rats.
103
3.2 HEPATIC CARBOHYDRATE AND FAT METABOLISM IN THE "FED"RAT
The role of glucose and gluconeogenic precursors in
hepatic glycogenesis in the 48h- starved rat has been studied (see
previous Section). The contribution of glucose to the net rate of
glycogen accumulation in the perfused liver was small,being zero in
terms of the net rates (as there was no glucose uptake); unidirectional
uptake was about one-third of the total rate. Since the level of the
enzyme glucokinase is reduced on starvation (Salas et al., 1963; Walker
& Rao, 1964), it was of interest to study the role of glucose and
gluconeogenic substrates in fed rats, where glucokinase and glucose
uptake could have a greater role.
3.2.1 Glucose metabolism in the perfused liver of
fed rats.
In fed rats, hepatic glucose uptake was faster than
in 48h- starved rats (Fig. 11). The existence of net gluconeogenesis
was demonstrated in livers from fed rats perfused with a mixture of
glycerol and pyruvate, plus lactate (maintained by infusion); this mixture
brought about net glucose output, in the absence of a decline in glycogen
(Fig. 11). If lactate was replaced by serine, net glucose uptake was
also reversed by this mixture of gluconeogenic precursors. The rate of
net glucose output in the experiments with the lactate - containing
mixture, calculated from Fig. 11 (disregarding the small amount of
net glycogen synthesis) was about 0.5p.mol glucose/Min /g liver, or
about 4jAmol/Min/iiver.
104
40
E
(I) 0 U _I 20
0
500
O
0 300
O
o
20 60 100 TIME (MIN )
Fig. 11 Glucose metabolism in the perfused liver of fed rats
Livers were perfused with glucose (initially 30mM) as described
in the text. The following additions were made, at the times indicated:
mixture of glycerol and pyruvate plus lactate (40min: see Table 11: A )
or serine (30min : see Table 41V ). In the experiments with the lactate
- containing mixture, glycogen in sequential biopsies was measured,
( G3 , broken line -- -). In two groups of perfusions, no additions were
made, in fed (o) and, for comparison, 48h- starved rats (0). Results
are from three perfusions.
105
3.2.2 Glycogen synthesis in the perfused liver
of overnight-starved rats.
A common procedure is to starve animals (or people)
overnight, in an attempt to stabilise the metabolic events of the "fed"
state. This depletes liver glycogen which permits the study of net
glycogen accumulation. Rates of glycogen synthesis were measured in
perfused livers from such rats, in conditions shown to be optimal
for net glycogen deposition in 48h- starved rats, and the role of
glucose and gluconeogenic substrates assessed.
In perfusions with glucose (NUM, no other substrates),
the rate was higher than in 48h- starved rats (Table 12). The rate of
glycogen accumulation was lowest in perfusions with glycerol-containing
uixture, and was unchanged by insulin (Table 12). This apparent
inhibitory action of glycerol was not due to diminution of the rate of
gluconeogenesis, the rate of which was 1.05+0.121(molAtin/g of fresh
liver (mean + S.E.M. of three perfusions with the glycerol mixture
measured in the absence of added glucose) i.e., more than three times
faster than the-rate of glycogen synthesis (0.281A-mol/min/g : Table 12).
In livers from 22h- starved rats, perfused at 14.00h with glucose plus
lactate, glycerol and pyruvate, the rate of glycogen accumulation was
higher than in 18h- starved rats (Table 12).
The above rates of glycogen synthesis, especially
in the presence of the glycerol-containing mixture, suggested that there
could be an inherent impairment in the capacity for net glycogen synthesis
in overnight-starved rats. However, experiments with anaesthetised intact
animals showed that this was not the situation. Net glycogen
accumulation, measured in sequential liVer biopsies in vivo, during glucose
infusions was 0.77+0.12 (4)tx.mol glucose/g/min at 30mM glucose, which is
Table 12. G1 co en synthesis in the perfused liver of overnight-starved rved rats
Livers were perfused as described in the text at 11.00h unless indicated. Substrates where
present were added after 15min. Glucose was initially 28rE and the gluconeogenic precursors, L(lactate), G(glycerol)
and P(pyruvate) were initially 5mM, 3.3mM and 1.7mM (respectively) and the infused (3ml/h) in a mixture containing
0.5M-sodium lactate, 0.33M-glycerol and 0.17M-sodium pyruvate. Mien the L and P mixture was added, lactate was
8mM and pyruvate 2mM and theywere infused (3m1/h) in a mixture containing 0.8M-sodium lactate and 0.2M-sodium
pyruvate. Fructose was initially 5mM and then infused (0.5M, 3m1/0. Insulin was added (250mU) every 15min.
Results are means +
Period of Additions to No. of Glycogen (,pmol of_glucose/g) Perfusate glucose Calculated rate starvation perfusate perfusions 20 min 50 min (mM) of net glycogen (h) 20 min 50 min accumulation
(median (left lateral ' cumol of glucose
lobe) lobe) /g/min)
18 None 8 6.3 17.6 26.3 23.7 0.38 + 0.09 18 L, G, P 6 2.0 10.6 27.1 29.6 0.28 + 0.05
18 L, G, P plus insulin
2 0 7.7 28.6 27.6 0.26
18 L, P 3 6.8 22.0 24.1 25.3 0.51 + 0.03
22* L, G, P 4 2.0 18.7 27.8 29.8 0.56 + 0.02
18 Fructose 3 5.3 27.0 27.7 32.1 0.73 4. 0.10
48 None 5 29.8 35.1 31.6 28.7 0.17 + 0.09
48 L, G, P 12 23.0 44.6 30.3 - 0.72 + 0.05
* perfused at 14.00h L H O
107
similar to the rate (0.6 - 0.7) in 48h- starved rats (see Section 3.1.6).
Also, if overnight-starved rats livers were perfused with glucose plus
fructose, maximal rates of net glycogen accumulation were observed
(Table 12), again showing that there is no inherent impairment in
glycogen synthesis capacity after starvation.
3.2.3 Fatty acid synthesis in the uerfused liver
of overnight-starved rats
The total rate of fatty acid synthesis was measured with
31120 (Salmon et al., 1974) in livers from overnight-starved rats,
perfused with glucose (30MM, plus the mixture of lactate, glycerol and
pyruvate). The rate was 0.5+0.3 (3)14.mol of long-chain fatty acid/g,
ywhich contrasts with rates of 3 - 5in fed rat livers in comparable conditions (Kirk, et al., 1975). These low rates may be explained by the
decline in hepatic glycogen, which serves as a source of acetyl residues,
and also exerts a stimulatory role in lipogenesis (Salmon et al., 1974).
108
3.3 DIE ACTIONS OF THE HORMONES OF THE
POSTERIOR PITUITARY GLAND ON HEPATIC
GLYCOGEN METABOLISM
1. Stimulation of hepatic glycogen breakdown
by (8 -arginine)-vasopressin and oxytocin.
2. The role of vasopressin in hepatic glucose
metabolism.
The stimulation of hepatic gluconeogenesis
by (8-arginine)-vasopressin.
The action of vasopressin and oxytocin on
glycogen synthesis in the perfused liver
and the intact rat.
5. The effect of vasopressin and other
glycogenolytic hormones on hepatic glycogen
synthetase and phosphorylase in vivo.
109
3.3 THE ACTIONS OF THE HORMONES OF THE POSTERIOR
PITUITARY GLAND ON HEPATIC GLYCOGEN METABOLISM
Antidiuretic hormone (vasopressin) is involved in
the processes of water balance and causes reabsorption of water in
the kidney in order to maintain a constant blood volume, compositiOn
and pressure. Glycogen is stored in the liver in association with
large quantities of water (Fenn, 1939) and since vasopressin blood
levels are increased during shock (e.g., haemorrhagic stress) .44-. was
of interest to evaluate the role of vasopressin, in hepatic
carbohydrate metabolism (see Introduction, Section 1.4.3). Also, since the
role of the posterior pituitary gland in metabolic events is not fully
clear, oxytocin action on the liver was studied.
3.3.1 Stimulation of hepatic glycogen breakdown
by (8 -arginine)-vasopressin and oxytocin
It has been reported (see Introduction, Section 1.4.3)
that vasopressin and oxytocin cause hyperglycaemia by stimulating the
breakdown of hepatic glycogen. It was of interest to study this
action in the perfused liver.
During perfusion of livers from fed rats the glucose
concentration in the medium rose initially from 5mM to 8mM and then
remained steady. If (8 -arginine)-vasopressin was then added in a
single dose to the medium there was a marked rapid efflux of glucose,
which fell off after about 40min (Fig. 12). This decline in glucose
efflux was not caused by lack of glycogen, as shown by glycogen
measurements at the end of the perfusions. The extent of the change in
glucose concentration was dose dependent, over the range 50 -609units/m1
of vasopressin (Fig.13), saturation occurring at 6001.kunits/ml.
110
20 —
1 40
80
120 TIME (MIN)
Fig. 12 Influence of va o ressin on the time course of
glucose output in the perfused liver of fed rats.
Livers were perfused as described in the text. The following
additions were made to the perfusion medium after 40min : vasopressin
at the following initial concentrations (p. units/ml) 1,000,0; 130,A ;300,0;
5,0 , or 0.7m1. of an extract of rat neuro-hypophysis 0 . Results
are from single perfusions.
0.5 VASOPRESSIN (mU/mI
1.0
Fig. 13 Dependence of the stimulation of hepatic glucose
output on vaso ressin concentration.
Livers from fed rats were perfused as described in the text.
After 40min, vasopressin was added, at various initial concentrations.
Each point represents the increment in glucose output during the next
40min , determined for each perfusion.
112
The most likely explanation for the rise in glucose concentration
in the medium was that breakdown of liver glycogen was being
stimulated by vasopressin. This was confirmed by measuring changes
in the hepatic content of glycogen during continuous infusion of
(8 -arginine)-vasopressin (Fig. 14). The rate of glycogenolysis was
calculated from the change in glycogen content of the two sequential
liver samples taken in each perfusion, presuming that no. large
differences in glycogen content exist between the major lobes. The
mean decrease in glycogen content in 60min. . was 49-1-14 (5)/A.mol of
glycogen glucose / g in control perfusions and 121±51 (4) in the
presence of vasopressin. The glycogen breakdown due to vasopressin
(about 7014mol of glucose / g) thus amounted to about 409kmol of glucose
(the average liver weight being 5.6g), which was sufficient to account
for the extra glucose (1501.4mol) which appeared in the medium. There
was no major effect of vasopressin on the lactate concentration in the about 40min
medium. The plateau in glucose efflux observed/after a single dose of
hormone (Fig.12) was still observed during (8 -arginine)-vasopressin
infusion, indicating that a single dose of hormone is sufficient for
a response to be observed, or that high concentrations of glucose inhibit
the vasopressin effect.
Apart from the fact that the posterior pituitary gland
secretes oxytocin as well as vasopressin, they are very similar
polypep-bides in structure and so it was of interest to replace vasopressin
by oxytocin in the above described system. A dose of 5000runits / ml
(10 times the saturating vasopressin dose) was added to the perfusion
medium; glucose efflux was negligible, but was observed at higher
concentrations (Fig.15).
LIV
ER
GLY
COG
EN
400
rn
a) 0
0) 200
'46
0 E
0
12
E E10 D
E
L.1.1 8
••••••••
e■I
0 4
tr) 0
2 -
A 4 ------A
A A
I
1
° - - 0 ----CI 0 • - - - - - • 0
0
113
0 20
40 60 80 . 100
TIME ( MIN)
Fig. 14 Effect of vasopressin on glycogen Content of
the nerfused liver.
Livers from fed rats were perfused as described in the text. After
40min vasopressin was added to an initial concentration of 700 punits /ml, and then infused at 200m units/h. An initial liver sample was
removed at 41min. The average liver weight for the remainder of '
the perfusion was 5.6g. Results are from three control perfusions
(open symbols), and four with vasopressin (filled-in symbols) : (0,0)
glucose; (A, A ) lactate; (0, 0 ) glycogen.
114
18
fc' --0- 9--1-1° A 0-1:1 A,4,/ ;
p,,N"- 0 ,/0 ■ --0 0 —0 0
0
0 20 40 60 80 100 TIME (MIN)
Fig. 15 Influence of oxytocin on the time course of
_lucose out ut in the erfused liver of fed rats.
Livers were perfused as described in the text. The
following additions of oxytocin were made to the perfusion medium
after 40min,at initial concentrations of lm units/m1(0);50m units
/ml (0) ; 100 m units/ml ( ); 500 m units/m1(A) . Results are
from single perfusions.
115
The commercially prepared vasopressin was checked
by a number of procedures. The amino acid composition after hydrolysis
in 6.14 liC1 was that expected of vasopressin thus excluding significant
contamination with additional peptide material. The glycogenolytic
effect of the preparation was largely destroyed by incubation in
10mM-thioglycollate (90min, 37°C, vasopressin 100m units Al), or by
combination of freezing and thawing, and standing at room temperature
for 20h. These results were compatible with the glycogenolytic agent
being vasopressin, which is known to be inactivated by these procedures.
An extract of rat neurohypophysis was prepared (in 0.5m1 of 0.1 M-HCl,
homogenised, centrifuged, and the supernatant neutralised) and on
addition to the perfusion caused glucose output (Fig.12) probably due
to vasopressin)since oxytocin (5m unit/m1) has a much less potent effect
on glucose efflux (Fig. 15).
No detectable alteration in the rate of flow of perfusion
mediumwas observed at any vasopressin concentration tested i.e.,
5 -10001kunitshal.
3.3.2 The role of vasopressin in hepatic glucose
metabolism
When fed rat livers were perfUsed with 30mM glucose,
glucose uptake was observed which was reversed by gluconeogenic
precursors (Section 3.2.1). In the presence of high glucose
concentrations (30mM) net glycogen accumulation is maximal and net
breakdown minimal (Glinsmann, et al., 196A and so this state might thus
be expected to influence the action of catabolic hormones. Such a
possibility was investigated in perfusions containing vasopressin
(Fig.16), which had a potent glycogenolytic action at autoregulatory
(9-12mM) glucose concentrations (Section 3.3.1). At 30mM glucose,
glucose output due to the hormone was negligible (Fig 16), although net
116
40
ID 1:3 El
---,... :1., 0 0 DI
E NN. 0 El ■-• w 0 CI U---....„.......0 D (-51 20`—
D (3 ul
I I I I I 1 40 80 120
TIME (MIN
Fig. 16 Effect of vaso ressin on lucose metabolism in
the perfused liver of fed rats.
Livers were perfused with glucose (initially
30mM) as described in the text. Between 40 and 60min 3 additions
of vasopressin, 110/m1, were made (0); control perfusions (❑ ).
Results are means of two perfusions with vasopressin, and three
controls.
0
117
glucose uptake was halted. A similar inhibition by glucose was observed
during infusion of the hormone (see previous section) when a decline in
glucose efflux was observed after 30mi .
3.3.3 The stimulation of hepatic gluconeogenesis by
(8-arginine)-vasopressin.
The observed stimulation of glycogen breakdown by vasopressin
raised the possibility that the hormone may act in a similar way and on
the same processes as•glucagon. The action of vasopressin on
gluconeogenesis was thus investigated.
The hormone was added in five doses at regular intervals due
to its rapid destruction by the liver (Little et al., 1966). The
concentration of (8-arginine)-vasopressin in the perfusion medium, measured
at the end of the perfusions, was not higher than that produced by each single
addition (M.L. Forsling, unpublished work), hence it was reasonable to
consider the hormone concentrations during the perfusion to be equivalent
to that produced by the initial addition. Gluconeogenesis from an infused
mixture of lactate, glycerol and pyruvate was stimulated by vasopressin
over a concentration range 30-150)u units/ml. (Fig. 17).
A similar stimulation (from 1.61+0.10 (6) to 2.46+0.09 (3)
min/g) mes.observed when two doses of the hormone were added after a
control rate of gluconeogenesis had been obtained with the gluconeogenic
mixture (Fig. 18). The effect of vasopressin on gluconeogenesis in the
absence of added substrate was also tested; no major stimulation occurred
(Fig. 18). Oxytocin (1000).Lunits/m1) had no effect on the rate of
gluconeogenesis from the glucogenic mixture.
O cs)
•■■
0 0
0
O
inc r
ease
0.5
0
2.5
GLY
CO
GE N
2.0
118
50 100 150
280
VASOPRESSIN(,ill ml
Fig. 17 Influence of vasopressin on gluconeogenesis or net glycogen accumulation in the perfused liver of starved rats.
Livers were perfused with gluconeogenic precursors (glycerol, lactate and pyruvate) as described in Table 12. Vasopressin was added at 10min intervals, from 10min after the start of the perfusion, to the concentration shown (abscissa). Two groups. of experiments were carried out: (a) 0 measurement of gluconeogenesis; rates were measured between 20 and 60min. Half the median lobe of the liver was removed at 20min (as in the standard procedure when measuring glycogen synthesis) and the glycogen content was negligible. Each point represents a single perfusion; (b) , measurement of net glycogen synthesis; perfusions contained 30mM glucose and gluconeogenic precursors, and results are means of three or four measurements (bars indicate S.E.M.).
119
T0 1 14
12
40 80 120 TIME (MIN)
Effect of va
(endo enous and from added substrate in the •erfused
liirer of starved rats.
Livers were perfused as described in the text and Fig. 17. Gluconeogenic substrates, glycerol, lactate and pyruvate were added at 20min and then infused (0, 0). Half the median lobe of the liver was removed at 20min : the remaining liver being 3.77 and 3.67g respectively.(8-Arginine)-vasopressin (570,u units/ml) was added at 60 and 75min (o). In one group of experiments (A) no substrates were added and vasopressin (1,000ju units/ml) was added at 40 and 70min; in these perfusions no liver was removed and the mean liver weight was 4.73g. Results are means of three perfusions except for the control period with substrates before vasopressin addition(6) and bars indicate S.E.M.
0
Fig. 18
120
3.3.4 The action of vasopressin and oxytocin on glycogen
synthesis in the perfused liver and the intact rat
It has been shown that net rates of glycogen accumulation in
the perfused liver require the presence of gluconeogenic percursors
as well as glucose (see Section 3.1.2). Yet (8 -arginine)-vasopressin
stimulates both gluconeogenesis (in the starved animal : results above)
and glycogenolysis (in the fed animal : results above). It was therefore
of interest to study the action of the hormone on net glycogen
accumulation in the perfused liver from a starved rat.
Glycogen accumulation was inhibited by (8-arginine)-vasopressin
in the perfused liver (Fig. 17) and the dependence of this effect on
the hormone concentration, was similar to that observed when
gluconeogenesis was studied. In these experiments there was no clear-
cut effect of vasopressin on the total glucose synthesised (i.e., the
sum of the change in liver glycogen plus blood glucose). (8 -Lysine)-
vasopressin also inhibited glycogen accumulation (Fig. 19),the dose
required for half-maximal synthesis being 20p units/ml compared to a
dose of 5011 units/M1 for (8-arginine)-vasopressin. This apparent
difference in sensitivity is diminished when the concentrations of the
hormones are considered on a weight basis (Fig. 19), inhibition occurring
over the range 60-300 pg/ml (vasopressin - like activities of the pure
hormones were taken as 400 U/mg and 250 U/mg, respectively; Boissonnas
et al., 1961).
Under the same conditions as described above, oxytocin inhibited
glycogen accumulation (Fig. 19)
(arginine or lysine forms), the
half-maximal about 25 m units/M
but at much higher doses than vasopressin being 400
saturating dosesAbout/M units/M1 and the
. Glycogen synthesis is thus inhibited
121
103 105 HORMONE IN MEDIUM (log j ml
10'
GLY
CO
GEN
A
Fig. 19 Effect oiLLEi me,son and oxytocin on glycogen
synthesis in the perfused livers of starved rats.
Livers were perfused with 30mM glucose and a gluconeogenic mixture of glycerol, lactate and pyruvate as previously described; glycogen synthesis rates were measured between 20 and 50min. (8 -lysine) -vasopressin (10 and oxytocin (0) were added at 10min intervals from 10min after the start of the perfusion, to the concentration shown. For comparison, the dependence of the effect of (8 -arginine) -vasopressin is included : broken line. Results are from individual perfusions except for controls (0), ' 14 observations.
122
over the range 5 - 400 m units/M1 or 10 - 900 neml (pure hormone has
a reported activity of 450 U/mg ; see Boissonnas et al, 1961).
The results with (8-arginine)-vasopressin in the fed and
starved animals indicate that the starved animal maybe more sensitive
to the hormone.
In order to confirm that the above observed actions of (8 -
arginine)-vasopressin occur in vivo the action of the hormone was
investigated in the intact rat. Although the ideal experiment would
have been to study its effect on fed liver glycogen, the glycogen levels
found in fed animals are variable and it is thus difficult to follow
changes in the intact animal. It was therefore, decided to use the 48h-
starved rat . Vasopressin and glucose were infused into the intact
anaesthetised rat and liver biopsies removed. Inhibition of glycogen
synthesis was observed (Table 13), confirming perfusion results.
3.3.5 The effect• of vasopressin and other glycogenolytic
hormones on hepatic .f..lycoFen synthetase and
12.. 1°18 17.--ivc) •
Vasopressin has been observed to cause hepatic glycogenolysis
at a minimal effective concentration. of the same order as that necessary
for glucagon stimulation of glycogen breakdown. Since glucagon
stimulates glycogen phosphorylase under these conditions, the effects
of vasopressin and other glycogenolytic hormones on glycogen synthetase
and phosphorylase were evaluated.
Since glucagon is known to have a rapid, direct hepatic action
the response of the enzymes of glycogen metabolism to the hormones were the
assessed by injection into/hepatic portal vein of an intact animal (see
Methods, Section 2.4.2). Vasopressin caused a small but significant
increase in phosphorylase (Fig. 20) although the effect was very short-
Table 13. Effect of 8-ar -vaso•ressin. on he•atic cozen s thesis in vivo
Glucose (1.5M) was infused intravenously (3m1/h) into anaesthetised 48h-starved rats and
two liver samples were removed sequentially from each rat. At the time of the second liver sample, the
average blood glucose concentration was 33mM. Result are the mean of six experiments, ± S.E.M.
Additions to infusion fluid Glycogen content Oz mol of glucose / g) Calculated rate of glycogen synthesis ( u mol of glucose
/g/min) After 10min. After 70min.
None 35 72 0.62+0.09
Vasopressin 32
44
0.19+0.11 (175m units/ml)
124
I I I I 2
TIME NM)
Fig .20. - The time course of has hor lase activation by
Llycogenolytic hormones.
Experimental details are as in the text. Hormones (giucagon
1.0ug,n; adrenalin, 1.5 x 10 -8 mol, 0 ; (8-arginine)-vasopressin,
10 m units, A or 100 m units, A ) were injected into the hepatic
portal vein and the liver removed after various times, control,(0).
Phesphorylase was measured in crude liver homogenates. Results are
mean values + S.E.M. (indicated by bars) of at least 3 experiments,
except for 100 m units vasopressin at 3 and 5 min : one observation each,
and 10 m units vasopressin at 0.5min (mean of 2 values : 11.65 and 11.95).
125
tern, reaching a peak after 30sec. A similar elevation was observed with
adrenalin. The glucagon effect was slower than the other two hormones
but much more prolonged and the actual increase in phosphorylase was much
greater.
None of the hormones tested had a significant effect on glycogen
synthetase (Table 14), considered either as actual activity of the Tian
form of the enzyme or as the percentage of the enzyme in the "a" form.
126
Table 14. Effect of 1,co enol tic hormones on the activities of
glycogen synthetase
Experimental details are as in the texband Jig. 19.
Glycogen synthetase was measured essentially as Method 1, described in
the Methods Section (2.5.4), except for the assay being carried out in
the crude liver homogenate. Results are means ± S.E.Ms with the number
of observations in parentheses. Mean percentages of synthetase "a" were
calculated from ro "a" values in individual samples.
Injection Time after injection (min)
n a_ G1 co thetase
"a" Total "a"
None 0 6 0.15 + 0.02 0.49 + 0.04 31 ± 3
nazi 0.5 3 0.20 ± 0.04 0.58 + 0.04 35 + 4
1 3 0.18 + 0.01 0.63 + 0.02 28+ 1
3 3 0.16 + 0.02 0.40 + 0.02 40 + 5
5 4 0.13 + 0.03 0.35 4- 0.01 36 ± 8
(8-arginine) vasopressin
0.5 2 0.18 0.47 38
10 units 1 3 0.14 + 0.01 0.54 + 0.02 27 + 1
3 4 0.11 + 0.02 0.35 + 0.05 30 + 5
5 4 0.16 ± 0.01 0.40 + 0.04 40 ± 4
(8-arginine)- vasopressin
0.5 4 0.18 + 0.01 0.56 + 0.04 32 ± 3
100 units 1 3 0.16 + 0.05 0.42 + 0.04 37 ± 5
3 1 0.12 0.52 '23
5 1 0.16 0.60 26
Glucagon 1 3 0.16 4. 0.01 0.49 4- 0.06 34 + 3 1.0 lug
3 3 0.24 + 0.02 0.50 + 0.09 49 + 5
5 3 0.16 + 0.01 0.39 + 0.04 43 + 4
10 1 0.14 0.36 38
Adrenalin 0.5 4 0.13 4 0.01 0.41 ± 0.05 32 + 4 8mol 1.5x10 _
1 3 0.14 + 0.01 0.30 4- 0.04 47 + 3
2 4 0.15 + 0.02 0.38 + 0.06 43 + 4
3 3 0.14 + 0.01 0.31 4. 0.02 46 + 1
5 4 0.13 + 0.02 0.33 + 0.08 44 + 6
127
3.4 HEPATIC GLYCOGEN METABOLISM IN THE STARVED
STREPTOZOTOCIN-DIABETIC RAT
1. Glycogen accumulation in the. perfused liver
from diabetic rats.
Glycogen synthetase and phosphorylase
activities in vivo and in the perfused liver
of diabetic rats.
3. Influence of glucose and fructose on the
activities of glycogen synthetase and
phosphorylase in vivo.
Hepatic glycogen accumulation in the intact
diabetic rat.
128
3.4 HEPATIC GLYCOGEN METABOLISM IN THE STARVED STREPTOZOTOCIN
-DIABETIC RAT
As decribed above, net glycogen accumulation has been observed
in the perfused liver from starved rats. Under optimal conditions,
insulin added in vitro had no effect but in suboptimal conditions (i.e.,
glucose less than 30mM or in the absence of gluConeogenic precursors)
a moderate stimulation .of synthesis was observed (see Section 3.1.3),
The effect of insulin in hepatic synthesis warranted further study as,
for a variety of reasons the role of insulin in hepatic carbohydrate
metabolism is uncertain (see Introduction, Section 1.4.1). Therefore,
since conditions had. been established forst-udyingnet glycogen accumulation,
glycogen synthesis and the role of insulin in the process were studied
inn-the streptozotocin-diabetic animal.
3.4.1 Glycogen accumulation in the perfused liver from
diabetic rats
In the livers of starved-diabetic rats, there was a marked
decrease in rates of net glycogen accumulation compared to rates in
starved normal rats (Fig. 21; Compare Tables 11 and 15),'in which a linear
time course of glycogen deposition was maintained for at least 80min
(Fig. 21). This defect in glycogen synthesis in the livers of diabetic
rats, observed in perfusions with glucose plus either fructose or a
mixture of C3-substrates (lactate, glycerol and pyruvate), was completely
reversed by the administration in vivo of a mixture of glucose and fructose
50min prior to perfusion (Table 15). Glucose alone (in a dose equal to
that of the glucose & fructose together, i.e., 3mmols) was not effective
in correcting the impaired synthesis, and neither was fructose (0.5mmol)
alone (Table 15). Insulin treatment for similar periods (50 or 75min)
0 60
0
11.1 0 o 40
?-1 0
ce
20
129
Fig. 21
20 60
100
TIME (MIN)
Tiniecoirseofneizlx2aeaEj.a.ccumulation in normal and
diabetic rats.
Livers were perfused with bicarbonate-buffered saline containing
erythrocytes, and additions as indicated. Two samples (median followed
by left lateral lobe), or three samples (above two preceded by.right
lobe) were removed in sequence. Pilled-in symbols represent normal
(starved) rat livers, and open symbols diabetic rat livers. Additions
were as follows: (i) glucose (30mM) plus C3-substrates (.4 , 0 , 6
perfusions; M , 12 perfusions ); (ii) supplemented medium, including
insulin, hydrocortisone and amino acids, plus glucose, 30mM (0, five
perfusions; 0, two, perfusions), or glucose (20mM initially) and
fructose, 5mM ( v , 4 perfusions). Other details are given in the
text. Results are mean values, and bars indicate the S.E.M.
Table 15. Glycogen accumulation during perfusion of livers from streptozotocin-diabetic rats
Livers from 48h-starved diabetic rats were perfused for 50min with bicarbonate-albumin-saline containing washed erythrocytes. Substrates were added after 15min. Glucose was initially 28mM, and the C3-substrates, lactate, glycerol and pyruvate, were initially 5mM, 3.3mM and 1.3mM (respectively),thei.infused. Fructose was initially 5mM, and then infused (0.5M73m01). Pretreatments were administered subcutaneously, 50min before perfusion (unless indicated): glucose plus fructose in 2m1 , (1.25M and 1.25M respectively)7fructose (2m1) 0.25M) or glucose (2m17 1.5M). Other details are in the text. Results are mean +
Glycogen content Cumol of Calculated rate glucose per g of fresh liver) of glycogen
accumulation
Glucose concentration
(11`1)
Perfusions with Flucose,
20min (median lobe)
50min (left lateral
lobe) lactate, glycerol and pyruvate
Pretreatment No of perfusions
None 6 3.5 9.3
Glucose, fructose 6 48.1 72.3
Glucose, fructose & anti-insulin serum 4 6.9 18.6 Glucose, fructose & control serum 4 18.6 39.1
Insulin (50min) 5 11.3 26.6 Insulin (75min) 3 13.0 31.6 Perfusions with glucose plus fructose
Pretreatment No of perfusions
None 7 7.1 17.7 Fructose 3 3.5 9.6 Glucose 3 42.9 54.3 Glucose, fructose 9 25.6 50.8
(jumol glucose/g/min) 20min 50min
0.19 + 0.04 27.1 + 0.9 28.7 + 1.3
0.81 + 0.12 29.4 + 1.2 32.0 + 1.4
0.39 4. 0.08 29.3 + 1.0 31.5 + 1.1
0.68 + 0.18 28.2 + 1.3 29.9 + 1.6
0.51 + 0.15 27.5 ± 0.8 29.0 + 0.9
0.62 4. 0.24 28.7 + 0.9 30.9 + 0.6
0.35 + 0.07 --- 25.6 ± 0.9 28.5 ± 1.1 0.20 + 0.06 30.4 + 1.6 33.7 + 1.3 0.38 + o.o5 30.6 ± 0.9 32.7 ± 0.9 H 0.84 + 0.13 31.0 4. 0.8 33.6 + 1.2 0
131
largely restored the rate of glycogen accumulation (Table 15).
The possibility that insulin was implicated in the response
to hexose pretreatment in vivo was investigated by administering anti-
insulin serum with the glucose and fructose; reduced rates of glycogen
synthesis were observed in the perfused liver compared to those after
treatment in vivo with glucose, fructose and control serum (Table 15).
The rate was not, however, reduced to that seen when there was no
treatment in vivo and the livers were perfused with C3-substrates. The
rate of glycogen accumulation, when fructose was present in the perfusion
medium, was higher than when lactate, glycerol and pyruvate were added,
as was also observed in normal (starved) livers (Section 3.1.5).
Insulin in the medium (in the absence of treatment in vivo) did not
increase the rate of glycogen synthesis from glucose plus fructose or
other substrates (Fig. 21., Table 17).
In an attempt to increase glycogen synthetic rates during
perfusion of livers from diabetic rats, the standard Medium was
supplemented with further substrates and hormones. Two groups of
perfusions were carried out. In one group, events during pretreatment
with glucose and fructose were simulated in the initial phase (about Ih)-
of perfusion : the medium contained fructose (initially 5E11), glucose
(20mM), and insulin, cortisol and amino acids (the latter three infused,
essentially according to John and Miller (1969)). After 50min , more
glucose was added to bring the concentration to about 30mM (optimal for
glycogen synthesis; see Section 3.1.2) and 03-substrates were added
(total 10111M initially, and then infused). Net glycogen accumulation
under these conditions, determined between 70 and 100min (Fig.21) was
no faster than in perfused livers of other -Untreated diabetic rats.
132
Thus fructose, glucose and insulin were not effective in restoring
glycogen synthesis when added to the perfusate.
In a second group of perfusions, the medium was supplemented
with insulin, hydrocortisone and amino acids (as above, but without
fructose) in addition to the standard condition of glucose (30mM) plus
C3-substrates. Liver samples were removed after 20 and 50min. There
was no improvement in the low rates of glycogen accumulation (Ilg. 21).
When the perfusion time was extended to 80min similar low rates of
glycogenesis were observed.
In control experiments, livers from normal (starved) rats
were perfused with thesupplemented medium; net glycogen synthesis (Fig.
21) occurred at a rate similar to that in perfusion; with glucose plus
C3-substrates (Fig.21 and Table 11) or fructose (Table 11). Thus,
supplementation did not inhibit synthesis in normal (starved) livers.
In all conditions, there was net output of glucose during
perfusion, as in livers of normal starved rats perfused in these
conditions (even during maximal glycogen deposition). If no glucose
was added to perfusions of diabetic rat livers, there was no glycogen
accumulation, and the rate of net gluconeogenesis (from C3-substrates)
was 1.11 0.20 (3))umol glucose/min/g of fresh liver (mean ± S.E.M.).
Thus the capacity for glucose synthesis was sufficient to support normal
net glycogen deposition (0.68 - 0.82)amol/min ,/g).
133
3.4.2 Glycogen synthetase and_philluhcmrlase activities
in vivo and in the perfused liver of diabetic
rats.
To gain insight into the impairment of net glycogen
accumulation in the perfused liver of diabetic rats the activities of
the enzymes glycogen synthetase and phosphorylase were measured.
Zacozmamthetase and phosphorylase activities in vivo.
The activities of glycogen synthetase and phosphorylase were
compared in intact normal and streptozotocin-diabetic rats (Table 16).
In starved diabetic rats, compared to normal starved rats, there was
a moderate decrease in the proportion of synthetase "a" due to an
increase in total activityl and an associated slight increase in
phosphorylase (expressed per g). These enzymes were assayed simultaneously
in the different lobes of the liver, as a pre-requisite to the
interpretation of their activities in lobes removed in sequence during
perfusion. There were no significant differences in the enzyme activites
between the various lobes, which allowed the enzymes and glycogen to be
measured in the same samples and the differences between sequential
samples to be assessed.
aycoreta.aeancl.ppahosholaseac'tyisia;the_pszf:used liver.
The enzymes glycogen synthetase and phosphorylase were assayed
in two sequentially removed samples during perfusion. After 50min perfusion
with glucose plus either fructose or C3-substrates (unsupplemented medium)
the proportion of glycogen synthetase in the "a" form (Method 1) was
51-60 (Table 17). In contrast to the result in normal (starved) rat
livers (Section 3.1.5) there was no clear correlation between glycogen
synthetase "a" activity (at 50min) and the rate of glycogen accumulation.
In most conditions the percentage of synthetase in the "a" form in the
initial sample (removed after 20min perfusion) was 23-40 which resembled
that in the intact diabetic animal (cf. Tables 16&17). In two groups of
Table 16. Activities of glycogen synthetase and phosphorylase in intact starved normal and starved
streptozotocin-diabetic rats.
Enzyme activities in simultaneously-sampled lobes of livers from 48h-starved rats were
assayed as described in the text. Results are means + S.E.M. from three normal and four diabetic rats. Mean /were
percentages of synthetase "a' calculated from % "a" values, in individual samples.
Glycogen synthetase (p.mol/imin/g : Method 1)
Glycogen phosphorylase (ipmol/min/g : centrifuged
Rat Lobe homogenate) "a" Total yo "a"
Normal Median 0.10 + 0.01 0.22 + 0.06 49 + 6 5.8 + 1.4
Normal Left lateral 0.13 + 0.01 0.32 + 0.05 45 4. 7 5.8 + 1.2
Normal Right 0.09 ± 0.02 0.20 + 0.04 43 3 4.5 + 0.7
Diabetic Median 0.14 + 0.01 0.37 + 0.01 39 + 5 6.1 + 0.8
Diabetic Left lateral 0.11 4- 0.02 0.33 + 0.02 34 + 8 6.4 + 0.6
Diabetic Right 0.13 + 0.04 0.35 + 0.07 38 ± 6 6.5 ± 0.3
Table 17. G1 cozen s thetase activi in the erfused liver of starved stre tozotocin-diabetic rats
Livers were perfused as described in Table 15., with various substrates and after different pretreatments.Glycogen synthetase was assayed (Method 1) in liver samples removed after 20 and 50 min. Details are in the text. Other results from these perfusions are included in Table 15. Results are means +
Glycogen synthetase (paolboin/g of fresh liver)
"a" Total % via!?
Calculated rate of glycogen accumulation
(pniol of glucose/ min per g)
20 min 50 min Perfusions with glucose, lactate, glycerol and pyruvate
"a" Total % "a"
Pretreatment No of perfusions
None 4 0.13+0.01 0.30+0.05 48+6 0.11+0.02 0.19+0.03 60j3 0.17+0.09 Glucose, fructose 4 0.13+0.02 0.38+0.07 35+6 0.16+0.01 0.27+0.03 64-19 0.86+0.17 Glucose, fructose& anti-insulin serum
4. 0.14+0.15 0.32+0.05 461.5 0.14+0.02 0.25+0.06 61±9 0.39+0.08
Glucose, fructose & 4 control serum 0.10+0.03 0.23+0.03 42+8 0.09+0.01 0.16+0.01 56±9 0.68+0.18
Insulin 4 0.13+0.01 0.29+0.02 46+2 0.15+0.04 0.23+0.05 64+6 0.59+0.16 Nbne * 3 0.16+0.04 0.23+0.04 68+6 0.17+0.04 0.22+0.05 78+4 0.13+0.02 Perfusions with glucose, plus fructose
Pretreatment No of perfusions
None 4 0.141:0.02 0.431-0.04 34±5 0.181-0.06 0.25-10.05 681-9 0.38+0.10 Fructose 2 0.09 0.33 28 0.09 0.18 51 0.19 Glucose 3 0.12±0.01. 0.33±0.06 39±7 0.13-10.02 0.20±0.01 66±15 0.38+0.05
.Glucose,fructose 4 0.16+0.04 0.2 650 0.121-0.01 0.19-10.02 64±5 0.86+0.21
* Supplemented medium, including insulin: samples at 70 and 100 min.: See Fig.20
136
perfusions, where the liver was in contact with fructose for about
100min (in vivo and or perfusion), the percentage of synthetase "a"
was higher (Table 17) : (i) pretreatment and perfusion with glucose and
fructose (65 and 64% "a"), (ii) no pretreatment in vivo, but prolonged
perfusion with glucose and fructose in supplemented medium (68.and 78%
"a"). These high values for synthetase "a" were associated with rapid
glycogen accumulation in the former situation, but not the latter.
The activities•of glycogen phosphorylase in liver samples
removed during perfusion were lower than those in intact diabetic rats
(cf. Tables 16 & 18). The activities at 20 or 50min did not conform
to any pattern (Table 18). Changes in glycogen synthetase and
phosphorylase during perfusion were assessed from the activities in
sequential liver samples. In the livers of diabetic rats,when net
glycogen accumulation was restored by pretreatment,there was an increase
in the proportion of glycogen synthetase "a" during perfusion (Table 17),
which presumably reflected the activating effect of substrates on
synthetase, as observed in perfused livers of normal (starved) rats
(Section 3.1.5). This response was much less marked in diabetic perfusions
where glycogen synthesis was low (Table 17). Taking all groups of
diabetic perfusions with glucose and C3-substrates, the increase in the
proportion of synthetase "a" between 20 and 50min was correlated with
the rate of net glycogen accumulation (Pig 22a). In perfusions with
fructose there was no such correlation (Table 17).
In the experiments described above, the alteration in %
synthetase "a" during perfusion was manifested by a fall in total enzyme
(Table 17) rather than in the absolute value of synthetase "a". This
may have reflected alteration in the distribution of synthetase in
Table 18. Phosphorylase activity in the perfused liver of starved streptozotocin-diabetic rats
Livers were perfused as described in Table 17 (and Table 15). Glycogen phosphorylase was
assayed in (centrifuged) samples for which the glycogen synthetase results are given in Table 17. Other details
are in Table 17, or the text.
Glycogen phosphorylase Rate of glycogen (Fmol/Min/g of fresh liver) accumulation
!amo1 glucose / 20min 50min min per g)
Perfusions with lucose, lactate,, glycerol and pyruvate
Pretreatment:
None 1.93 + 0.48 1.40 + 0.23 0.17
Glucose, fructose 2.92 ± 0.58 1.72 + 0.52 0.86 \
Glucose, fructose,anti -insulin serum 3.12 ± 0.51 2.53 ± 0.75 0.39
Glucose, fructose, control serum 1.78 ± 0.23 1.46 ± 0.24 0.68
Insulin 2.23 + 0.52 1.25 ± 0.24 0.59
None * 2.62 ± 0.87 2.34 ± 0.52 0.13
'Perfusions with glucose plus fructose
Pretreatment:
None 2.55 + 0.06 1.38 + 0.21 0.38
Fructose 2.09 1.47 0.19
Glucose 1.72 + 0.53 1.51 + 0.47 0.38 H
Glucose plus fructose 2,39 + 0.25 1.48 + 0.20 0.86 ■' 4
* Supplemented medium, samples at 70 and 100 min: See Fig.21
•
_ 138
For Figure 22. See Over—leaf
Fig. 22. Relationship between net hepatic glycogen
accumulation and glycogen synthetase and
phosphorylase in diabetic rats.
In livers perfused with glucose plus 03-
substrates, for which glycogen changes are given in Table 15,
and measurements of glycogen synthetase and phosphorylase
activities (pmol/g/min) in Tables 17 and 18, mean values for
the change between 20 and 50min perfusion in (a) the
percentage of glycogen synthetase "a" and (b) the quotient
(percentage synthetase "a")/(phosphorylase) were calculated.
Rats received either no pre-treatment (o, no additions; A 7
supplemented medium; see Fig. 21) or insulin ( A ), or glucose
and fructose (o ), or hexoses plus either anti-insulin serum
( a ) or control serum ( 0 )0 Other details are in the text.
Bars indicate S.E.M., and the numbers of perfusions are in
Table 17. Lines were fitted by regression analysis of values
from individual perfusions : (a) r = 0.54, p < 0.01
(23 observations); (b) r = 0.64, p < 0.01 (23 observations).
0
O
3 0 CD
C0 co
3 0
CD
•;•• c0
<A
tn
O
O co
O
CHANGE IN ENZYME ACTIVITY IN 30 MIN PERFUSION
Increase in (% synthetase"al(phosphorylase) Increase in % glycogen synthetase "a" O 0 11 8 0
140
centrifuged homogenates, or could have been a consequence of the
action of modifiers in the homogenates if UDPG was not at saturating
concentrations (see discussions by Bishop and Larner, 1967; Blatt and
Kim, 1971b). To distinguish between these possibilities the details
of the synthetase assay were altered (Method 2) and the enzyme measured
in key groups of perfusions. In these uncentrifuged homogenates, there
was no decline in total synthetase during perfusion (Table 19). However,
the main finding of the experiments using Method 1 were confirmed, i.e.,
the proportion of synthetase "a" (and the absolute value) increased
duririg perfusions of livers from diabetic pretreated rats, but not in
livers of un -treated animals (Table 19). The rates of glycogen deposition
resembled those in Table 15; with insulin in the perfusate, the average
rate was 0.26pmol of glucose/Min/g measured in three perfusions, confirming
the lack of a direct hepatic action.
There was a decline in the activity of phosphorylase (assessed
between 20 and 50min) during perfusion of diabetic rat livers (Table 18)
as would be expected as glycogen synthetase and phosphorylase often
exhibit reciprocal changes. The extent of the changes in each enzyme
were correlated (Fig. 23). There was relatively little decline in
phosphorylase during perfusion of liver from diabetic rats with glucose
plus 03-substrates (0.53 + 0.26 (4) pmol/min/g fresh liver) compared
to thegreaterdecrease in perfusions after pre-treatment in vivo (1.20
0.38 (4) jumol/min/g), in which net glycogen accumulation was more rapid
(Table 18). Taking all groups, there was not however, any clear-cut
inverse correlation between the net rate of glycogen synthesis and the
decline in phosphorylase activity during perfusions (with or without
fructose in the perfusate).
Table 19. Glycogen synthetase activity in the perfused liver of starved streptozotocin-
diabetic rats.
Livers were perfused as described in Tables 15 and 17, with glucose, lactate, glycerol
and pyruvate, and glycogen synthetase was assayed (Method 2). In one group (*) insulin was infused into the
perfusion medium: 1.5m1A of a solution containing 330mU/M1., following an intial dose of 500mU. Other details
are in the text. Results are means +
Glycogen synthetase cumol/min/g of fresh liver : Method 2)
Pretreatment (in vivo). No of perfusions.
20 min 50 min
"a" Total % "a" "tar" Total % nal?
None 0.52 + 0.09 0.84 + 0.11 62 + 7 0.55 + 0.40 1.05 + 0.12 53 4. 4
Glucose, fructose 3 0.64 + 0.05 0.98 + 0.09 66 ± 6 0.83 + 0.10 1.11 + 0.08 .75 + 6
Insulin (IU, 75 min, before perfusion) 4 0.60 + 0.16 1.12 + 0.23 52 + 3 0.78 + 0.15 1.12 + 0.18 69 + 5
None * 6 0.51 + 0.07 0.99 + 0.07 51 + 6 0.52 + 0.09 0.99 ± 0.08 52 + 6
* Insulin added to perfusion medium
0 A
z 0 L.
w
z
C)
z-
CH
AN
GE
1 N
GLY
CO
GE N SYN
TH
ETASE "a
"
+20
0
142
•1.40
0
+0.4 0 -OA -0.8 -1.2 -2.0
CHANGE IN PHOSPHORYLASE ACTIVITY IN 30 MIN PERFUSION
Fig. 23. Relationship between the changes in glycogen synthetase
and phosphorylase during liver perfusion in diabetic rats
Livers were perfused with glucose plus 03-substrates, with or
without pre-treatment in vivo (see Tables 15, 17 and 18, Pig. 22). an Chges in the activities of glycogen synthetase (percentage "a" and
phosphorylase Oa mol/min/g) between 20 and 50min are presented for the individual perfusions described in Tables 17 and 18. Symbols are as in Fig. 22. Other details are in the text. The line was fitted by regression analysis, taking into account variation in both axes : r = 0.67, p < 0.001. (23 observations).
143
The above results demonstrated that the response of enzymes
to glucose and 03-substrates during perfusion was impaired in diabetic
rat livers, anoVres was tored after pretreatment in vivo with hexoses or
insulin. A combined measure of the functional state of both synthetase
and phosphorylase is the quotient (% synthetase "a")/(phosphorylase). The
change in this quotient between 20 and 50min perfusion with glucose plus
C3-substrates was significantly correlated with the net rate of glycogen
accumulation in all groups if diabetic rat livers (Fig. 22b).
3.4.3. Influence of glucose
of glycogen synthetase andhosse in vivo
The impairment in glycogen accumulation observed in the
perfused liver of diabetic rats was reversed by the administration in
vivo of a mixture of glucose and fructose 50min prior to perfusion (see
Section 3.4.1),It was therefore of interest to study the effect of this
treatment in vivo on glycogen synthetase and phosphorylase.
In normal rats, glucose and fructose each induced conversion
of synthetase to the "a" form (Table 20). Glucose and fructose together
brought about the greatest "activation" of the enzyme (Table 20) in
agreement with observations in the perfused liver (Table 11; Section
3.1.5). In diabetic rats these effects were less marked (Table 20).
Administration of the hexoses, either alone or in combination, also
produced a decline in phosphorylase activity in normal rats (Table 20),
although the effect of glucose plus fructose was not greater than either
hexose alone. A similar response was seen in the diabetic rat indicating
that control of phosphorylase is not impaired.
Table 20. Influence of :lucose and fructose on lons thetase and •hosh lase in
intact starved normal and streptozotocin-diabetic rats.
Enzyme activities in simultaneously-sampled lobes from 48h-starved rat livers were
assayed as described in the text. - Treatments were administered in 2m1, S.C., 50 min before sampling: glucose
(2.5M), fructose (0.25M) or glucose plus fructose (1.25M and 0.25M respectively). For comparison, two values
from Table 16 are included. Other details
all groups except fructoss-treated diabetic
are in the text. Results are means
rats (5 animals).
Glycogen synthetase (% "a"). (Method 1)
± S.E.M. from three rats, in
Glycogen phosphorylase centrifuged homogenate) pmol min g)
Rat Treatment Median lobe Left lateral lobe Median lobe Left lateral
lobe Normal -* 49 + 6 45 + 7 5.8 4- 1.4 5.1 4. 1.2
Normal Glucose 77 + 2 70 + 4 2.1 + 0.3 1.5 + 0.2
Normal Fructose 60 ± 5 69 .... 4 2.7 + 0.4 2.1 ± 0.2
Normal Glucose plus fructose 83 + 7 81.- 6 2.3 + 0.1 2.7 + 0.2
Diabetic -* 39 + 5 34 + 8 6.1 + 0.8 6.4 ± 0.6
Diabetic Glucose 50 ± 5 57 + 7 2.6 + 0.4 2.6 4. 0.1
Diabetic Fructose 46 ± 7 46 + 6 3.5 ± 0.6 2.5 ± 0.4
Diabetic Glucose plus fructose 47 4- 6 48 ± 3 3.7.4- 0.3 3.1 4. 0.6
* From Table 16 for comparison 1-,
145
3.4.4 Hepatic glyco gen accumulation in the intact
diabetic (starved). rat.
For comparison with resultsin the perfused liver, glycogen
accumulation was followed in anaesthetised intact streptozotocin-
diabetic rats. An impairment in the capacity for net glycogen synthesis
was observed during infusion of either glucose or glucose plus a mixture
of glycerol, serine and pyruvate, following pre treatment in vivo
(Table 21). As in the experiments with the perfused liver, fructose
and glucose were required to bring about rapid glycogen deposition
in diabetic rats (Table 21).
Table 21. Glycogen accumulation in intact starved normal and starved streptozotocin-diabetic rats.
Glycogen accumulation was followed in intact rats. Rats received various pre-treatments (S.C.,1°
in two portions of Iml , unless indicated; see Table 15), were then anaesthetised with Nembutal, and received
an infusion (3ml/h) into a tail vein of glucose (1-1.25K), fructose (0.25K) or gluconeogenic C3-substrates (a
mixture of glycerol, serine and pyruvate, each 0.33M, also used for pre-treatment). The first liver sample was
removed within about 15min of anaesthesia, and the second lh later. The indicated time of pre-treatment refers
to the interval between injection and the first sample. Other details are in the text. Results are means S.E.M.
Rats Pre-Treatment Infusion Glycogen content ()umol
No. of of glucose per g of experiments fresh liver
Sample 1 Sample 2
Calculated rate of glycogen accumulation
ol glucose,/ g/mmn)
Final blood glucose concentration
(lox)
Glucose
Glucose plus fructose
Glucose
Glucose
Glucose plus C3 -substrates
Glucose plus fructose
Glucose plus fructose
Normal None
Normal Glucose plus fructose (60min)
Diabetic Glucose 12min)
Diabetic Glucose (53min)
Diabetic Glucose plus C -substrates 7
Diabetic Glucose plus fructose (i.g., 35min)
Diabetic Glucose plus fructose (60min)
35.2 72.3 0.62 4. 0.09 32±4 4 110.6 155.0 0.74 4. 0.15 36 4. 7
2 5.2* 8.4* 0.06 25
3 77.2 70.6 0 37 ± 5 3 11.5* 17.9* 0.12 4. 0.10 64 ± 17
4 26.6 63.2 0.61 4. 0.18 61+9
72.1 109.9 0.63 ± 0.08 40 ± 3
(i.g.,
* Samples 53min. apart Abbreviations : s.c., subcutaneously; i.g., intragastrically.
rn
147
3.5 HEPATIC CARBOHYDRATE METABOLISM IN THE STARVED
ADRENALECTOMISED RAT
1. Glycogen accumulation in the perfused
liver from adrenalectomised rats.
2. Glycogen synthetase and phosphorylase
activities in vivo and in the
perfused liver of adrenalectomised
rats.
148
3.5 HEPATIC CARBOHYDRATE METABOLISM IN THE STARVED
ADRENATRCTOMISED RAT
Net rates of glycogen accumulation were obtained in the
perfused liver of normal starved (see Section 3.1) and diabetic
starved rats (see Section 3.4), and the relationships established
between glycogen and the enzymes of its metabolism, glycogen synthetase
and phosphorylase. Since the role of adrenal cortical steroids in
glycogen metabolism is not clear (see Introduction, Section 1.4.2)
it was of interest to study glycogenesis and its control in the et
adrenalectomised rat under conditions optimal fo rates of glycogen
accumulation in other states.
3.5.1 Glycogen accumulation in the_perfused
liver from adrenalectomised rats
Glycogen synthesis was studied in the perfused liver from
starved adrenalectomised rats in the conditions which allowed net
rates of accumulation in the normal starved rat; no synthesis was
observed (Table 22). Supplementation of the perfusion medium with
hydrocortisone, amino acids and insulin did not increase the rate of
glycogensis. Pre-treatment of the rat in vivo for 50min with glucose
and fructose prior to perfusion of the liver, (a treatment which restored
synthesis in the diabetic rats : see Section 3.4.1) increased synthesis
in the presence or absence of hydrocortisone or insulin in vitro,
but not to the rates observed (Table 22) in sham-operated rats. In
all these perfusions glucose concentration was 27.5-30.5mM. There was
slight glucose output during perfusion of sham-operated or pre-treated
adrenalectomised rats (Table 22). Thus the net carbon source of
glycogen in these perfusions was C3-substrates, rather than glucose,
as in similar perfusions of starved (Section 3.1), or starved diabetic
(Section 3.4) rat livers. In livers from untreated adrenalectomised
rats (where there was no glycogen synthesis : Table 22), there was no
149
For Table 22, see over-leaf
Table 22 Glycogen accumulation in the perfused liver from adrenalectomised starved rail
Livers from 48h- starved adrenalectomised rats were perfused as described in the text.
In all perfusions, glucose was initially 28mM and3-substrates (lactate, glycerol and pyruvate) were
5mM, 3.3mM and 1.3m1v1 (respectively) and then infused at 3m1/h from 15min. The full supplement medium
containing insulin, amino acids and hydrocortisone was as described in the Methods (Section 2.3.2).
Pre--treatment was administered subcutaneously at the time indicated before perfusion : glucose plus
fructose in 2m1 (1025M and 0.25M respectively) and 1U of insulin. Results are means -I- S.E.M. with
the number of observations in parentheses.
Livers were perfused with glucose (28mM), fructose (5mM) and insulin (added
and infused) for 2h,3-substrates were added and infused with insulin from 135min and liver and medium
samples were taken at 140 and 170min.
Livers were perfused with glucose (28mM), fructose (5mM), C3-substrates (10mM),
amino acids mixture (4 x "normal"), hydrocortisone and insulin (added and infused). After 60min perfusion,
amino acids (2 x "normal"), fructose (5mM) and 03-substrates (10mM) were added. The first and latter of
these were also added after 120min perfusion. C3-substrates were infused from 165min, liver biopsies and
medium samples being removed at 180 and 240min.
Sham-operated controls.
208+1.5 2,1+1.4 27.5+0.8 26.6+0.9 -0.02+0.02 (6)
Full supplement
1.5+1.0 .1.7+0.6 28.5+0.6 28.9+0.7 0.01+0.04 (5)
2.8+1.2 7.4+1.8 29.1+1.2 32.0+1.5 0.16+0.05 (6)
Full supplement
5.8+2.0 15.1+4.3 27.5+1.1 28.1+1.0 0.31+0.09 (9)
Insulin 6.0+1.4 10.5:0.9 29.4+1.9 30.9+1.6 0.15+0.13 (3)
Hydro-cortisone
3.2+1.0 9.6+1.5 28.3+1.3 30.8+1.2 0.21+0.05 (4)
Hydro-cortisone
6.0+4.8 14.4+6.7 28.3+1.2 30.0+0.9 0.28+0.09 (s)
Hydro- cortisone &
1.6+0.5 9.0±3.4 26.6+0.9 28.4+1.1 .0025+0.10 (6)
Insulin
*Insulin & fructose
*0.6+0.1 *1.5+0.5 *32.7+0.7 *32.9+0.9 0.03+0.01 (3)
+Fructose & +18.2 +14.2 +32.5 1-33.3 -0.07 (2) Full supplement
19.3+4.9 38.8+6.0 28.4+0.7 50.4+0.9 0.65+0.06 (11)
Fructose & glucose
Fructose &
50 glucose
Fructose & gltcose
Fructose &
50 glucose
Fructose, 60 glucose & insulin
Fructose, 60 glucose & insulin
Fructose, 45 glucose & insulin
Glycogen content Pre-Treatment Time Additions to (rol of glucose/g fresh liver)
(min) perfusion 20min 50min 20min
(median lobe) (left lateral lobe)
Calculated rate of glycogen accumulation
50min (pmol glucose/g/min)
Glucose in medium (DE)
151
change in glucose concentration during perfusion. If glucose was
not added to such perfusions (with the mixture of lactate, glycerol
and pyruvate), the rate of net gluconeogenesis was about 1.5pmol/dmin (C.J
Kirk : unpublished results). Thus the capacity for gluconeogenesis
in these perfusions was sufficient to have supported glycogen synthesis.
In order to evaluate the impairment in glycogen accumulation
in the perfusion, adrenalectomisedrats were treated with hydrocortisone
in vivo for varying times prior to liver perfusion. Net rates of
glycogenesis (measured between 20 and 50min perfusion) comparable to
those observed in normal animals, were obtained when the animal was
pre-treated for 4h before perfusion (Fig. 24). When hexoses were
given with the steroid the restoration in glycogen accumulation was
seen earlier (Fig. 24) : significant rates being observed between 20
and 50 min of perfusion, after 2h pre-treatment in vivo.
The above time course of restoration of synthesis by
hydrocortisone suggested that net rates of glycogen accumulation were
not restored by the initial attempt at restoration with substrates,due
to the short pre-treatment time. Since it is likely that the hexose
treatment in vivo causes insulin secretion, which could have a role
in restoration, insulin was given in addition to the glucose and
fructose, and the rates of glycogenesis measured in perfusion after
varying times of pre-treatment in vivo. After 2h, in vivo treatment,
normal net rates of glycogen accumulation were obtained in the
perfusion (Fig. 24) which were not increased by longer treatment before
perfusion. Insulin in vitro had no effect on the rate of synthesis
after lh pre-treatment (cf. Table 22 & Fig. 24) although the addition
of hydrocortisone to the perfusion medium perhaps had a small effect
in the short -term (cf. Table 22 & Fig. 24).Due to this apparent
stimulation of hepatic glycogen accumulation by hydrocortisone in vitro
(during perfusion for 50min) the effect of this hormone in perfusion
was assessed over a longer time period. In the presence of amino acids,
LL. 0
w F-
GLY
CO
GEN
SYN
THES
IS
.152
0 1 • 2 3. 4 LENGTH OF PRETREATMENT PRIOR TO PERFUSION ( H
Fig. 24 Time course of restoration of net rates oflaa22gen.
position in the perfused liver from adrenalectbmised
rats.'
Rats were treated subcutaneously with fructose, glucose and insulin, ( as described in Table 22),o ; hydrocortisone succinate,
A 10mg; or fructose, glucose and hydrocortisone succinate (l.2517, 0.25M and 10mg, respectively)pyith the addition of insulin for lh value;or no pre-treatment, 0. After various times of pre-treatment of the rat, the liver was perfused with glucose (28mM) and C7-substrates as indicated in Table 22, and liver biopsies taken at 20 .ind 50min. Results are means of at least 3 experiments and bars indicate the S.E.M.
153
fructose, insulin and hydrocortisone the rate of glycogen synthesis
measured in liver biopsies removed after 3 and 4h perfusion was howevernegligible (Table 22).
Net rates of glycogen accumulation were therefore observed
in the perfused liver only after treatment in vivo with fructose,
glucose and insulin for 2h. When these in vivo events were simulated
in the perfusion and liver samples removed at 140 and 170min no
synthesis was obtained (Table 22).
For comparison, it may be mentioned that studies in the
intact anaethetised adrenalectomised rat have demonstrated that
hepatic glycogen accumulation is also impaired in vivo in short-term
experiments (C. Kirk, unpublished results). Thus when glucose was
infused for lh and liver biopsies removed for glycogen determination,
the rate of synthesis was negligible (about 0.03 pmol glucose/g/min)
in the adrenalectomised rat (cf. Table 21, Section 3.4.4). If
glucose was administered (sub-cutaneously or intra-gastrically) 1.5 -
2h before anaesthesia ratesof about one-third normal (0.1-0.37 iumoVe min) were observed on subsequent glucose infusion. These results
support (by a similar procedure involving sequential biopsies in
vivo) the find.irg that glycogen accumulation is impaired in the perfused
liver from adrenalectomised rats and that synthesis is not restored
in the short-term, by treatment with hexoses.
3.5.2 Glycogen synthetase and activities
in vivo and in the erfused liver of
adrenalectomised rats.
As .a pre-requisite to the interpretation of the activities
of glycogen synthetase and phosphorylase in lobes of the liver
removed in sequence during perfusion, and to gain insight into the
impairment of net glycogen accumulation in the perfused liver, tie
enzyme activities were measured in the intact rat.
154
SycoaensynIhetase and hos hor lase activities in vivo
Glycogen synthetase.and phosphorylase activities were not significantly
different in the sham-operated animal and normal rat (Table 23, cf.
Table 16 Section 3.4.2). In the starved adrenalectomised rat the
activity of (supernatant) hepatic glycogen synthetase (total) was increased,
producing a significant decrease in the % "a" form (Table 23). The
relatively greater recovery of (total) synthetase in the high-speed
supernatant fraction in adrenalectomised rats (Table 23) may reflect
their lower glycogen content, i.e., there would be less "particulate"
synthetase. No difference between the various lobes of the liver in
either group of rats was observed; this allowed differences between
activities in sequential liver biopsies to be assessed as rates of change.
amogen svnthetase and phosBhozylase activities in the erfused liver.
The enzymes of glycogen metabolism were measured in two sequentially
removed samples during perfusion. As was noted in normal (starved) rat
livers (Table 11; Section 3.1.5) but not in diabetics (Table 17; Section
3.4.2) there was a correlation between the % "a" form at 50min and the
rate of glycogen accumulation (Table 24). This correlation was also
observed when fructose was substituted for C3-substrates in the perfusate,
unlike the results obtained in the diabetic perfusions (see Table 17,
Section 3.4.2). Since these assays were carried out in the uncentrifuged
homogenate, no loss of total enzyme activity during perfusion was recorded
(see also Section 3.4.2); thus a correlation between the actual activity
of the "a" form at 50min and rate of synthesis was also observed (Table
24). In most conditions the percentage of synthetase "a" in the initial
sample (removed after 20min perfusion) was 21-38% which was slightly greater
than that in the intact adrenalectomised rat (cf. Tables 24 & 23),
assuming that centrifugation did not decrease % "a" (see Section 3.1.4).
The activities of glycogen phosphorylase in liver samples
removed during perfusion were lower than those in the intact
Table 23. Glycogen synthetase and thosphorylase activities intact sham adrenalectomised
and adrenalectomised rats.
Enzyme activities in simultaneously-sampled lobes of livers from 48h-starved rats were
assayed as described in the text. Results are means + S.E.M. from three rats. % "a" was calculated for each
sample in order to obtain the S.E.M. for the mean values.
Glycogen synthetase (Method 1) Phosphorylase Rat Liver lobe (pnol / min g) (Centrifuged homogenate)
1•11.1■1111.1•1■011■1•••■■••••■•••■••••••••••••••• ••
"a" Total % n d "a" (pmol/minig)
Sham adx. Median 0.13 + 0.03 0.23 + 0.01 57 + 8 4.68 + 1.30
Sham adx. Left lateral 0.13 + 0.02 0.26 + 0.06 55 + 8 4.59 + 0.60
Sham adx. Right 0.13 + 0.03 0.28 + 0.01 47 + 7 5.40 + 0.24
Adx. Median 0.10 + 0.02 0.45 + 0.02 23 + 6 5.07 + 0.28
Adx. Left lateral 0.11 + 0.03 0.46 + 0.02 24 + 8 5.09 + 0.45
Adx. Right 0.09 + 0.02 0.47 + 0.04 21 + 6 5.01 + 0.53
Abbreviation : Adx. adrenalectomised.
Table 24. Glycogen synthetase activity in the perfused liver from starved adrenalectomised rats.
Livers were perfused as described in Table 22 and pre-treatment with F & G (fructose and glucose) was for 50min prior to perfusion. Glycogen synthetase was assayed (Method 1; uncentrifuged homogenate) in the same liver samples as was glycogen. Results are means + % "a" was calculated for each perfusion in order to obtain an estimate S.E.M. for the mean value.
Glycogen synthetase (pmol/Min/g) Rate of Pre-Treatment . Additions. to Rat No. of 20min L_50min glycogeh
the perfusion perfusions "a" Total % "a" ITan Total % "a" synthesis (Fmol glucose/ min/g)
Sham 3 0.34+0.11 0.56+0.12 56+11 0.53+0.09+ 0.67+0.05 79+8 0.61+0.08
- - Sham 3* 0.20+0.02 0.32+0.02 61./3 0.19+0.02 0.26+0.04 74+5 0.68+0.15
- - Adic. 3 0.12+0.01 0.48+0.03 250 0.14+0.01 0.55+0.05 25+2 -0.04+0.03
- *.13 Fructose Adx. 3 0.10+0.03 0.47+0.08 21+4 0.14+0.02 0.49+0.04 28+4 0.00+0.02
F & G - Adx. 3 0.14+0.05 0.52+0.10 26+5 0.26+0.05 0.66+0.06 38+4 0.19+0.10
F & G Full supple Adx. 4 0.21+0.04 0.53+0.03 38+6 0.35+0.08 0.73+0.02 47+11 0.31+0.17 -ment
- Full supple Adx. 3 0.13+0.01 0.43+0.05 31±3 0.12+0.01 0.44+0.05 270 -0.01+0.04 -ment
F & G Insulin Adx. 3 0.18+0.05 0.62+0.01 29+8 0.24+0.07 0.67+0.08 34+7 0.15+0.13
F & G Hydro- cortisone
Adx. 3 0.19+0.05 0.49+0.04 37+7 0.25+0.05 0.55+0.03 46±9 0.34+0.07
* Enzyme assayed in centrifuged homogenate (data not included in Fig. 25, 27 & 28)
..1' Fructose substituted for C3-substrates in perfusate
Abbreviations: adx. adrenalectomised sham. sham operated animal rn
F. fructose
G. glucose
157
adrenalectomised rat (cf. Tables 25 & 23), despite the fact that the
perfused liver samples were not centrifuged before assay. The activities
at 20 or 50min did not conform to any pattern (Table 25) as was observed
in the diabetic perfusions (Table 18; Section 3.4.2).
Changes in glycogen synthetase and phosphorylase during
perfusion were assessed from the activities in sequential liver samples.
When net rates of glycogen accumulation were observed in the livers of
sham-operated animals there was an increase in the percentage of glycogen
synthetase "a" during perfusion (Table 24). This response was less
marked in the adrenalectomised perfusions where glycogen synthesis was
low (Table 24). Taking all groups of perfusions, the increase in the
proportion of synthetase "a" between 20 and 50min (presumably reflecting
the enzyme activation by substrates) was correlated with the rate of net
glycogen accumulation (Fig. 25a). A similar enzyme sensitivity was
observed in the diabetic rat liver perfusions (see Fig. 22a; Section
3.4.2).
As has been noted previously (see Section 3.4.2) when the
glycogen synthetase activity was assayed in the centrifuged homogenate
the alteration in % "a" during perfusion was manifested by a fall in total
enzyme (Table 24) rather than in the absolute value of synthetase "a".
This fall was not observed when the crude liver homogenate was assayed,
when an increase in the absolute activity of the "a" form as well as the
percentage was seen during perfusion (Table 24). This increase in the
activity of the "a" form was correlated with the rate of glycogen
accumulation (Fig. 25b).
• When net rates of glycogen accumulation were measured in
the perfusion there was a decline in the activity of phosphorylase
(assessed between 20 and 50min); an increase occurred when there was a
negative or small rate of glycogen synthesis (Table 25). The change in
activity of phosphorylase between 20 and 50min was correlated with the
Table 25. Glycogen phosphorylase activity in the perfused liver from starved adrenalectomised rats.
All details are in Tables 24 & 22. Phosphorylase was assayed in Uncentrifuged homogenates
of the same liver samples as was glycogen synthetase and glycogen.
Phosphorylase ol/min/g)
Pre-Treatment Additions to the perfusion
Rat No. of perfusions 20min 50min
Sham 3 3.05+0.09 2.69+0.11
Sham 3* 2.31+0.40 1.65+0.11
Adx. 3 1.74+0.14 2.25-3.17
Fructose Adx. 3 2.26+0.52 3.16+0.73
F & G Adx. 3 1.76+0.0s 1.93+0.29
F & G Supple- mented
Adx. 4 2.32+0.07 2.49+0.10
Supple- mented
• Adx. 3 1.90+0.23 2.72+0.52
F & G Insulin Adx. 3 1.95+0.33 2.48+0.52
F & G Hydro- cortisone
Adx. 3 2.36+0.25 2.18+0.09
Rate of glycogen synthesis
Oumol glucose/Min/ g)
0.61+0.08
0.68+0.15
-0.04+0.03
0.00+0.02
0.19+0.10
0.31+0.17
-0.01+0.04
0.15+0.13
0.34+0.07
Fructose substituted for 03-substrates in perfusate
Enzyme assayed in centrifuged homogenate (data not included in Fig. 26)
Abbreviations : adx: adrenalectomised sham: sham operated animal Ft fructose G: glucose
For fig. 25 see Over-Leaf
159
Fig. 25 Change in sensitivity of glycogen synthetase
during net
in perfused livers from adrenalectomised rats.
In livers perfused with glucose plus C3-substrates,
for which glycogen changes are given in Table 22, and measure-
ments of synthetase in Table 24, mean values for the change
between 20 and 50min perfusion in (a) the percentage of glycogen
synthetase "a" and (b) the activity of glycogen synthetase "a"
(Ilmol/min/g) were calculated. Rats received either no pre-
treatment (0, no additions; A, supplemented medium : see Methods;
D , Sham-operated animals), or fructose and glucose (0, no
additions; A , supplemented medium; G3 , insulin; V , hydrocortisone).
Other details are in the text. Bars indicate S.E.M. and the
number of perfusions within each group are in Table 24. Lines
were fitted by regression analysis of values from individual
perfusions: (a) r = 0.72, p< 0.001 (22 observations);
(b) r = 0.73, p< 0.001 (22 observations).
Increase in % glycogen synthetase. "a"
0
❑ I
uo!p
alnw
mpo
a CD
0
2 - o
ca
3 o ca
CHANGE IN GLYCOGEN SYNTHET,ASE ACTIVITY N 30 .MIN PERFUSION
0
Increase in glycogen synth eta se "a" 0 0 O 0
(IA 1>--1
161
rate of glycogenesis (Fig. 26). A similar fall in phosphorylase
activity due to substrates in perfusate was observed during perfusion
of diabetic livers (see Section 3.4.2). Although the enzyme was assayed
- in the uncentrifuged homogenate of adrenalectomised rat livers the
decrease was less, for a corresponding rate of net glycogen accumulation,
i.e., a smaller decrease in phosphorylase activity was observed than in
the(centrifuged) samples from diabetic rat liver perfusions. As might
be expected, the alterations, during perfusion, in both glycogen
synthetase and phosphorylase were correlated (Fig. 27a & b).
The above results demonstrated that the response of enzymes
to glucose and 03-substrates during perfusion was impaired in adrenalect-
omised rat livers. A combined measure of the functional state of both
synthetase and phosphorylase is the quotient % glycogen synthetase "a"
/phosphorylase or glycogen synthetase "a" / phosphorylase. The change
in these quotients between 20 and 50min perfusion were significantly
correlated with the net rate of glycogen accumulation in all groups of
adrenalectomised rat livers (Fig. 28). When the results are compared
with those from diabetic rat liver perfusions (Fig. 22b, Section 3.4.2)
it would appear that the quotient % glycogen synthetase "a" / phosphorylase
did not increase by as much during adrenalectomised rat liver perfusions,
as was observed in the diabetic rat. As has been seen, the change in %
"a" during perfusion of adrenalectomised rat liver was similar to that
obtained in perfused diabetic rat livers. However, phosphorylase did
not appear to respond to substrates to the normal degree at agiven
glycogen synthesis rate : this was illustrated previously (Fig. 26) and
in the proportional changes of both enzymes during perfusion (cf. Fig 27 &
23),It would therefore seem that'in the adrenalectomised rat liver, the
response of glycogen phosphorylase to activitors or inhibitors is impaired,
as reflected in the lower change in activity, at a given glycogen synthetic
rate, compared to that in normal (or diabetic) rats.
0
IN 3
0 M
IN PE
RFU
SIO
N
+1. 0 162
.111■,•1
0. 4
+0.2
0.1 0.2 0.3 0 = 0.'5 0.6 0.7
•••••■11.
-0.6-- RATE OF NET GLYCOGEN ACCUMULATION (11MOLI GI MIN )
Fig. 26 Response of -phosphorylase activity during net rates sgely_ coon. accumulation.
In livers perfused with glucose plus 03-substrates, for which
glycogen changes are given in Table 22, and measurements of phosphorylase in Table 25, mean values for the change between 20 and 50ain perfusion in phosphorylase activity (p mol/min/g) were calculated.. Symbols are as in Fig. 25. The line was fitted by regression analysis of values from individual perfusions; r = 0.75, p < 0.001 (22 observations). Bars indicate S.E.M. and the number of perfusions in each group are in Table 25.
OF_
PHO
SPHO
RYLA
SE
163
For Fig. 27 see Over—Leaf
Fig. 27
Relationship be
synthetase and phosphorylase during liver
perfusion. in adrenalectomised rats.
Livers were perfused with glucose plus C3
substrates, with or without pre-treatment in vivo (see Tables
22, 24 & 25). The mean, changes in activities of glycogen
synthetase (% "a" and "a") and phosphorylase between 20 and
50min were calculated (rol/min/g see Fig. 25 and 26). Symbols
are as in Fig. 25. The lines were fitted by regression analysis
of values from individual perfusions, taking into account the
variation in both axes : (a) r = 0.77, p < 0.001 (22 obser-
vations) ;(b) r = 0.75, p< 0.001 (22 observations).
CHANGE IN % GLYCOGEN SYNTHETASE "a"
IN 30 MIN PERFUSION
0 Cn
0
CHANGE IN GLYCOGEN SYNTHETASE "a"
ACTIVITY IN 30 MIN PERFUSION
0
NO
ISIld
213c1
CH
AN
GE IN
PH
OSPH
ORY
LASE A
CTIVITY
C • •
165
For Fig. 28 see Over—Leaf
Fig. 28. • Changes in combined response of both glycogen
synthetase and phoshorla
net glycogen accumulation in the perfused
liver from adrenalectomised rats.
All details and symbols are given in Tables 22,
24 and 25, and Fig. 25. Values for the ratio % glycogen
synthetase "a" / phosphorylase, glycogen synthetase "a" /
phosphorylase were calculated for 20 and 50min liver samples,
and the differences between these values obtained. Bars
indicate S.E.M. and the number of perfusions in each group
are in Tables 25 and 26. Lines were fitted by regression
analysis of values from individual perfusions (a) r = 0.71,
p < 0.001 (22 observations); (b) r = 0.70,p < 0.001 (22
observations).
0 CHANGE IN ENZYME ACTIVITY IN 30 MIN_ PERFUSION
Increase in ( synthetase "a")/(phusphorylase)
2 co
I 1 I I I I I • I 06
Increase in ( o synthetase'crY(phosphorylase)
w.
Uol
iorI
LWID
DIO
O 1-0 O AD O 1.3
O
0
O
167
3.6 AMINO ACID BALANCE IN THE PERFUSED LIVER
1. Amino acid metabolism in fed, starved
and diabetic rat livers.
2. Urea formation in the perfused liver.
3. The role of the liver in amino acid
utilisation.
168
3.6 AMINO ACID BALANCE IN THE PERFUSED LIVER
Net rates of glycogen accumulation were obtained in the
perfused liver of starved rats at rates which occur in vivo (see
Section 3.1). In these experiments synthesis was studied over a
relatively short time period, in the absence of any evaluation of
protein status. Since many hormones are known to have effects on
protein synthesis (enzymes for example) it seemed appropriate that
conditions of protein balance be established in the perfusion so that
any relevance to the action of hormones could be evaluated. One phase
of this study was to assess amino acid release and uptake, at different
initial amino acid concentrations, in different nutritional and
hormonal states. These measurements are relevant to the understanding
of events in livers perfused with amino acids ("supplementation" :
Section 3.4.1), in which net rates of glycogen synthesis were measured,
and so are presented here.
3.6.1 Amino acid metabolism in fed, starved and diabetic
rat livers.
Changes in perfusate amino acids at four times "normal" concentrations
Livers from fed, 48h-starved or streptozotocin -diabetic rats
were perfused in the presence of 30mM glucose, gluconeogenic subStrates
(lactate, glycerol and pymuvate : infused) and amino acids (see Methods
Section 2.3.2). When fed rat livers were perfused in the presence of
amino acids at four times "physiological" concentrations, most were
taken up by the liver (for representative graphs see Fig. 29). Although
there was a delay in the onset of tyrosine uptake (probably due to its,
formation from phenylalanine), after 3h perfusion it reached an
equilibrium at an approximately normal plasma level (Fig. 29). The
0.6
0.3
Cl -U <{
'0 . Z 03 - .
=E <{
=E :::>
0.1 o· w =E
0.9
Fig. 29
TYR
-0 ~~
~00 .~
MET
I LE
~O:::-!b~ o~
°"'-0 ________ 0
ARG
GLU
20 60 120 180
TIME(MIN)
PHE
0.3
0.1
HIS
~
ALA. "/
-11. 3
0--
Metabolism of amino acids during perfusion at high
initial concentrations.
Livers of rats (180-200g; average liver weight about 7g)
were perfused as described in the text vTi th glucose plus a mixture
of lactate, glycerol and pyruvate. Initial amino acid concentration?
vTere about four times "normal" arterial. blood levels. Resul ts are
means of selected values from ~hree perfusions. . Livers vlere from
fed diabetic ( l1 ), 48h- starved normal (0) or fed nomal ( 0 ) rats.
169
170
"branched-chain" amino acids showed little change (for isoleucine,
see Fig. 29), and the amino acids, alanine and glutamic acid, were
transported out of the liver (Fig. 29).
In perfusions of livers from starved or diabetic rats,
the rates of uptake of the amino acids were in general faster than those
,for fed rats, particularly in the case of tyrdsine and lysine (Fig. 29).
Alanine was utilised by the starved rat liver in contrast to those
from fed or diabetic rats (Fig. 29); glutamate was released in both
groups. Glucose and ketone bodies were likely to be major products of the
amino acids used.
The uptake or release of "branched-chain" amino acids by
the liver provides a measure of protein change since they are not
significantly synthesised or utilised by other processes. In these
perfusions the initial concentrations were (approximate nM) : valine,
1200; isoleucine, 400 : leucine, 800. The concentrations in perfused
livers of fed or starved rats did not change significantly during 100min
perfusion whereas the levels increased by 10-20% during perfusion of
diabetic rat livers, suggesting that some proteolysis was occurring.
Release of amino acids during perfusion.
As was seen above, most amino acids (when present at four.
times "normal" concentrations) were taken up by the liver. It was
therefore of interest to evaluate what occurred when no amino acids
were present. In the case of livers from fed animals, most amino
acids normally found in blood appeared in the perfusate and reached
"normal" plasma levels (Table 26). Both glutamic acid and alanine were
however, released in substantial quantities and after 80-100min perfusion
the concentrations were double those of "normal" plasma (Table 26).
171
Table 26. The concentration of amine acids after perfusion
in the presence of "normal" levels or no added
amino acids
Livers were perfused as described in Fig. 29, or
text. Results are means of least 3 perfusions + S.D, except for results
given'for starved or diabetic rat liver perfusions which are means of
2 perfusions. Concentration of amino acids in plasma (1)
No added Amino acids initially "normal" amino acids
Fed Starved Diabetic Fed
Amino acids Final Initial Final
(80-100min.) (5 min) (80-100min)
ASP 69 + 6 68 21 + 4 <20 <20
ASN 44 + 15 130 95 + 20 - _
THRE 103 ± 18 181 62 ± 15 - -
SER 283 ± 20 350 110 + 25 - -
GLU .350 + 52 142 1110 + 156 140 <20
GLN 342 + 32 620 693 + 120 -
GLY . 140 + 34 55o 160 + 25 25 <20
ALA 692 + 82 374 1280 + 54 335 205
VAL 243 + 39 300 279 + 58 260 425
ILE 171 ± 25 125 100 + 25 85 185
LEU 260 14- 52. 266 275 + 30 200 295
MET <20 120 23 + 5 <20 <20
TYR 62 + 15 105 60 + 15 20 <20
PTIE 93 + 12 96 7o + 25 3o 45 HIS 87 + 16 85 65 + 12 20 '35
LYS 220 + 60 420 220 + 40 90 140
ARG <20 110 420 420 420
ORNITH 90 + 20 - - - -
NH3 841 + 75 - - - -
172
The "branched-chain" amino acids were also released, reaching
"physiological" levels, indicating that lysis of endogenous protein
was a major source of all the amino acids released (Table 26).
Livers from starved or diabetic rats released no significant
quantities of amino acids except for valine, leucine and isoleucine
which attained concentrations (results not shown) similar to those
found for the fed liver. Therefore, hepatic proteolysis, under these
conditions, was not altered by starvation or diabetes.
Chan es in erfusate amino acids at "normal" concentrations.
In order to fully assess protein balance in the perfusion,
amino acids were added to the perfusate medium, at "normal" plasma
concentrations. In the perfusions with fed rat livers most amino acids
were taken up (Table 26). Histidine, phenylalanine and glutamine did
not however, change significantly(Table 26). As was noted when four
times "normal" levels of amino acids were added to the perfusate, the
concentrations of the "branched-chain" amino acids, valine, leucine
and isoleucine remained unchanged and alanine and glutamic acid were
released by the liver (Table 26).
When the livers of starved rats were perfused, the amino
acid profiles were similar to those obtained with fed rat livers (Table
26). In general, there appeared to be greater uptake or less release of
amino acids e.g., the concentrations of alanine and glutamic acid did not
change significantly, whereas they were released in fed perfusions
(Table 26). The levels of "branched-chain" amino acids hardly changed.
All amino acids were removed by the diabetic rat liver except
for the "branched-chain" and alanine, which did not alter significantly
(Table 26).
173
3.6.2 Urea formation in the perfused liver
The urea concentrations were measured in the perfusate
plasma, to permit an assessment of nitrogen balance during perfusion.
In perfusions with no added amino acids, the urea level in perfusate
reached 1.5 - 2.4MM (Table 27). In the presence of "normal" levels of
amino acids the urea increased during perfusion of starved or diabetic
rat livers whereas it did not in fed perfusions (Table 27). When four
times "normal" concentrations were in the perfusate, urea increased in
all groups of perfusions, although the urea produced by starved or
diabetic rat livers reached higher levels (Table 27). In all
perfusions with four times "physiological" Amino acids there was a
rapid efflux of urea (Fig. 30), which stabilised during perfusion of
livers from fed rats, but continued to rise during 2h perfusions of
starved or diabetic rat livers.
The urea produced exceeded amino acid uptake (mols N)
in most perfusions (Table 28), except those in which there were four
times "normal" levels of amino acids where nitrogen balance was
attained (Table 28). In perfusions of liver from diabetic rats the
change in amino acid nitrogen exactly corresponded to the production of
urea nitrogen, whereas in those from fed or starved animals the
uptake of amino acids exceeded urea production (Table 28). This latter
finding could reflect the occurrence of hepatic protein synthesis.
174
Table 27. Effect of added amino acids on urea production
in the perfused liver.
Livers were perfused as described in Fig. 29
and in the text. Urea was measured after 80-100min perfusion. Results
are means ± S.E.M. with the number of observations in parenthesis.
EMLIEPLM112.121)
Initial amino acid concentration
Rat None added "Normal" 4 x "Norman'
Starved 1.5 ± 0.2 (3) 3.3 (1) 5.5 + 0.6 (3)
Fed. 1.9 (2) 1.8 (1) 3.7 ± 0.5 (4)
Diabetic 2.4 (2) 3.8 (2) 6.1 + 0.5 (3)
E LU
To
T
A
1 150 50
TIME ( MIN ) 100
Fig. 30. The effect of four times "normal" concentrations
of amino acids on urea production in the erfused
liver.
Livers ( ❑ , fed; 0,48h- starved and A diabetic)
were perfused as described in Fig. 29 and in the text. Samples
of perfusate were removed at the times indicated and assayed for
urea. Results are means of at least 3 observations and the bars
indicate S.E.M.
175
176
Table 28. Nitrogen balance in the perfused liver.
Amino acid change was calculated from measured
initial values and those after 80-100min perfusion; some of the
relevant changes are shown in Fig. 29 and Table 26. Glutamine and .
lysine were considered to provide two mols of nitrogen, histidine
three and arginine four. Urea change was calculated from the
concentration at 80-100min (there being none present initially);
see Table 27.
Unaccounted changes in N during perfusion (which may
reasonably be regarded as reflecting protein metabolism) are expressed
as -ve if amino acid uptake was less than urea production, i.e., if
proteolysis apparently exceeded protein synthesis (and +ve in the
converse case)
Rat
Change in amino- N (m mols)
Amino acid Urea Not accounted for as urea or amino acid.
Amino acids initially four times
Fed
Starved
8.7
14.8
7.4
11.0
+ 1.3
+ 3.8 "normal"
Diabetic 12.2 12.2 + 0.0
Amino acids initially
Fed 0.2 3.6 - 3.4
"normal" Starved 3.2 6.6 - 3.4
Diabetic 3.3 7.6 - 4.3
3.6.3 The role of the liver in amino acid utilisation
177
The present experiments demonstrate the influence of
different hormonal and nutritional states on the uptake of amino
acids by the liver, in conditions where an ample supply of circulating
carbohydrate is available. Breakdown of hepatic protein was minimal
in all groups of perfusions in the presence of amino acids at one or
four times"normal" concentrations, as shown by the insignificant
increases in "branched-chain" amino acids during perfusion. In the
absence of added amino acids, valine, leucine and isoleucine were
however, released by the liver to reach equilibrium at normal plasma
concentrations (present results; see also Mallette et al., 1969).
The presence of glucose and gluconeogenic substrates therefore
inhibited the proteolysis, (measured by "branched-chain" amino acids
released during perfusion) which occurs in the absence of added
carbohydrate, and in the presence or absence of amino acids (Bloxam,
1971). A marked efflux of "branched-chain" amino acids has also
however, been observed in the presence of added carbohydrate and
absence of amino acids (Schimassek & Gerok, 1965), which is inconsistent
with the present findings. One explanation could be the different
perfusion techniques.
The present results confirm that the perfused liver
can regulate the perfusate concentrations of amino acids by net
movement into or out of the cells according to the perfusate
concentrations (see also Bloxam, 1971). Two amino acids which did not
appear to be regulated by the liver were alanine and glutamic acid
which were consistently released during perfusion of livers from,
fed rats, in the presence or absence of added amino acids (see also
178
Schimassek & Gerok, 1965). Such an efflux did not occur when the
livers were from diabetic or starved rats, other than in the presence
of four times "normal" levels of amino acids. It would appear there-
fore that the release of glutamic acid at least, could be a
consequence of the liver glycogen rather than the carbohydrate in
the perfusate. This conclusion is sustained by the results of Mondon
and Mortimore (1967) who observed a notable glutamate release from
the perfused fed rat liver in the absence of added carbohydrate. The
release of alanine on the other hand, is probably a result of the
circulating carbohydrate substrates. It is likely that the efflux of
both glutamic acid and alanine observed from diabetic and starved
rat livers2in the presence of excess amino acids and gluconeogenic
substrates,was due to transamination of pyruvate and Krebs cycle
intermediates occurring faster than these amino acids could be utilised.
The observation however, that alanine was released by perfused livers
from diabetic (but not starved) rats isnot explicable on the known
characteristics of the diabetic state and could imply that alanine is
not an important gluconeogenic precursor in this state.
It has been suggested that there is a permeability barrier
to and from the liver cells for glutamate (Hems et al., 1968).. However,
the present results and those of Schimassek and Gerok (1965) and of
Mondon and Mortimore (1967) illustrate that glutamate can be released
by the liver, and that when present at physiological levels,in the
absence of added substrate2 can be rapidly taken up (Bloxam, 1971).
CHAPTER FOUit
DISCUSSION
180
CHAPTER FOUR
DISCUSSION
4.1 HEPATIC GLYCOGEN METABOLISM IN THE NORMAL RAT
1. The circulating precursors of hepatic glycogen .
2. The role of glucokinase in glycogen accumulation;
3. Control of hepatic glycogen synthesis.
4.2 THE ROLE OP INSULIN IN HEPATIC CARBOHYDRATE METABOLISM
1. Insulin and hepatic glycogen metabolism in the normal starved rat.
2. Hepatic glycogen accumulation in the starved diabetic rat.
3. Properties of glycogen synthetase and phosphorylase in the perfused liver of diabetic rats.
4.3 THP, ROLE OF ADRENOCORTICAL STEROIDS IN HEPATIC GLYCOGEN.
METABOLISM
1. Glucocorticoids and hepatic glycogen metabolism in the normal (starved) rat.
2. Hepatic glycogen accumulation in the starved adrenalectomised rat.
3. The characteristics of glycogen synthetase and phosphorylase in the perfused liver of adrenalectomised rats.
4.4 THE ROLE OF THE HORMONES OF TEE POSTERIOR-PITUITAZY.GLAND
IN REPATIC CARBOHYDRATE METABOLISM
1. The posterior-pituitary gland hormones and the metabolism of liver carbohydrate.
2. The mechanism of vasopressin action.
161
4.1 HEPATIC GLYCOGEN METABOLISM IN THE NORMAL RAT
4.1.1 The circulating precursors of hepatic glycogen
The use of an isolated liver preparation in which net
glycogen deposition occurred at physiological rates and in which all
aspects of glycogen metabolism apparently retained the features present
in vivo, has permitted the characterisation of glycogen metabolism in
rats. An initial question concerns the nature of the circulating
precursors of glycogen. The experiments reported (Section 3.1.2)
suggest that much of the hepatic glycogen that accumulates in the rat
on re -feedingl after a period of starvation,could be initially derived
from gluconeogenesis (defined as the new synthesis of glucose in either
mono- or poly saccharide form). This follows from the facts that the
fastest rates of glycogen deposition in the perfused liver were obtained
with gluconeogenic precursors in the perfusion medium as well as glucose,
and that there was no net uptake of glucose in this situation, despite
its high concentration (25-30MM). These characteristics of hepatic
glycogen synthesis may reasonably be presumed to exist in vivo, because
the rate of glycogen deposition, its dependence on glucose concentration,
and time course, were similar in the perfused liver and in the intact
rat.
It is reasonable to suppose that gluconeogenesis could have a
major role in the redeposition of hepatic glycogen after re-feeding
with material other than glucose. It is also possible that the
continuation of gluconeogenesis (from "endogenous" precursors) could
contribute to hepatic glycogen accumulation after the post-starvation
ingestion of glucose. In this situation, the increase in circulating
glucose concentration would relieve the liver of its function to release
free glucose into the blood, and the newly formed hepatic glucose
182
phosphates would be directed to form glycogen. The present experiments
suggest that this could be the initial sequence of events in vivo on
re-feeding with glucose. This possibility was suggested from experiments
on intact rats by Olavarria et al., (1968),* and was also raised by
Jeffcoate and Moody (1969) and Zaragoza -Hermans and Felber (1972).
Although the concentrations of circulating precursors in
the present experiments were high compared with those in the intact rat,
the findings may nevertheless be related to the starvation-re-feeding
situation in vivo. The question of particular interest is whether glucose
in the blood, after ingestion by a starved rat, is taken up by the liver
to form glycogen. In the perfusion experiments, glucose and gluconeogenic
precursors were present at near-saturating concentrations and the liver
exhibited a total preference for the gluconeogenic substrates. In the
intact 48h- starved-refed rat, although the blood concentrations of
gluconeogenic substrates are lower than were employed in perfusion),
so is that of glucose. Hence it is likely that in the initial phase of
re-feeding with glucose in vivo, the preference for gluconeogenic
precursors would prevail and that net glucose uptake by the liver would
be negligible.
During starvation, glucose tolerance is impaired in comparison
with the fed state. The present results suggest that, in the initial
phase of assimilation of a glucose load by tissues of starved animals,
there is no significant net uptake of glucose by the liver, although
there may be a marked diminution in the hepatic release of glucose into
blood, caused largely by the increased concentration of blood glucose
and perhaps by insulin. This conclusion is in accord with the main
explanation for the glucose intolerance of starvation proposed by Mahler
and Szabo (1970).
183
On ingestion after starvationl glucose is degraded by
extrahepatic tissues into potentially gluco genic precursors such
as lactate and alanine. These precursors, diluted in the general pools
of "endogenous" precursors, would eventually become incorporated into
liver glycogen. The extent of this process has been measured by
Friedmann et al., (1965), and was estimated at 20-40% of the glycogen
synthesis. This result is not incompatible with the present conclusion
that hepatic glycogen accumulation, after ingestion of glucose, may
initially derive largely from the continuing flux of "endogenous"
gluconeogenic precursors.
After a period of deprivation of dietary glucose, hepatic
metabolism adapts on eventual re-feeding with glucose, so as to decrease
the rate of gluconeogenesis. The present experiments suggest that
the initial phase of such an adjustment between "starved" and "fed"
states (defined with regard to dietary glucose) need not involve a
cessation of hepatic gluconeogenesis but rather its continuation for
perhaps several hours, contributing to the reaccumulation of glycogen.
During this time the activity of glucokinase in liver may slowly
increase (Salad et al., 1963; Sharma et al., 1963). In the event of
persistent repletion with dietary glucose, net uptake of glucose by the
liver could eventually supervene. Such a slow adaption to re-feeding
with glucose is likely to be of greater value to an animal than a more
rapid response to a possibly transient surge of glucose in the portal vein.
The existence of net gluconeogenesis in livers from fed rats
is difficult to demonstrate, in the presence of the usual associated
glycogen breakdown. The present experiments (Section 3.2.1) with a
glucogenic mixture, in which net glycogen breakdown was suppressed by
high glucose concentrations, showed however, that net gluconeogenesis
can occur in such livers and is not inhibited by high glucose concentrations.
184
The existence of gluconeogenesis in the fed state is in general
agreement with previous experiments with 14C-labelled gluconeogenic
precursors (Clark et al., 1974;EXton et al., 1972b)..
The results obtained with overnight-starved rats (Section
302.2).confirm the importance of gluconeogenesis in glycogen deposition,
although the influence of gluconeogenic precursors on hepatic glycogen
synthesis is much less marked and that of glucose more, than in 48h-
starved rats. In this condition there would not have been time for a
full adaptation of the enzyme3for the metabolism for 03-substrates,
which occurs on longer starvation. Even between 18 and 22h following
food deprivation, there was an increase in the capacity to form glycogen
from the mixture of glycerol, lactate and pyruvate. On the other hand,
the accumulation of glycogen in perfusions containing fructose showed no
difference between 18 and 48h- starved states, in agreement with the
general concept that adaptations in this part of the gluconeogenic
pathway are not as great as in the enzymes which metabolise pyruvate.
On considering the calculated rates of glucose release from
these livers, i.e., fed, 18h- starved and 48h- starved, adaptations to
starvation in the pathway of gluconeogenesis can be seen. The rate of
glucose release from the fed rat liver was 4.0iumol/min which is about
half the rate of gluconeogenesis in matched 48h- starved rats (about
8.0 prnol/min per liver). In overnight starved rats the rate was 5.6)1=1/
min per liver. This difference between fed and starved animals is of
the same order as that obtained when fed rats were depleted of liver
glycogen before liver perfusion (Ross et al., 1967). It would appear
therefore that there is some adaptation of the pathway of gluconeogenesis
to different nutritional states but that the capacity for the process is
always present. This is in keeping with the assayable activities of
regulatory enzymes of gluconeogenesis, and the extent of changes on e.g.,
starvation.
185
In general, the precursors of hepatic glycogen appear to
be gluconeogenic, the net contribution of glucose being minimal perhaps
even in the fed state.
4.1.2 The role of glucokinase in glycogen accumulation
Hepatic glucokinase activity is not always sufficient to
account for the uptake of glucose to form glycogen after the post-
starvation ingestion of glucose. It is known that glucokinase activity
falls on starvation and diabetes.(Salas et al., 1963; Walker & Rao, 1964),
although levels are sufficient to support some phosphorylation of glucose,
they would not allow net glycogen deposition. The experiment reported
14 1 \ with k C) -glucose in the perfusion medium (Section 3.1.2) showed that
glucose uptake (uni -directional, rather than net) did occur during
maximal glycogen synthesis from gluconeogenic precursors in the liver
from 48h- starved rats,and that glucose as a carbon source contributed
about one-third of the total carbon.
Alternative pathways of glucose uptake have been suggested
(see Friedmann et al., 1967; Ryman & Whelan, 1971) to circumvent the
low activity of glucokinase. The present experiments may however, resolve
this difficulty, in suggesting that gluconeogenesis could make a major
contribution to hepatic. glycogen synthesis on administration of glucose to
starved rats or insulin to diabetics. Thus there is no requirement
for glucokinase activity to be as rapid as the rate of glycogen deposition.
If indeed the net uptake of glucose by the liver to form
glycogen is of minor importance, even in fed non-ruminant mammals, the
question arises of the significance of the glucokinase enzyme. This
could reside in the control of glycogen metabolism by glucose, which
could involve phosphorylation of glucose (see Section 4.1.3).
186
4.1.3 Control of hepatic glycogen synthesis
The influence of increasing glucose levels has been
extensively studied in various experimental systems and it has been
found that glycogen synthesis is enhanced in a dose-dependent manner*
(see Introduction, Section 1.3). A similar result, in the presence of
gluconeogenic substrates, is reported here (Section 3.1.5).
On glucose ingestion after starvation the initiation of
glycogen synthesis is due to activation of glycogen synthetase (see
Hers et al., 1970, for review; Hornbrook, 1970) and inhibition of
phosphorylase (Hornbrook, 1970). The present results suggest that this
initiation of glycogen synthesis could largely be a direct consequence
of an increase in glucose concentration in the hepatic portal vein,
since maximal glycogen accumulation required the presence of glucose
in the medium, but no added hormones (Section 3.1.2). The rate of
total gluconeogenesis (glucose plus glycogen) was not significantly
altered by the presence of glucose in the medium (as was observed by
Exton & Park, 1967; Haft, 1967) which is in accord with the possibility
that glucose (or its metabolic products) may primarily activate the
glycogen synthetase system. Such activation of glycogen synthetase
(concomitant with an inhibition of phosphorylase activity) by glucose
(Buschiazzo et al., 197.0; Glinsmann et alo, 1970; Miller et al., 1973)
and fructose (Valli et al., 1974) has been demonstrated in the isolated
perfubed liver. It has also been demonstrated that by increasing the
glucose concentration a more extensive activation of glycogen synthetase
and inhibition of phosphorylase is obtained (Glinsmann et al., 1970).
It appears from the present results that gluconeogenic substrates,
including fructose,enhance the response of glycogen synthetase to glucose,
since there was less activation by glucose or gluconeogenic substrates
alone. The C3-substrates did not however, enhance the response of
187
phosphorylase to glucose (Section 3.1.5).
The "a" and "b" forms of glycogen synthetase and phosphorylase
are interconverted by protein kinases and phosphatases (see Introduction,
Section 1.2) and glucose appears to affect the activity of these glycogen
metabolising enzymes by increasing the activity of glycogen synthetase
phosphatase (DeUulf & Hers, 1967a & 1968b) and phosphorylase phosphatase
(Stalmans et al., 1970 & 1974a). Since glucose action on phosphorylase
phosphatase decreases the amount of phosphorylase "a" (active form of the
enzyme) this in turn will cause deinhibition of glycogen synthetase
phosphatase by phosphorylase "a" (Stalmans et al., 1971 & 1974a).
The question arises as to whether it is glucose specifically,
or a metabolite which initiates glycogen synthesis. The glucose
concentration-dependence of synthesis suggested that glucose phosphorylation
may be relevant in the initiation of synthesis, since it resembled that of
glucokinase (Walker & Rao, 1964). Consideration of the role of glucose in
inducing or initiating processes in other tissues may shed light on this
question. For example, studies on glucose-induced insulin release in the
pancreas may bear on this problem. The effect of glucose on insulin release
is critically dependent on extracellular concentrations of the sugar and it
was therefore suggested (Cerasi & Luft, 1970; Landgraf et al., 1971) that
pancreaticA5-cells possess a glucoreceptor system (which, through molecular
interaction with glucose at an appropriate concentration of glucose; leads
to activation of insulin release). It has also been proposed however, that
it is a glucose metabolite that activates insulin release (rather than glucose
itself). These hypotheses have been termed regulator site and substrate
(or metabolite) site, respectively (see Randle & Hales, 1972 for further
discussion). The initial suggestion for the operation of the substrate site
mechanism came from the observation that sugars readily metabolised by
mammalian tissues (for example
188
glucose, mannose and fructose) elicit insulin release (see MacDonald
et al., 1975). Sugars that are not readily metabolised (e.g.,
galactose and 5-0-methylglucose) or that are only phosphorylated
(2-deoxyglucose) did not stimulate insulin release (Coore & Randle,
1964; Grodsky et al., 1965). The regulator (glucoreceptor) site
hypothesis has obvious attractions, for it involves direct interaction
of glucose with a specific receptor. If such a receptor exists, it
would have a pattern of specificity however, which has not been found
among glucose transporting systems or enzymes. At present therefore,
evidence suggests that it is a metabolite of glucose (probably not glucose
6-phosphate), which activates insulin release.
As has been seen in the present work and noted above, some
phosphorylation of glucose occurs during glycogen synthesis. Since
the rate of glycogen accumulation in the perfused liver was dependent
on the glucose concentration over the range 10-30mM, which corresponds
to the !Km? (for glucose) of hepatic glucokinase (Di Pietro et al.,
1962; Walker & Rao, 1964), phosphorylation of glucose and perhaps
further metabolism of the hexose (as was proposed above with regard to
the stimulation of insulin secretion) may be implicated in the initiation
of hepatic glycogen synthesis by glucose. This would accord a role to
glucokinase in the regulation of glycogen metabolism. A number of
glucose-like substances (those readily metabolised, non-readily
metabolised or only phosphorylated) do not, however, substitute for
glucose in shortening the lag phase of glycogen synthetase activation in
vitro (DeWulf et al., 1970) or inactivate glycogen phosphorylase (Stalmans
et al., 1970). It has recently been shown that a few derivatives of
glucose do, however, stimulate the phosphorylase phosphatase reaction,
1, 5 -anhydroglucitol being the most effective (Stalmans et al., 1974b).
This indicates that it maybe glucose itself and not a metabolite, which
189
is involved in the activation of glycogen synthetase, inhibition of
phosphorylase and hence initiation of synthesis.
The concentration of glucose 6-phosphate in the cell has •
been frequently considered as an important factor in the control of
glycogen synthesis from UDPG. According to this hypothesis, the leyel
of glucose 6-phosphate in the liver would be increased after a glucose
load, as a direct result of hyperglycaemia, whereas the concentration of
UDPG would possibly be lowered. However, both metabolites decrease in
the liver of fed animals and the concentration of UDPG is diminished
whereas that of glucose 6-phosphate is already low, in the fasted mouse
(DeWulf & Hers, 1967a). These observations rule out glucose 6-phosphate
as a stimulatory agent. In vitro studies have also mitigated against
the importance of glucose 6-phosphate as an activator of glycogen
synthetase "a" and "b" and indicate that Pi is the more important enzyme
regulator (DeWulf et al., 1968). The fact that both UDPG and glucose
6-phosphate decrease after a glucose load indicate that a pull mechanism
exists such as would be obtained by a stimulation of the last steps of
glycogen synthesis. In the present work a decrease in the hepatic UDPG
concentration was observed when glycogen synthesis was maximal, although
there was an increase in glucose 6-phosphate (Section 3.1.5).This latter
finding is probably due'to the presence of gluconeogenic substrates in
the perfusion medium and suggests that in these conditions glucose 6-
phosphate could affect the activity of glycogen synthetase. For
example, the "b" form could be activated by glucose 6-phosphate, since
conversion to "a" is never complete (up to about 80% in the present
experiments).
190
It is not known how glucose directs hexose phosphates
to glycogen rather than glucose. It is known however, that a "pull"
on glycogen precursors would occur due to glucose-activated glycogen
synthetase, and it is possible that the inhibition of glucose 6-
phosphatase is involved (see Cahill et al., 1959; Ryman & Whelan, 1971).
It was found in the present work that when glycogen accumulation
occurred during perfusion there was less glucose release, but that the
total glucose formed (glucose and glycogen) at different initial glucose
concentrations, was constant (except when the glucose concentration was
40MM or there were no gluconeogenic precursors). This could indicate
that the perfusate glucose has to reach a certain concentration before
. the hexose phosphates are directed to glycogen in preference to glucose.
It is noteworthy that high concentrations of circulating glucose (40-50MK)
inhibited glycogen deposition although not gluconeogenesis : this could
.reflect an osmotic action of such a high solute concentration, or more
complex effects of glacose.Overall several aspects of the regulation of
hepatic metabolism by glucose remain unsolved (as in the case of
initiation of insulin secretion by glucose in the pancreatic /3 -cell)
191
4.2 THE ROLE OF IlibULIN IN HEPATIC CARBOHYDRATE METABOLISM
402.1 Insulin and hepatic glycogen metabolism in the •
normal starved rat.
There is controversy about the possible existence of a rapid,
direct action of insulin on the liver in post-starvation glycogen
deposition. In support of this possibility, hepatic glycogen synthetase
can be activated by insulin in certain conditions (Bishop et al., 1971),
glycogen synthesis can be prevented by anti-insulin serum (Steiner, 1964)
and insulin can stimulate glycogen synthesis from glucose in the isolated
perfused liver of starved rats (Sakai et al., 1958). It is established
that insulin can inhibit hepatic glycogen breakdown in the perfused
liver in some conditions (see Mondon & Burton, 1971, for references), as
it can in the intact animal (see Madison (1969) and Steele (1966)for
reviews). However, there is also evidence against a direct hepatic
action of insulin in the initial phase of post-starvation glucose
assimilation. For example, insulin does not activate glycogen synthetase
in the perfused rat liver (Glinsmann et al., 1970), and anti-insulin
serum does not alter the immediate hepatic fate of administered glucose
in rats (Moody et al., 1970). In rats with portal-caval anastomosis,a
combination of high circulating insulin an normal glucose concentrations
is associated with decreased hepatic glycogen synthesis (Assal et al.,
1971).
Since the immediate post-starvation synthesis of glycogen
appears to be largely a result of gluconeogenesis (even after glucose
ingestion), insulin, which can inhibit gluconeogenesis in the isolated
liver (Exton et al., 1970), would not be expected to stimulate this
process. In the present experiments with the perfused liver, glycogen
192
synthesis was rapid in the absence of added hormones, and insulin did
not influence glycogen synthesis under optimum conditions (Section 3.1.3).
Hence in the immediate post-starvation situation it appears unlikely
that circulating insulin exerts a major direct stimulatory effect on
liver glycogen synthesis. This is in general agreement with the
conclusions of Glinsmann et al., (1970), Hers et al., (1970) and
Moody et al., (1970). However, it is possible that insulin may exert
a moderate hepatic effect during post-starvation glycogen deposition,
in view of its known action in preventing glycogen breakdown. In the
present experiments insulin( albeit at a high concentration) produced
a moderate stimulation of glycogen deposition in suboptimal conditions,
i.e., low glucose concentration or no C3-substrates. Also, the
present data do not exclude a possible hepatic action of insulin in vivo
in combination with other hormones (e.g., in counteracting the glycogenolytic
action of glucagon Glinsmann & Mortimore, 1968; Mackrell & Sokal, 1969).
Regarding the effect of anti-insulin serum, obtained by Steiner (1964),
this could reflect extrahepatic events that indirectly resulted in a
glycogenolytic action on the liver, or the relatively long duration of
the experiment (3.5h), since the present results are not incompatible
with a longer-term action of insulin on liver glycogen metabolism. Evidence
discussed in the next section supports the first of these possibilities.
193
4.2.2 IleaaLticalysogen accumulation in the starved diabetic rat.
Imaientofheat3.c12 bhesisindiabetesrm . For
a variety of reasons (see Introduction, Section 1.4.1) it has not been
clear whether there is a significant and inherent alteration in
maximal rates of net hepatic glycogen accumulation in diabetes. The
streptozotocin-diabetic animal provides a suitable model for the
clarification of this problem (Junod et al., 1969). The changes in
glycogen metabolism described previously (Section 3.4) reflect the
consequences of insulin-deficiency diabetes, rather than hepatic toxicity
of streptozotocin since they were reversed by insulin and hexoses in
vivo.
The results (Section 3.4.1) show that there is a marked
inherent impairment in the maximum rate of net glycogen accumulation
in the liver of starved diabetic animals. In all groups of perfusions
of the livers of diabetic rats (in the absence of pre-treatment in vivo),
rates of net glycogen synthesis were low compared to those in normal
(starved) rats. This result confirms, by the measurement of net glycogen
deposition in optimal conditions, the inherent impairment of hepatic
glycogen synthesis in diabetes which has been inferred from measurements
of incorporation of 14C from 14
C- labelled precursors into glycogen in
the perfused liver (Exton et al., 1972b & 1973a) or in liver slices (for
reviews see Levine & Fritz, 1956;.Renold et al., 1956; Steiner, 1966).
In general, measurements of net glycogen synthesis in intact
diabetic rats (Friedmann et al., 1963 & 1967; Hornbrook, 1970; Longley
et al., 1957), do not show such a marked impairment of synthesis as that
revealed in the perfused liver. Thus the present perfusion experiments
demonstrate the extent of the potential decrease in the maximal rate of
glycogen synthesis in diabetes, that is inherent to the liver, and its
194
rapid reversibility (by treatment in vivo).
In experiments with intact diabetic animals, the decrease
in the rate of hepatic glycogen synthesis can be shown by measurements
of incorporation of 14C from C from C- labelled precursors into glycogen
(Friedmann et al., 1970; for reviews of earlier work, see Renold et
al., 1956; Steiner, 1966). The general implication of these
observations is that in diabetes, the hexose -phosphate products of
gluconeogenesis (produced at an increased rate) are directed towards
free glucose formation rather than glycogen (although the proportional
contribution of gluconeogenesis to glycogen formation is increased :
Stetten & Boxer, 1944). Results obtained in the perfused liver confirm
this suggestion, in that during maximal glycogen synthesis, the net carbon
sources of glycogen are gluconeogenic precursors, in either normal
(starved) rat livers (see Section 3.1.2) or in diabetic rat livers
(see Section 5.4.1). Net glucose uptake by the liver is not of significance
in diabetes. However, glucose must be present during perfusion to permit
glycogen deposition, even if net uptake of glucose does not occur. The
mechanisms underlying this requirement for glucose are not clear (see
also Seglen, 1974).
Studies of net glycogen metabolism in the liver of intact
diabetic animals show that as well as there being a defect in the rate
of glycogen accumulation (see Results, Section 3.4.4), there is a
limitation in the amount of glycogen which can be stored (Friedmann,
et al., 1967; Hornbrook, 1970). The present experiments did not however,
bear on this latter aspect of hepatic glycogen metabolism in diabetes.
Restoration of hepatic :1 cocren accumulation in diabetes.
Restoration of net glycogen synthesis in the perfused liver of
starved diabetic rats was achieved by two procedures in vivo : (i) pre-
' treatment with glucose plus fructose, (ii) insulin pretreatment (50-75min).
This effect of insulin would be expected, as it was repeatedly been shown
195
that the consequences of diabetes due to pancreatic A -cell toxins may
be reversed by insulin in vivo.
The results shed light on the origin of the inherent
alteration in hepatic glycogen synthesis rates in diabetes. Since
insulin in vitro produced no restoration of synthesis in perfusions
lasting up to 100min , but induced restoration within about 75min ,
if administered in vivo, it follows that a decline in the short-term
direct actions of insulin on the liver cannot be a major contributory
factor (Levine & Fritz, 1956). This conclusion is in agreement with the -
lack of action of insulin under optimal conditions for glycogen synthesis
in the perfused liver of normal (starved) rats (see Discussion, Section
4.2.1) and is not vitiated by the observation that insulin can suppress
the breakdown of glycogen in the perfused liver (see Exton et al., 1970)
or in hepatocytes (Akpan et al., 1974). Similarly in studies with 140-
labelled precursors, impairments in glycogen synthesis were not corrected
by insulin in the perfused liver (Exton et al., 1973a; Haft, 1968) or
liver slices (Renold et al., 1955).
Although the rats employed in the present experiments were
diabetic (based on the normal criteria), there was no necessity for
insulin maintainence, and thus they were not severely diabetic. However,
the impairment in hepatic glycogen synthesis was marked (albeit rapidly
reversible). The liver may therefore be sensitive (indirectly) to small
falls in plasma insulin (in the intact animal). It is likely that
starvation of the rats, in further lowering the circulating insulin level,
contributed to the development of the alteration in glycogen metabolism.
Thus it might be expected that the consequences of diabetes (insulin lack)
on glycogen metabolism in "fed" rats may be less consistent. These
factors might explain some of the. variability in resported effects of
insulin deficiency on hepatic glycogen metabolism.
J 196
The present experiments also show that hepatic glycogen
synthesis is not sensitive to a decline in direct long-term insulin action.
This follows because the alterations in diabetic animals were reversed
within 75min by procedures in vivo (tantamount to re-feeding), through
mechanisms which did not include direct hepatic insulin action.
Among possible factors acting in vivo, by which insulin may
affect the liver indirectly, the present experiments exclude any simple
role for adrenal corticosteroids (since hydrocortisone did not restore
synthesis of glycogen), or a fall in the blood concentration of glucagon
or free fatty acids, since prolonged perfusion in the absence of these
agents did not result in increased net glycogen accumulation. Hence
extra-hepatic effects or unknown co-factors appear to be implicated in
those rapid insulin actions which ultimately affect liver glycogen
synthesis in intact animals. These mechanisms are entirely obscure.
The restorative effect of hexoses on net glycogen synthesis
(measured during eventual perfusion) appears to involve a specific action -
of fructose. Thus the action of fructose plus glucose could not be
reproduced by an equivalent quantity of glucose, and anti-insulin serum
did not completely prevent their action. Also, in the intact diabetic
rat,glucose alone or plus a mixture of C3 -substrates, did not restore net
glycogen synthesis as effectively as did glucose plus fructose. This
effect of fructose is reminiscent of its action in diabetic rats in
restoring rates of fatty acid synthesis (Baker et al., 1952). The
mechanism of its action is likely'to include production of glucose (and
related metabolites) within the liver, since the liver is a main site of
fructose utilisation, and in diabetes this process is not impaired in the
same way as that of glucose assimilation (Chernick & Chaikoff, 1951;
Miller et al., 1952). Additional factors may be implicated in the
restorative effect in vivo of fructoseand glucose. One such factor is
197
likely to be insulin, which was effective over the same period
in vivo, and the restorative action of the hexoses was shown to be
partially prevented by anti-insulin serum. Insulin secretion, although
impaired, can be stimulated by administration of glucose to streptozotocin
-diabetic rats (Junod et al., 1969 )and it is known that fructose
may potentiate this action (Curry et al., 1972; Dunnigan & Ford, 1975).
4.2.3 Properties of almogmEalletase and phosphorylase
in the erfused liver of diabetic rats.
There could be various origins for the impairment in the
capacity for net glycogen synthesis in the liver of diabetic rats. The
present experiments suggest that the explanation may reside at least
partly in an inadequate responsiveness of the hepatic glycogen synthetase
system to substrates. Thus during perfusion of liars from diabetic rats,
in the absence of pre-treatment, there was an impairment in the increase
of glycogen synthetase "a" (active form of the enzyme : Herb et al.,
1970) which was observed in response to glucose plus 03-substrates in
normal rat livers. Similarly,if glucose alone is added to the medium
perfusing the liver of diabetic. rats (i.e., in conditions, which are not
conducive to maximal glycogen synthesis) impairment in the response of
synthetase and phosphorylase to glucose may be observed (Miller et al.,
1973). In the present experiments, the impairment in response of
synthetase to substrates in diabetes was not complete, since fructose (in
the presence of glucose) was able to activate synthetase in the perfused
liver of diabetic rats (in the absence of pre-treatment in vivo).
An impaired activation of glycogen synthetase was also demonstrated in
the intact rat, being most noticeable in the relative lack of response to
glucose plus fructose, compared to that in normal (starved) rats. This
is in general agreement with the results of Steiner (1964 & 1966),
Kreutner & Goldberg (1967) and Miller et al., (1973). The response of
198
phosphorylase to glucose and fructose was not impaired.
Insulin administration in vivo rapidly restored the sensitivity
of glycogen synthetase and phosphorylase to glucose plus C3-substrates
in the diabetic liver (as it restored the associated low rates of net
glycogen synthesis). This is unlikely to reflect a direct hepatic action
of insulin, as was shown above (Section 4.2.2); thus the high proportions
of synthetase "a" in the livers perfused with supplemented medium
(including insulin) probably involved the action of fructose, observed
in perfusions with fructose and glucose alone. Insulin effects on
glycogen synthetase in isolated liver preparations (in contrast to frog
liver : Blatt & Kim, 1971a) are in general not extensive (Akpan et al.,
1974; Miller & Larner, 1973). An exception to this generalisation is
that insulin can counteract the rapid (inhibitory) actions of glucagon or
adrenalin in the perfused liver on glycogen synthetase (Hostmark, 1973;
Miller & Lamer, 1973); this type of action may have a role in the rapid
effect of insulin in intact diabetic animals, in which circulating glucagon
concentrations are increased (Unger, 1972).
If direct, short-term insulin action on hepatic glycogen synthesis
(and synthetase)is not significant (in rats at least) then it follows that
the effect of insulin on hepatic glycogen synthetase in intact animals
(Bishop et al., 1971; Blatt & Kim, 1971b; Miller & Lamer, 1973; Nichols
& Goldberg, 1972) does not involve direct action. This conclusion endorses
in general that derived by Hers et al., (1970) from experiments in intact
mice. Similar considerations apply to the decrease in hepatic synthetase
phosphatase activity of diabetic animals (Bishop, 1970; Gold, 1970b);
this activity can be restored by insulin in vivo, but perhaps not as a
result of direct insulin action on the liver.
199
The behaviour of glycogen synthetase and phosphorylase in vivo
and in perfusion suggest that fructose can exert specific effects on
their activities (see Van delBerghe et al., 1973) and in restoring their
efficacy in the liver of diabetic rats. Thus fructose in combination
with glucose in the perfusate produced the highest proportional
activities of glycogen synthetase "a", in the liver of normal rats (in
vivo or in perfusion). Also, in diabetic rats, fructose plus glucose
in vivo resulted in the highest initial glycogen synthetase "a" in
subsequent perfusiohs containing fructose. Anti-insulin serum did not
completely prevent the action of fructose and glucose in vivo in
increasing the response of synthetase "a", during perfusion with C3-substrates,
suggesting that the effect of fructose (plus glucose) in correcting the
impaired sensitivity of synthetase and phosphorylase to substrates in
diabetes involves actions in vivo additional to the stimulation of insulin
secretion. These are likely to include specific hepatic effects, such
as a decline in the hepatic cyclic AMP content (which is raised in diabetes :
Pilkis et al., 1974) and an increase in hepatic fructose 1-phosphate contest;
both these changes occur in response to fructose, being relevant to the
associated decrease in glycogen phosphorylase activity (Thurston et al.,
1974; Van den Berghe et al., 1973).
The attainment of a.high proportion of synthetase "a" during perfusion
with fructose shows that the enzymic apparatus for glycogen synthesis is
not fundamentally lacking in the diabetic rat liver (see also Hornbrook,
1970). This also follows from the rapidity of the correction of impaired
glycogen synthesis by treatment in vivo. Yet, such activation by fructose
was not a sufficient condition for maximal net glycogen accumulation. Thus
in diabetes, the close relationship between synthetase "a" and the rate
of hepatic glycogen synthesis need not hold. A possible explanation for
this is the following. Maximal glycogen synthesis in the livers of normal
200
(starved) animals involves glucose and other carbon sources acting in
combination. Glucose, while not providing carbon for glycogen synthesis,
appears to exert an "initiatory" effect on synthesis by influencing the
fate of newly-synthesised hexose phosphate, i.e., in directing this
hexose to glycogen rather than free glucose. Presumably, in the diabetic
rat liver, a component of this overall process is impaired. This
deficiency cannot be in the process of gluconeogenesis, which is enhanced
in diabetes. Also, the effect of fructose in the perfused liver suggests
that the glycogen synthetase system is not defective. The defect which could not
be corrected in the perfused liver may therefore involve the initiatory or
cofactor role of,glucose, which is implicated in net glycogen deposition
in the normal liver and apparently in maintaining the full sensitivity of
synthetase (see Section 4.1.3). This would be in accord with the
established impairment in hepatic glucose uptake in diabetes (see Renold
et al., 1956; Steiner, 1966), and would also explain Why the hepatic
glycogen deposition in diabetes can be corrected more rapidly than is
accounted for by the restoration of glucokinase activity (Steiner, 1964)
since this regulatory role of glucose could involve free glucose (Hers
et al., 1970) as well as metabolic products. The nature of the glucose
effect on the liver is obscure; correction of this impaired glucose
effect in diabetes also involves factors which are not clear, including
insulin, but not solely (if at all) as a result of direct hepatic insulin
action.
201 •
4.3 TBE ROLE OF A METABOLISM
AL CORTICAL STEROIDS IN HEPATIC GLYCOGEN
4.3.1 Glucocorticoids and hepatic glycogen metabolism in the normal starved rat.
The stimulation of hepatic glycogen synthesis by glucocorticoids
was described in 1940 by Long, Katzin and Fry. Since then it has been hydro
shown that/col*;scale increases the conversion of circulating glucose into
liver glycogen (Ashmore et al., 1961; Gold & Segal, 1966), by activating
glycogen synthetase (DeUulf & Hers, 1967b, and 1968b) and inhibiting
phosphorylase (Stalmans et al., 1970).
In the present experiments with the perfused liver from
starved rats, glycogen synthesis was rapid in the absence of added hormones
and hydrocortisone did not alter synthesis under optimal conditions (see
Fig. 21, Section 3.4.1). This result is not however, incompatible with
glucocorticoids having a role in glycogen synthesis in the longer-term or
in vitro in suboptimal conditions, but suggests, in particular, that they
have no direct short-term effect on hepatic glycogeneds in normal rats.
4.3.2- Hepatic glYcogen accumulation in the starved
and adrenalectomised rat.
Impairment of hepgaticicoens -yadrenano.
Aspects of the role of adrenal cortical hormones in glycogen metabolism
are not fully clear. For example,'it is uncertain whether alterations
observed in hepatic metabolism,after adrenalectomy, are inherent to the .
liver.
The present results with starved adrenalectomised rats showed
that there was a marked inherent impairment in glycogen synthesis in all
groups of liver perfusions, even after pre-treatment in vivo with
hydrocortisone and or insulin, for periods of up to lh. Since the
202
metabolic changes due to adrenalectomy are not rapidly corrected (by
substrates in vivo, or even by the replacement of the missing hormone),
this hormone-deficient state would appear to be a more severe condition
than diabetes, where "normal" rates of glycogenesis were obtained after
lh treatment in vivo with glucose and fructose (see Discussion, Section
402. ).
This total loss of glycogen synthetic capacity represents a
much more severe alteration in hepatic metabolism following adrenalectomy
(plus starvation) than has been reported so far for any other metabolic
process (e.g., gluconeogenesis, ketogenesis, fatty acid metabolism) . The
impairment in gluconeogenesis,for example,which is relevant to glycogen
deposition in the present conditions (see Discussion, Section 4.1.) is
not severe,and is insufficient to explain the total loss of glycogen
synthetic capacity after adrenalectomy.
Hydrocortisone,in vitro, did not increase glycogen accumulation
in short or longer-term experiments,confirming the lack of a direct hepatic
effect on this process (see also Exton et al., 1973b). A similar lack of
response of glycogen synthesis to hydrocortisone was observed when the
livers of starved or diabetic rats were perfused. Although steroid-induced
glycogenesis has been postulated to be mediated by insulin (Exton et al.,
1973b, Hornbrook, 1970; Kreutner & Goldberg, 1967; Nichols & Goldberg,
1972), insulin did not have any effect when added in vitro. The nature of
the insulin effect in the intact animal is therefore obscure.
The marked impairment in glycogen accumulation observed in
perfusion was confirmed in vivo, showing that the impairMent was not a
phenomenon of perfusion as an experimental system or due to the omission in
of circulating factors. A similar impairment/in vivo glycogen synthesis
(although not glucose synthesis) from 140-labelled alanine or pyruvate,
has been observed (Friedmann et al., 1965). There is evidence however,
to suggest that glycogen synthesis is rapidly restored in the intact rat, for
203
after intragastric glucose, hepatic glycogen levels increase, although
not until 2-3h. Such "normal" rates of glycogenesis may fall off after
8h (Friedmann et al., 1967). It was therefore suggested that the defect
in adrenalectomy resides in the storage of glycogen rather than in its
synthesis.
Restoration of hepatic glycogen synthesis in adrenalectomy.
Restoration of net rates of hepatic glycogen accumulation in perfusion
were achieved by a number of procedures in vivo (a) pre-treatment with
hydrocortisone for 3-4h; (b) pre-treatment with hydrocortisone, fructose
and glucose for 2h or (c) pre-treatment with insulin, fructose and glucose
for 2h.
(a) The role of glucocorticoids in hepatic glycogen synthesis.
The restorative effect of hydrocortisone on hepatic glycogenesis was as
previously reported (Kreutner & Goldberg, 1967; Nichols & Goldberg, 1972).
However, since the hormone in vitro did not affect synthesis in perfusions
lasting up to 4h (treatment in vivo for this time restoring synthesis), /apparently
the restorative effect cannotbe due to a direct action on the liver (see
above).
time course of The/restoration of net rates of glycogen synthesis by steroid
was shortened by the addition of fructose and glucose during pre-treatment.
These hexoses were found to restore synthesis in the livers from diabetic
animals and it was suggested that fructose may have a specific action,
which could include stimulation of insulin secretion (see Discussion, Section
4.2. ). In adrenalectomised rats, insulin secretion is impaired (Malaisse
et al., 1967), the response of the pancreas to glucose being restored by
hydrocortisone. Since the adrenalectomised rats used in them studies
were starved for 48h prior to use, endogenous substrates would be depleted,
expecially since steroid lack would inhibit protein breakdown and thus
limit the availability of amino acids. Hydrocortisone treatment would cause
breakdown of protein in muscle (Long et al., 1940), and the amino acids
released would stimulate insulin secretion, (Floyd et al., 1966),
204
as well as providing a carbon source for gluconeogenesis. The
facilitation of glycogenesis by the hexoses was therefore probably
due to the fact that they were present initially in the pre-treatment
in the presence of hydrocortisone, so that insulin secretion could be
stimulated immediately (and possibly potentiated by fructose). A
specific role for fructose in the restoration of synthesis was not however,
investigated.
(b) The role of insulin in hepatic glycogen accumulation in
adrenalectomised rats. Several series of experiments have suggested that
glucocorticoid - activated glycogenesis is mediated by insulin (see
above) and that the impairment in hepatic glycogen accumulation in
adrenalectomised rats is a consequence of insulin deficiency. The present
experiments confirm this suggestion, in that insulin (plus hexoses to
prevent fatal hypoglycaemia) administered to the intact animal, restored
. the capacity for net hepatic glycogen synthesis. The rate obtained was
identical to that found after hydrocortisone and hexose treatment.
At 'first sight, it would be paradoxical for insulin deficiency
to provide the sole explanation for the total loss of glycogen synthetic
capacity since the defect in starved diabetic animals is less severe.
However, it is possible that insulin lack in 48h-starved adrenalectomised
rats is indeed more severe than in the matched diabetic animals used
previously (they were only moderately diabetic and did not require insulin
maintainence). On starving diabetic animals, glycogen mobilisation is
slow (Friedmann, et al., 1963) and hyperglycaemia correspondingly slow
to disappear; thus insulin may still be detected in blood of starved
streptozotocin-diabetic rats (Junod et al., 1969; Schein et al., 1971 )
whereas the concentration in starved adrenalectomised rats is very low
(Van Lan et al., 1974).
205
A near total lack of circulating insulin would not matter in
the starved adrenalectomised state since without adrenal glands serious
ketonaemia would not supervene, adrenal gland hormones having a cofactor
role in the provision of substrate for hepatic ketogenesis (Chernick et al 1972).
However, since negligible rates of synthesis were obtained
after insulin in vivo for ih, it would seem unlikely that the metabolic
changes due to adrenalectomy are only a result of insulin deficiency.
It would also appear that an unknown factor (or factors) is implicated in
the restoration of glycogen accumulation (which could be a site of lesion
in adrenalectomy), since no direct hepatic effect of insulin was observed.
4.3.3 The characteristics of glycogen synthetase and
phosphorylase in the perfused liver of
adrenalectomised rats.
The present experiments suggest that the impairment in hepatic
glycogen synthesis is due to an impairment in the responsiveness of both
hepatic glycogen synthetase and phosphorylase to substrates. Thus during
the perfusion of livers from adrenalectomised rats, in the absence of
pre-treatment in;vivo for more than lh, there was an impairment in the
increase in activity of glycogen synthetase "a" and fall in phosphorylase
seen in response to glucose and 03-substrates in the normal starved'
animal. A similar impairment in glycogen synthetase was observed in the
diabetic rat,although it was not as complete as the enzyme responded to
fructose (for Discussion see Section 4.2.). .
Loss of control of both glycogen synthetase and phosphorylase
by glucose has been previously reported in the perfused liver of 18h-
starved adrenalectomised rats (Miller et al., 1975). These enzymes did
however, respond normally in the fed adrenalectomised rat. A comparable
difference between the fed and starved state was observed in the activity
of glycogen synthetase phosphatase; the enzyme was reduced in the fasted
206 •
rat but not in the fed (Glinsmann, et al., 1970; Gruhner & Segal, 1970;
Mersmann & Segal, 1969). These results could implicate insulin in the
maintainence of glycogen synthetase and phosphorylase activity.
The action of steroids on hepatic glycogen phosphorylase activity
is not completely clear. As has been shown, the response of phosphorylase
to substrates (i.e., inactivation) is impaired on adrenalectomy (see also
Miller et al., 1973). This result suggests glucocorticoids act on
phosphorylase phosphatase, which is known to increase after steroid
administration in vivo (Stalmans et al., 1970). However, the activation
of hepatic phosphorylase by glucagon (Saitoh & Ui, 1975), or adrenalin
and cyclic AMP (Schaeffer et al., 1969),is lost on adrenalectomy,
implicating steroids in the conversion of "inactive" phosphorylase to
the "active" form i.e., at phosphorylase kinase. Measurements of hepatic
phosphorylase activity in the adrenalectomised rat have shown normal
levels of "active' enzyme, although there is a reduction in the "inactive"
form (Schaeffer et al., 1969). This reduction in activity of "inactive"
phosphorylase explains the lack of response of the enzyme to glucagon,
adrenalin and cyclic AMP. It would therefore appear that adrenal cortical
steroids are necessary for the maintenance of enzyme activity and are
involved in both the activation and inhibition of hepatic phosphorylase
activity.
Overall, glacocorticoids appear to be implicated in both sides of
glycogen turnover, in affecting the activities of both glycogen synthetase
and phosphorylase, in a manner which is not fully clear.
It has already been shown that the metabolic alterations in
hepatic glycogen metabolism due to adrenalectomy are not reversed rapidly,
even with hydrocortisone in vivo, and since there appears to be a total
impairment in the responsiveness of the enzymes glycogen synthetase and
phosphorylase, adrenal corticosteroid deficiency maybe considered as a
207
more severe state than diabetes.
Taken overall, the present results are compatible with the
notion that circulating insulin levels control the capacity for
hepatic glycogen synthesis, in relation to the dietary state, and to
pancreatic or adrenal endocrine status. This insulin effect appears
not be be di-rect on the liver, and its nature remains to be elucidated.
A similar problem exists for the short-term role of insulin in hepatic
ketogenesis (McGarry et al., 1.975).
208
4.4 THE ROLE OF THE HORMONES OF THE POSTERIOR-.
PITUITARY GLAND IN HEPATIC CARBOHYDRATE METABOLISM
4.4.1 Theos ard hormones and
the metabolism of liver carbohydrate.
The experiments described above (Section 3.3) show
that (8 -arginine) -vasopressin can stimulate the breakdown of liver
glycogen. This confirms the results obtained with higher concentrations
of vasopressin in liver slices (Heidenreich et al., 1963) and in the
perfused liver (Vaisler et al., 1965a) and the inference that was drawn
from experiments in vivo in which vasopressin was injected into the
hepatic portal vein (Bergen et al., 1960). Inhibition of glycogen
accumulation was observed in the perfusion and during intravenous
infusion'of vasopressin in vivo, showing that glycogen turnover (i.e.,
synthesis and breakdown) is affected by the hormone, and also that
vasopressin can affect hepatic glycogen metabolism in the intact animal.
It would seem likely that this prevention of glycogenesis in the starved
rat is due to the same action of vasopressin as in the atimulation of
glycogenolysis in the fed rat.
The effects of (8-arginine)-vasopressin observed in
the perfused rat liver were obtained with concentrations which can occur.
in vivo during "shock" or "stress" (see. Ginsburg, 1968), but especially
during acute haemorrhage (Forsling et al., 1971; Ginsburg & Heller, 1953).
Similar concentrations are necessary for the pressor action of the
hormone, exerted on extrahepatic tissues (see Altura, 1970; Dekanski,
1952). It is unlikely however, that the blood levels of vasopressin
in the unstressed animal are high enough to affect the liver. This may
also be the situation in regard to adrenalin action.
209
The action of vasopressin on hepatic carbohydrate metabolism
may be relevant to states of "shock" or "stress" for two different
reasons. Firstly, tissues require glucose during "stress", and the
present results show that vasopressin may contribute to the hyperglycaemia
usually associated with these states of "shock" (see Johnson, 1972),
by activating glycogenolysis and increasing gluconeogenesis. Secondly,
there are large quantities of water stored in conjunction with glycogen,
amounting to 2-4 times the weight of glycogen (Fenn, 1939). In a fed
rat of about 200g the hepatic glycogen content may be as much as ig and
the associated water up to 3m1. It is known that this water can be
mobilised e.g., during perfusion (Mortimore, 1961) or starvation (Herrera
& Freinkel, 1968). Therefore on stimulation of glycogenolysis by
vasopressin this hepatic water (which could amount to as much as one-third
of the plasma volume) would be released into the blood, without
prejudice to organ function. This action of vasopressin, in the
maintenance of plasma volume, would be especially relevant during, for
example, haemorrhage. During hypovolaemic stress, the hyperglycaemia
would serve to maintain extracellular fluid volume (including blood),
by osmotic action. These actions of vasopressin are therefore consistent
with the accepted role of the hormone in the preservation of plasma
volume, composition and pressure.
More generally, the action of vasopressin on the liver represents
a third main function of this hoimione in the rat, in addition to its
antidiuretic and pressor effects.
Since (8-arginine)-vasopressin caused glycogen breakdown, and
also moderate stimulation of gluconeogenesis (in livers from starved rats
that were perfused in the absence of added glucose), its actions on the
liver resemble those of glucagon (Exton et al., 1970). The minimum
210
effective circulating concentrations of these hormones (in the rat)
are of the same order, i.e., about 100pg/M1, as 11. units of (8-arginine)
-vasopressin corresponds to 2.5pg (for discussion of hepatic sensitivity
to glucagon see &ton et al., 1970; Sokal, 1966). As with glucagon (Pauk
& Reddy, 1971), the liver of the starved rat appeared to be more
sensitive to vasopressin than the liver of the fed rat.
The results obtained with oxytocin confirm those obtained
in vitro by other workers (Heidenreich et al., 1963; Vaisler et al.,
1965b), although Heidenreich et al., (1962) found no effect of oxytocin
in vivo, on the blood glucose concentration of the rat (presumably
reflecting the insensitivity of the liver to this hormone). Since
massive doses of the hormone were required to inhibit glycogen
accumulation in the perfusion, this action probably has no physiological
significance.
Hepatic and vascular responses to hormones of the posterior-
pituitary gland may be compared. The, two tissues exhibit (i) comparable
sensitivity to (8-arginine)-vasopressin (both responses .being much less
sensitive than that of the kidney) and (ii) insensitivity to oxytocin.
These observations suggest that the receptors to vasopressin in blood
vessels and liver may resemble each other in their characteristics, but
perhaps do not resemble the renal receptor to the hormone.
4.4.2 The mechanism of vasopressin action
In the kidney, vasopressin can stimulate glycogen breakdown
(Darnton, 1967) and gluconeogenesis (Stumpf et al., 1972). Since both
the renal action of vasopressin (Beck et al., 1971) and the hepatic
action of glucagon and possibly adrenalin (Exton et al., 1970) may be
211
mediated by an increase in the tissue concentration of cyclic AMP,
it appeared possible that the action of vasopressin on liver
carbohydrate metabolism would be similarly mediated. However, hepatic
cyclic JP does not rise in response to vasopressin (Kirk and Hems,
1974), so that a different "second messenger" may be implicated, in
vasopressin action.
A small increase in hepatic phosphorylase activity was
observed when vasopressin was administered in vivo. The activation of
the enzyme was not as significant as that obtained with glucagon, under
the same conditions, although the response and time course of activation
were similar to those obtained with adrenaline This provides a further
similarity in the actions of vasopressin andadrenalin on the rat liver
in addition to the lack of a role of cyclic AMP; thus the increase in
this "messenger" due to adrenalin, in comparison with that produced
by glucagon, is small (Kirk & Hems, 1974). It has been suggested that
adrenalin action on hepatic glucose output (cc receptor-mediated effect)
probably does not involve cyclic AMP as a messenger (Sherline et al.,
1972; Tolbert & Fain, 1974; Tolbert et al. 1975).
An observed increase in hepatic phosphorylase activity could
in general be due to either of two mechanisms : an increase in protein
kinase or a decrease in phosphorylase phosphatase activity. However,
in keeping with its lack of effect on hepatic cyolic AMP, vasopressin
(administered in vivo) does not activate hepatic protein kinase activity
(Keppens & DeNulf, 1975). Glucagon (Keppens & DeWulf, 1975) and
adrenalin (see Butcher et al., 1972) do, however, increase protein kinase,
the rise being more marked with glucagon than with adrenalin; this is in
agreement with the different increases in cyclic AMP concentrations
known to be produced by these hormones (Exton et al., 1971). The effect
212
of vasopressin on phosphorylase phosphatase has not been assessed.
It is known however/ that stimulation of the splanchnic nerve causes
activation of hepatic phosphorylase without any increase in cyclic AMP
(Shimazu & Amakawa, 1968 & 1975). It has been shown that the activity
of phosphorylase kinase is not altered, but that the activity of
phosphorylase phosphatase decreases promptly and markedly, on
splanchnic nerve stimulation. Glucagon and adrenalin have no significant
effect. From such circumstantial considerations,it seems plausible that
vasopressin could increase hepatic phosphorylase activity by inhibiting
phosphorylase phosphatase. This remains to be tested.
The metabolic actions of both vasopressin and adrenalin can
occur in the absence of any change in total hepatic blood flow, but it
may be that an explanation for the activation of phosphorylase by these
hormones lies in hypoxia produced by redistribution of blood flow, away
from the periphery of the lobes. Such a conclusion was drawn from
experiments showing that adrenalin inhibited hexobarbitone metabolism
by the perfused liver although it did not in rat liver slice studies
(Boobis & Fowls / 1974).
Glucagon, adrenalin and vasopressin are hormones which cause
hepatic glycogenolysis by increasing phosphorylase activity, and increase
gluconeogenesis. It does appear however, that although they have common
effects they all act via different mechanisms :. glucagon acting on
protein kinase activity by increasing cyclic AMP/ vasopressin possibly
decreasing phosphorylase phosphatase activity, and adrenalin acting via
local
hypoxia (cyclic AMP independent route). Many aspects of these
actions clearly still require elucidation.
213
REFERENCES
Akpan, J.0., Gardner, R.& Wagle, S.R. (1974) Biochem. Biophys. Res. Commun. 61, 222
Altura, B.M. (1970) Amer. J. Physio1.211, 222
Appleman, M.M., Krebs, E.G. & Fischer, E.H. (1966) Biochemistry 5, 2101
Ashmore, J., Stricker, F., Love, W.C. & Kilsheimer, G. (1961) Endocrinology 68, 599
Assal, J.P., Levrat, R., Stauffacher, W.& Renold, A.E.. (1971) Metab. Clin. Exp. 20, 850
Baker, N., Chaikoff, I.L. & Schusdek, A. (1952) J. Biol. Chem. 194, 435
Ballard, F.J. & Oliver, I.T. (1964) Biochem. J. 92, 131
Baron, D.N. & Roberts, P. (1963) J. Physiol. (London) 165, 219
Beck, N.P., Kaneto, T., Zor, U., Field, J.B. & Davis, B.B. (1971) J. Clin. Invest. 50, 2461
Bergen, S.S., Sullivan, R., Hilton, J.G., Willis, S.W. & Van Itallie, T.B. (1960) Amer. J. Physiol. 129, 136
Birch, G.G., Lee, E.Y.C. & Hems, D.A. (1974) Int. J. Biochem, 5, 867
Bishop, J.S. (1970) Biochem. Biophys. Acta 208, 208
Bishop, J.S. & Lamer, J. (1967) J. Biol. Chem. 242, 1354
Bishop, J.S. & Larner, J. (1969) Biochem. Biopys. Acta 171, 374
Bishop, J.S., Goldberg, N.S. & Lamer, J. (1971) Amer. J. Physiol. 220, 499
Blatt L.M. & Kim, K-H.(1971a) J. Biol. Chem. 246, 4895
Blatt, L.M. & Kim, K-H.(1971b) J. Biol. Chem. 2462. 7256
Bloxam, D.L. (1971) Br. J. Nutr. 26, 393
Boissonnas, R.A., Guttmnnn, St., Berde, B.& Konzett, H. (1961) 12ERILLI, 377
Boobis, A.R. & Fowls, G.(1974) Biochem. Pharmacol. 23, 3377
Burn, J.H. (1923) J. Physiol. London) 6/1, 324
Buschiazzo, H., Exton, J.H. & Park, C.R. (1970) Proc. Nat. Acad. Sci. U.S.A. 65, 383
214
Butcher, R.W., Robi.son, G.A. & Sutherland, E.W. (1972) in: Biochemical Action of Hormones (G. Litwack ed.) Vol. 2, p.219 Academic Press, New York and London.
Cahill, G.F., Hastings, A.B., Ashmore, J.& Zottu, S .(1958) J. Biol. Chem. 230, 125
Cahill, G.F., Ashmore, J., Renold, A.E. & Hastings, A.B. (1959) Amer. J. Med. 26, 264
Cash, W.D. & Kaplan, M.H. (1964) EndocrinolUY 74, 803
Cerasi, E. & Luft, R. (1970) Horm. Metab. Res. 2, 246
Chernick, S.S. & Chaikoff, I.L. , 1951) J. Biol. Chem. 188, 389 Chernick, S.S., Clark, C.M., Gar =ex, R.J. & Scow, R.U.C1972) Diabetes Clark, G.A. (1928) J. Physiol. (London) 64, 324 21, 946
Clark, D.G., Rognstad, R.& Katz, J. (1974) J. Biol.222Chen42, 2028
Claude, H.& Baudouin, A. (1912) C.R. Soc. Biol. 7,3, 732.
Coore, H.G. & Randle, P.J. (1964) Biochem. J1 93L 66
Cori, G.T. & Cori, C.F. (1940) J. Biol. Chem. 135, 733
Cori, C.F.,Cori, G.T. & Green, A.A. (1943) J. Biol. Chem. 151, 31
Curry, D.L., Curry, K.P. & Gomez, M. (1972) Endocrinology •12 1493
Darnton, S.J. (1967) 211aELI.jup. Physiol. 52, 392
Das, I.& Hems, D.A. (1974) Horm. Metab, Res. 6, 40
Dekanski, J. (1952) Brit. J. Pharmacol..1, 567
DeWulf, H. (1971) The Control of Glycogen Metabolism. Ph.D., Thesis. Vander, Louvain
DeWUlf, H. & Hers, H.G. (1967a) Ear. J. Biochem. 2, 50
DeWalf, H. & Hers, H.G. (1967b) Ear. J. Biochem. 2, 57
DeNUlf, H. & Hers, H.G.- (1968a) Ear. J. Biochem, 6, 552
DUalf, H. & Hers, H.G.: (1968b) Ear. J. Biochem. 6, 558
DeWalf, H., StaImans W.& Hers, H.G. (1968) Ear. J. Biochem. 6, 545
DeValf, H., Stalmans, W.& Hers, H.G.. (1970) Ear. J. Biochem. 15,, 1
DiPetro, D.L., Sharma, C.& Weinhouse, S. (1962) Biochemistry. 1, 455
Dunnigan, M.G. & Ford, J.A. (1975) Endocrinol. Metab. 40, 629
Eisenstein, A.B., Spencer, S., Flatness, S.& Brodsky, A. (1966) Endocrinology 79,.182
215
Exton, J.H. & Park, C.R. (1965) J. Biol. Chan. 240, 955
Exton, J.H. & Park, C.R. (1957) J. Biol. Chem. 242, 2622
EXton, J.H. & Park, C.R. (1968) J. Biol. Chem. 243, 4189
Exton, J.H., Mallette, L.E., Jefferson, L.S., Wong, E. H-A., Friedmann, N., Miller, T.B. & Park, C.R. (1970) 2.,...ecentPiorm. Res. 26, 411
Exton, J.H., Robison, G.A., Sutherland, E.W. & Park, C.R. (1971) J. Biol. Chem. 2461 6166
Exton, J.H.9 Friedmann, N., Wong, E.H-A., Brineaux, J.P., Corbin, J.D.. &.Park, C.R. (1972a) J. Biol. Chem. 24, 3579
Exton, J.H., Corbin, J.G. & Harper, S.C. (1972b) J. Biol. Chem. 247, 4996
Exton, J.H., Harper, S.C., Tucker, A.D. & Ho, R-J. (1973a) Biochem. Biophys. Acta. 329, 23
Exton, J.H., Harper, S.C., Tucker, A.L., Flagg, J.L. & Park, C.R. (1973b) Biochem. Biophys. Acta 329, 41
Fenn, W.O. (1939) J. Biol. Chem. 128, 297
Fischer, E.H., Pocker, A. & Saari, J.C. (1970) Essays in Biochemistry 6, 23
Floyd, J.C., Fajans,S.S., Conn, Knopf, R.F. & Hull, J. (1966) J. Clin. Invest. 45, 1487
Forsling, M.L., Martin, M.J. & Burton, A.M. (1971) J. Endocrinol, 51, 413
Friedmann, B., 238, 2899
Friedmann, B., 3729
Goodman, E.H. & Weinhouse, S. (1963) J. Biol. Chem.
Goodman, E.H. & Weinhouse, S. (1965) J. Biol. Chem. 210
Friedmann, B., Goodman, E.H. & Weinhouse, S. (1967) Endocrinology 81, 486
Friedmann, B., Goodman, E.H. & Weinhouse, S. (1970) Endocrinology 86, 1264
Gilboe, D.P. & Nuttall, F.Q., (1973) Biochem. Biophys. Res. Commun. a, 164
Gilboe, D.P. & Nuttall, F.Q. (1974) Biochem. Biophys. Acta 338, 57
Ginsburg, M. (1968) Handb. EXp. Pharmacol. 23, 286
Ginsburg, M.& Heller, M. (1953) J. Endocrinol. 9, 274
Glinsmann, W.H. & Hern,E.P. (1969) Biochem. Biophys. Acta 36, 931
Glinsmann, W.H. & Mortimore, G.E. (1968) Amer. J. Physiol. 215, 553
216
Glinsmann, W.H., Hern, E.P. & Lynch, A. (1969) Amer. J. Physiol. 216, 698
Glinsmann, W.H., Pauk, G. & Hem, E./1(1970) Biochem. Biophys. Res. Commun. 39, 774
Gold, A.H. (1970a) Biochemistry 9, 946
Gold, A.H. (1970b) J. Biol. Chem. 245, 903
Gold, A.H. & Segal, H.L. (1966) Mol. Parmacol. 2, 84
Good, C.A., Kramer, H. & Somogyi, M. (1933) J. Biol. Chem. 100, 485
Grodsky, G.M., Batts, A.A., Bennett, L.L., Vbella, C., McWilliams, N.B. & Smith, D.P. (1963) Amer. J. Physiol. 205, 638
Gruhner, K. & Segal, H.L. (1970)1.BI ta222Biochen , 508
Haft, D.E. (1967) Amer. J. Physiol. 213, 219
Haft, D.E. (1968) Diabetes 17, 244
Heidenreich, 0., Kook, Y.& Reus, E. (1962) Naunyn-Schmiedeberg's Arch. Exp. Pathol. Pharmakol. 243, 136
Heidenreich, 0., Kook, Y., Baumeister, L. & Keller, P. (1963) Naunyn-Schmiedebergls Arch. Exp. Pathol. Pharmakol. 245, 321
Hems, D.A. & Brosnan, J.T. (1970) Biochem. J. 120, 105
Hems, R., Ross, B.D., Berry, M.N. & Krebs, H.A. (1966) Biochem. J. 101, 284
Hems, R., Stubbs, M. & Krebs, H.A. (1968) Biochem. J. 107, 807
Herrera, E. & Preinkel, N. (1968) Biochem. Biophys. Acta 170, 244
Hers, H.G., Deliulf, H., Stalmans, W & Van den Berghe, G. (1970) Advan. Enzyme Regu1. 8, 171
Hildalgo, J.L. & Rosell-Perez, M. (1971) Rev. Esp. Fisiol. 27, 343
Hizukuri, E. & Larner, J. (1964) Biochemistry 3, 1783
Hohorst, H.J. (1963) in: Methodsof H.U., ed.) p266, Academia Press, New York
Hornbrook, K.R. (1970) Diabetes 19, 916
Hornbrook, K.R., Burch, H.B. & Lowry, 0.H. (1965) Biochem. Biophys. Res. Commun. 18, 206
Hornbrook, Burch, H.B. & Lowry, 0.H. (1966) Nol. Pharmacol. 22. 106
217
Hostmark, A.T. (1973) Acta physiol. Scan& 88, 248
Illiano G., Tell, G.P.E., Siegel, M.I. & Cuatrecasas, P. (1973) Proc. Nat. Acad. Sci. U.S.A. 70, 2443
Imrie, C.G. (1929) J. Ph. siol. (London) 67, 264
Jeffcoate, S.L. & Moody, A.J. (1969) Diabetologia. 5, 293
Jefferson, L.S., Exton, J.H., Butcher, R.W., Sutherland E.W. & Parks C.R. (1968) J. Biol. Chem. 243, 1031
Jeffrey, P.L., Brown, D.H. & Illingworth Brown, B. (1970) Biochemistry 9, 1403
John, D.W. & Miller, L.L. (1969) J. Biol. Chem. 244 6134
Johnson, I.D.A. (1972) Advan. Clin.2221:15, 255 Junod, A., Lambert, A.E., Stauffacher, W. & Renold, A.B. (1969) J.
Clin. Invest. 48, 2129
Keppens, S. & DeWulf, H. (1975) F.E.B.S. Letters_21, 29
Kirk, C.J. & Hems, D.A. (1974) F.E.B.S. Letters .AZ, 128 Kirk, C.J., Verrinder, T.& Hems, D.A. (1975) Biochem. Soc. Trans.
.3.2. 515
Krebs, H.A. & Henseleit, K. (1932) itopzzSELefsaphmi.o1z....Chem. 11-22' 33
Krebs, H.A.t Dierks, C. & Gascoyne, T. (1964) Biocham.11....21 112 Kreutner, W. & Goldberg, N.D.(1967) Proc. Nat. Acad. Sci. U.S.A. ..1515
Landau, B.R. (1965) Vitam. Horm. New York) 23, 1
Landgraf, R., Kotler-Brajtburg, J.& Hatschinsky, F.M. (1971) Proc. Nat. Acad. Sci. U.S.A. 8, 536
Larner, J. & Villar-Palasi, C. (1971) Current Topics in Cell. Reg. 3, 195
Lee, E.Y.C. & Whelan, W.J. (1966) Arch. BiochenL..131ophya...116, 162
Lejeune, N., Thines-Sempoux, D.& Hers, H.G. (1963) Biochem. J. 86, 16
Leloir, L.F., (1964) in: Ciba Foundation Suiposium on Control of Glycogen Metabolism (W.J. Whelan & M.P. Camerc7r7e77777.18 Churchill, London
Leloir, L.F. (1967) E-LLsp3229LD....isii.Monorah2E2 1, 3
Leloir, L.F. & Carlini, C.F. (1957) J. Amer. Chem. Soc. 79, 6340
Leloir, L.F. & Goldemberg, S.H. (1960) J. Biol. Chem, 23, 919
218
Little, J.B., Klevay, L.M.,Radford, E.P. & McGandy, R.P. (1966). Amer. J. Physiol. 211, 786
Levine, R. & Fritz, I.B. (1956) Diabetes 5, 209
Long, C.N.H., Katzin, B. & Fry, E.G. (1940) Endocrinology 26, 309
Longley,R.W., Bortnick, R.I. & Roe, J.H. (1957) Proc. Soc. Exp. Biol. Med. 94, 108
MacDonald, M.J.,Ball, T.N.Lauris, V. & Steinke, J. (1975) Biochem. 188
Mackrell, D.J. & Sokal, J.E. (1969) Diabetes 18, 724
Maddaiah, V.T. & Madsen, N.B. (1966) J. Biol. Chem. 241, 3873
Maddaiah, V.T. & Madsen N.B. (1968) Can. J. Biochem. 46, 521
Madison, L.L. (1969) Archt ZE:berIrlad.123" 284
Mahler, R.J. & Szabo O. (1970) latLLIJILD1L Eat 11, 271
Malaisse, W.J., Malaisse-Lagae, F.,McCraw, E.F. & Wright, P.H. (1967) Proc. Soc. E2g. Biol. Med. 124, 924
Nanette, L.E., Exton, J.H. & Park, C.R. (1969) J. Biol. Chem. 244, 5724
Mansford, K.R.L. (1967) Diabetes 16, 475
Marsh, W.H., Fingerhut, B. & (1965) 21122 21.,..ien21.222, 624
McGarry, J.D. Wright, P.M. & Foster, D.W. (1975) J. Clin. Invest. Z,. 1202
Merlevede, W., Goris, J. & DeBrandt, C. (1969) Ear. J. Biochem. ILL 499
Mersmann, H.J. & Segal, H.L. (1967) Proc. Nat. Acad. Sci. U.S.A. 211. 1688
Mexsmann, H.J. & Segal, H.L. (1969) 1701
Miller, L.L.,Bly, C.G., Watson, M.L. & Bale, W.F. (1951) L...}22,c laa,21, 431
Miller, M., Drucker, W.R., Owens, J.E., Craig, J.W. & Woodward, H. (1952) J. cliilLaampIl 51, 115
Miller, T.B. & Larner, J. (1973) J. Biol. Chem. 248, 3483
Miller, T.B., Hazen, R. & Larner, J. (1973) Bi_o12111.Bi.ophystRes. Common. 7, 466
Mills, G.T. & Smith, E.E.B. (1963) in: Methods of Ersti-acAnalys1- (Bergmeyer, H.U. ed. ) P . 581, Academic Press, London
Mondon, C.E. & Burton S.D. (1971) Amer. J. phapiol. 220, 724
219
Mondon, C.E. & Mortimore G.E. (1967) 173
Moody, A.J., Jeffcoate, S.L. & Volund, A. (1970) Horm. Metab. Res. 22. 193
Mortimore, G.E. (1961) Amer. J. Physiol. 201 1315
Nichols, W.K. & Goldberg, N.D. (1972) BiocheiBiohm"42-22ba2, 245
Nordlie, R.C. (1968) in: Control of Glycogen Metabolism (W.J. Whelan ed.) Universitets for laget, Oslo, p. 1531Academic Press, London,
Nuttall, P.Q., (1972) in: Handbook of Physiolam, Section 7 (Endocrinology, Greep, R.O. & Astwood, E.B., eds.) Vol. 1. p.395, American Physiological Society.
Olavarria,J.M., Godeken, 0.G.R., Sandruss, R. & Flawia, M. (1968) Biochem. Biophys. A2ta165, 185
Porter, K.R. & Bruni, C. (1959) CancerEu: la, 997
Pauk, G.L. & Reddy, W.J. (1971) Diabetes 20, 129
Pilkis, S.J. & Park, C.R. (1974) Ann. Rev. Pharmacol.....1i, 365
Pilkis, S.J., Exton, J.H., Johnson, R.A. & Park, C.R. (1974) Biochem. Biophys. Acta 343, 250
Rall. T.W. & Sutherland, E.W. (1958) J. Biol. Chem. 212., 1065
Randle, P.J. & Hales, C.N. (1972) in: HandbaokaLEILliala, Section 7 (Endocrinology, Greep, R.O. & Astwood, E.B. eds.) Vol. 1 p.219,American Physiological Society
Renold, A.E., Teng, C-T., Nesbett, F.B. & Hastings, A.B. (1953) J. Biol. Chem. 204, 533
Renold, A.E., Hastings, A.B., Nesbett, F.B. & Ashmore, J. (1955) J. Biol. Chem. 213, 135
Renold, A.E., Ashmore, J. & Hastings, A.B. (1956) Vitam. Horm. New York) 14, 139
Ross, B.D., Hems. R., Freedland, R.A. & Krebs, H.A. (1967) Biochem. 869
Ruderman, N.B. & Herrera, M.G. (1968) .limer:L.layIsio1„...2141 1346
Ryman, B.E. & Whelan, W.J. (1971) Advan. Enzvmol. Rel. Areas Mol. Biol. 34, 285
Saitoh, Y. & Ui, M. (1975) B5ochem. Biophys. Acta 404.2. 7
Salas, M., Vinuela, E. & Sols, A. (1963) .L1uLshm,2322 3535
Salmon, D.M.W., Bowen, N.L. & Hems, D.A. (1974) Biochem. J. 142, 611
Salsas, E. & Lamer, J. (1975) J. Biol. Chem. 250, 1833
220
Schaeffer, L.D., Chenoweth, M. & Dunn, A. (1969) Biochem. Biophys. Acta 192, 292
Scharff, R. & Wool, I.G. (1966) Biochem. J. !II, 173
Schein, P.S., Alberti, K.G.M.M. & Williamson, D.H. (1971) Endocrinology 8g, 827
Schillinger, E., Loge, E., Schroder, E., Klieger, E. & Lubke, K. (1972) EuIt Bj.ocl11.2 473
Schimassek, H. (1963) 131.ochenit21 336, 460
Schimassek, H.& Gerok, W. (1965) Biochem. Z. 343, 407
Seglen, P.O. (1974) Biochem. Biomes. Acta M, 317
Sharma, C., Manjeshwar, R. & Weinhouse, S. (1963) J. Biol. Chan. 238, 3840
Sherline, P., Lynch, A. & Glinsmann, W. (1972) Endocrinology21, 680
Shimazu, T. & Amakawa, A. (1968) Biochem. Blophys. Acta 165, 335
Shimazu, T & Amakawa, A. (1975) Biacilein.33i2=stAs-La325...2„ 242
Soderling, T.R. & Park, C.R. (1974) Advan.aclic Nucleotide Res. AL 284
Sokal, J.E. (1966) Endocrinoloy 78, 538
Sokal, J.E., Miller, L.L. & Sarcione, E.J. (1958) Amer. J. Physiol.115.9 295
Sokal, J.E., Sarcione, E.J. & Henderson, A.M. (1964) Endocrinology 930
Soskin, S. (1941) Physiol. Rev. 214 140
Soskin, S., Essex, H.E. Herrick, J.P. & Mann, F.C. (1938) Amer. 11:21-asj,..2LL2.1.2.. 558
Stalmans W., BeWulf, H., Lederer, B. & Hers, E.G. (1970) Eur. J. Biochem. 151. 9
Stalmans, DeWulf,H.& Hers, H.G. (1971) 582 Stalmans, W., DeWuif, H.,Hue, L..& HersiII.G.(1974a) Eur. J. Biochem. 41, 127
Stalmans, W., Laloux, M. & Hers, H.G. (1974b) Eur. J. Biochem. Ai. 415
Steele, R. (1966) Eraeb. PhzE121. Biol. Chem. Exp. Pharmakol. 57, 91
Stehle, R.L. (1950) ItiVitam.Hor c8, 216
Steiner, D.F. (1964) Nature (London 204, 1171
221
Steiner, D.P. (1966) Vitam. Horm._.(New ysistl21.2. 1
Steiner, D.F. & King, J. (1964) J. Biol. Chem. 239, 1292
Steiner, D.F., Rauda, V. & Williams R.H. (1961) J. Biol. Chem. 216z.299
Stetten, De.W & Boxer, G.E. (1944) 21131-01tSL511.169 271
Stumpf, B., Bole, A. & Seubert, W. (1972)n.33529zBiochei Commun. 49, 164
Sutherland, E.W. & Wosilait, W.D. (1956) J. Biol. Chem. 218 459
Tan, A.W.H., & Nuttall, F.Q., (1974) Fed. Proc. .1311 1894A,
Tata, J.R., (1964) 2L2.0.11229 284
Thomas, A.J. (1969) in: Automation Mechanisation and Data Handling in Microbio Ed. by Soc. Applied Bacteriology, Technical Handbook. Academic Press.
Thurston, J.H.,Jones, E.M. & Hauhart, R.E. (1974) Diabetes 23 597
Tolbert, M.E.M. & Fain, J.N. (1974) J. Biol. Chem. 249 1162
Tolbert, M.E.M.,Butcher, P.R. & Fain, J.N. (1973) J. Biol. Chem. 248a 5686
Unger, R.H. (1972) Advan. Metab. Dis. 6, 73
Vaisler, L., Costiner, E. & Gheorghiu, C. (1965 ) St. Cercet. Endocrinol. 16, 403
Vaisler, L., Costiner, E. & Gheorghiu, C. (1965b) St. Cercet. Endocrinol. 16, 557
Van den Berghe, G., Hue, L. & Hers, H.G. (1973) Biochem. J. 214., 637
Van Lan, V., Yamaguchi, N., Garcia, M.J., Ramey, E.R. & Penhos, J.C. (1974) Endocrinology 2A, 671
Villar-Palasi, C. & Larner, J. (1960) Bi2...Llen....11 21 ta, 171
Vinuela, E., Salas, M. & Sols, A. (1963) lz Bio.238" 1175
Walaas, 0., Walaas, E. & Gronnerod, 0. (1974) -1,-9:UZda1212. 93
Walker, D.G. (1963) 13iochenys. Actl_71, 209
Walker, D.G. & Rao, S.(1964) Biochem. J. 90, 360
warn, A.K., Siebler, G., Zepf, E. & Schimassek, H. (1974) 52P.... Sela211g.....1Pilsen1- 7 59 353
222
Winternitz, LW., Long, C.N. & Dintzis, R. (1957) Endocrinology dla 724
Whelan, W.J. (1971) Biochem. J. 122, 609
Wolf, D.P., Fischer, E.H. & Krebs, E.G. (1970) Biochemistry 9, 1923
Wosilait, W.D. & Sutherland,E.W. (1956) J. Biol. Chem. 218, 469
Zaragoza-Hemmans, N. & Felber, J.P. (1972) Horm. Metab. Res. 4, 25