Regulation of insulin effector systems in the brain


Material Information

Regulation of insulin effector systems in the brain
Physical Description:
v, 137 leaves : ill. ; 29 cm.
Mudd, Laura Mary, 1958-
Publication Date:


Subjects / Keywords:
Receptor, Insulin   ( mesh )
Protein Kinase C   ( mesh )
Brain -- enzymology   ( mesh )
Physiology thesis Ph.D   ( mesh )
Dissertations, Academic -- Physiology -- UF   ( mesh )
bibliography   ( marcgt )
non-fiction   ( marcgt )


Thesis (Ph.D.)--University of Florida, 1989.
Bibliography: leaves 123-136.
Statement of Responsibility:
by Laura Mary Mudd.
General Note:
General Note:

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001106705
oclc - 25072955
notis - AFK3052
System ID:

This item is only available as the following downloads:

Full Text








A B S T R A C T ................................................ ... ...........................................................iv


I. INTRODUCTION........................................................ ............................. 1

Insulin and Insulin Receptors in the Brain........................................ 2
Insulin in the Brain: Potential Functions......................... ............ 4
Protein Kinase C (PKC): Regulation of Insulin and IGF I
R eceptors....................................................... .......................... 6
Ligand-Receptor Interactions.......................................................8
Phosphorylation of Receptors.................... .......... .............. 8
Receptor-Induced Effects......................... .........................10
Effects ofGrowth Factors on PKC.................................... ..11
PKC in the Brain: Distribution........................................ ........12
PKC Regulation of Glucose Uptake and Insulin Receptors
in the B rain................................................. ................. ...............13
Cultured Brain Cells: A Model for the Study of Neurotrophic/
Neuroactive Substances.......................... ..............................14

II. METHODS.................................. ..........................................................20

Preparation of Primary Neuronal Cultures from Rat Brains......20
Preparation of Primary Astrocytic Glial Cultures from Rat
Immunocytochemistry of Neuron-Specific Enolase/Glial
Fibrillary Acidic Protein.......................... ........................ 22
Neuronal Depolarization..........................................................23
Insulin Radioimmunoassay.......................... .......................................... 23
Characterization of Immunoprecipitable Insulin by HPLC..........24
Labelling of Immunoreactive Insulin in Neuronal
Characterization of the Regulation of Neuronal Insulin Release
by Glucose................................................................................. 25
2-Deoxy-D-glucose (2-dGlc) Uptake...............................................26
lodination of Insulin................................................ ........................26
Insulin Binding........................................................ ..... .................. 28
Protein Determinations...................................... ............................29

PKC Im m unocytochemistry.................................. ............................... 31
Statistical Analysis........................... ...... ..........................31


Introduction.................................... .................................................... 37
Results......................................................................... ............. ........... 39
D iscussion...................................... ...................................................... 40

IN PRIMARY CULTURE......................... .................................51

Introduction........................................... ................ ........................... 51
R esults....................................................................... ................. .......... 52
Discussion....................................... ..................................................... 53

E S T E R S ........................................................... ...... ......................71

Introduction.................................... ..................................................... 71
Results.............................. ...............................................72
Discussion....................................... ..................................................... 72


Introduction.................................... .................................................... 76
R esults........................................... ....................................................... 77
D iscussion............................ ................................... ................................ 79

AND GLIAL PKC........................................................... 97

Introduction............................................................. ........................... 97
R esults............................................................................ ............................. 98
Discussion...................................... ..................................................... 99

VIII. DISCUSSION AND SUMMARY...............................................................113

R EFERENC ES ....................................................................... ................................. 123

BIOGRAPHICAL SKETCH........................ ...... ..........................137

Abstract of Dissertation Presented to the Graduate School of
the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy



Laura Mary Mudd

May, 1989

Chairman: Mohan K. Raizada
Major Department: Physiology

In spite of insulin's effects on glucose uptake in the periphery and the fact
that the brain is a glucose-obligate organ, the brain was long believed to be
insulin-independent because insulin does not cross the blood-brain barrier.
Many reports over the course of the past ten years have localized both insulin
and insulin receptors in the brain, although the source and function of the
peptide in the brain and the mechanisms by which receptors are regulated have
remained the subjects of some mystery.
It was found that an immunoreactive insulin was released from cultured
neurons under depolarizing conditions, that labelled leucine could be
incorporated into the released peptide and that the peptide coeluted with a rat
insulin standard when the two were applied to a reverse-phase high pressure
liquid chromatography column. Thus, the peptide resembles insulin
chromatographically as well as immunologically, is synthesized in neuronal
cultures and is released under depolarizing conditions. In the brain, insulin
may act as a neurotransmitter.

Protein kinase C (PKC) occurs in many tissues, but it occurs at highest
concentration in the brain. It has been shown to regulate glucose uptake and
the insulin receptor in many peripheral tissues. PKC stimulates both glucose
uptake and insulin binding in glial cells without affecting either in neurons. The
stimulation of insulin receptor binding in glial cells is not accompanied by an
increase in activity; thus it appears to inactivate the receptors. Activation of PKC
by phorbol esters, as evidenced by translocation from the cytosol to the
membrane, occurs in both neurons and glial cells, although the fraction of
cytosolic, and thus, translocatable, PKC is much smaller in the former. In glial
cells, Translocation of PKC precedes other effects of PKC, and thus, may be
involved in the mechanism. Insulin stimulates the level of immunoreactive PKC
in glial cells, but not in neurons; thus, in glial cells, this appears to be a
feedback loop whereby insulin regulates its own receptor. In neurons, both
arms of the circuit, that is, regulation of the PKC concentration by insulin and
regulation of the insulin receptor by PKC, are missing. That neuronal and glial
insulin receptors are differentially regulated is not surprising as the receptors
differ from one another with respect to both structure and function.


Historically, insulin has been used in the treatment of diabetes
mellitus and psychosis. Pancreatic insulin was first isolated in
1922 at which time it was used clinically in the treatment of Type I
diabetes (1). In classical Type I diabetes the insulin-secreting 8-
cells of the pancreas are destroyed and the patients become
dependent on exogenous insulin to control their blood sugar. Several
years after Banting and Best began using insulin to treat diabetes,
Schmidt (2) and Sakel (3) separately reported successful treatment
of psychotic patients with insulin in the presence and absence of
carbohydrate, respectively, that is, treatment with and without
hypoglycemic shock. Insulin-induced hypoglycemic shock was used
successfully in the treatment of schitzophrenia and depression as
well as psychosis until the development of electroshock therapy and
more sophisticated psychoactive drugs (4).
The mechanisms of pancreatic insulin synthesis and release, as
well as physiological effects on liver, muscle and adipose tissue,
have been well characterized. Pancreatic insulin is synthesized as
preproinsulin, cleaved to proinsulin almost immediately and

packaged into secretary granules where the C-peptide is cleaved
prior to secretion (5). Basal beta-cell secretion of insulin is about
0.5 U/hour. This can be increased 10-30 times with acute
stimulation. Glucose is the most potent stimulator of pancreatic
insulin release in man (6) although amino acids (especially arg, lys,
leu and val), glucagon, B-adrenergic and vagal stimulation of the
pancreas stimulate secretion as well (7,8). Somatostatin,
serotonin, dopamine, prostaglandin E and splanchnic stimulation are
inhibitory to insulin release (8,9). Chronic glucose stimulation
induces a biphasic release of insulin (10); this is thought to be due
to the release of short- and long-term insulin stores (8).
The first step in the action of insulin on the peripheral target
tissues is its interaction with specific receptors on the cell
surface. This interaction of insulin with its receptor stimulates
autophosphorylation of a tyrosine residue (11) as well as receptor
tyrosine kinase activity (12). Binding of insulin induces an increase
in glucose and amino acid uptake by liver, muscle and adipose
tissues and consequently an increase in the synthesis of glycogen
and protein with a corresponding decrease in lipolysis and
gluconeogenesis (13).

Insulin and Insulin Receptors in the Brain
In spite of insulin's effects on glucose uptake in the periphery
and the fact that the brain is a glucose-obligate organ, the brain was
long believed to be insulin-independent because insulin does not
cross the blood-brain barrier. Havrankova et al., established the
presence of insulin in the brain at concentrations that averaged 25

times that of plasma insulin (14). Since then many reports have
appeared in the literature confirming this finding although the
actual amount of insulin in the brain is the subject of some
controversy (15,16). Insulin is found in cultures from mouse brain
and neuronal cultures from rat brain (17,18). The presence of this
insulin-like peptide in the brain raises two major questions; what
is the source of this peptide and what is its function?
The following observations suggest that insulin may be
synthesized in the brain: (a) central insulin concentrations appear to
remain constant in pathological situations in which peripheral
insulin concentrations vary widely (19); (b) insulin-like activity
has been demonstrated in cultured brain cells which are removed
from pancreatic insulin for weeks (18); (c) radioactive amino acids
can be incorporated into an insulin-like peptide in cultured neurons
by a cycloheximide-sensitive process (18) and (d) there is an mRNA
in the brain which hybridizes with insulin cDNA (20-4). In addition,
our experiments have shown that insulin is synthesized and released
by neurons (25). Furthermore, when neuronal cultures were labelled
with leucine and depolarized, a labelled peptide was released which
could be precipatated by an insulin antibody. Cycloheximide
decreased insulin synthesis by 80% (18).
In recent years specific insulin receptors in the brain have been
demonstrated conclusively (26). These receptor sites, which are
distributed non-uniformly throughout the brain (27), are evenly
distributed between neurons and glial cells (28-30). Studies have
shown that neuronal and glial insulin receptors are structurally and
physiologically distinct in several respects. The brain insulin

receptor is generally similar in subunit structure to the more
familiar peripheral insulin receptors although the alpha-subunit is
somewhat smaller in the brain (31) due to decreased glycosylation
of the receptor in the brain. In particular, the neuronal alpha- and
beta-subunits are smaller, while the glial subunits resemble those
from the liver (29,32) in size. That the glial receptor is
structurally similar to the peripheral receptor is not surprising,
since insulin appears to have traditional metabolic effects in glia
and neuromodulatory effects in neurons. Insulin stimulates glucose
uptake by glia, but not by neurons (28) and inhibits norepinephrine
uptake by neurons, but not by glia (30).

Insulin in the Brain: Potential Functions
Insulin in the central nervous system (CNS) has been implicated
in the control of brain growth and development (28,33-5),
catecholamine release (36-9), and diabetes (40) in in vitro studies
and satiety (41) in in vivo studies. In addition, insulin may itself
act as a neurotransmitter in the CNS. In vitro studies have been
essential in developing an understanding of insulin's actions in the
CNS at the cellular level. With regard to insulin's role in growth and
development, insulin stimulates DNA, RNA and protein synthesis in
mixed brain cell cultures and astrocyte glial cultures. It also
stimulates 2-deoxy-D-glucose uptake in astrocyte glial cultures
(28,33). These are all events which are associated with cell growth
and proliferation. In addition, insulin has differentiative effects on
developing rat brain. When immature rat retinal cultures were
exposed to insulin, precocious synaptic release of acetylcholine was

evoked within one hour while control and glutamate treatment of the
cells elicited no such response (34). Insulin induces ornithine
decarboxylase activity and neurite outgrowth in cultured embryonic
mouse brain cells and rat brain cultures (35). Thus, insulin induces
maturation of brain cells in culture as well as growth.
Insulin alters the content of serotonin and catecholamines in the
brain by increasing the rate of uptake of their precursors (36-8). It
also stimulates the release of dopamine, epinephrine and
norepinephrine from hypothalamic slices (39) and changes the firing
rates of neurons in the striatum (42) and hippocampus (43). Insulin,
then, definitely acts as a neuromodulator in the CNS and it may act
as a neurotransmitter itself. Recently, the neuromodulatory effects
of insulin have been suggested to be mediated by insulin receptors
present on the neurons (44). Insulin has, then, satisfied several
criteria by which putative neurotransmitters are classified as such:
it is synthesized in neurons, released under depolarizing conditions,
bound with specificity and high affinity to a receptor, degraded and
modulates neuronal activity.
CNS insulin may be involved in the pathophysiology of obesity and
Type II, or insulin-independent, diabetes. In vivo studies in baboons
indicate that insulin has a direct effect on satiety (41). When
insulin was infused into the lateral cerebral ventricles, a
significant, dose-related decrease in food intake and body weight
was observed. This may occur by interaction with neurons in the
hypothalamus. Injection of insulin in low doses causes electrical
activity to increase in hypothalamic neurons (45). In addition to its
association with obesity, central insulin may play a role in Type II

diabetes. Lesions of the ventromedial hypothalamus cause
hypersecretion of hypothalamic insulin (46).
In studies involving cultures of brain cells from diabetic mice
the ratio of externalized insulin receptors to total insulin receptors
was substantially decreased in cultures from diabetic mice versus
control cultures (40). As obesity is a major risk factor for Type II
diabetes there may be a relationship between these two effects of
central insulin. This hypothesis is supported by the observation that
insulin receptor number is significantly decreased in the olfactory
bulb of obese rats (47).

Protein Kinase C (PKCI: Regulation of Insulin and IGF I Receotors
PKC is a serine/threonine kinase which is present in many
tissues but occurs at highest concentrations in brain (48). PKC is
calcium-dependent and 1,2-diacylglycerol (DAG) a product of
membrane phospholipid metabolism, increases the affinity of the
enzyme for calcium (49-50). Reports of the molecular weight of
PKC vary. The different values may reflect the method by which the
relative molecular weight (Mr) is determined, as evidenced by a
study in which values of 77,000 and 82,000 daltons were obtained
from sucrose density gradient and polyamide gel electrophoresis,
respectively (51). Differences may also be attributable to subunit
aggregation (52) or to the existence of different isozymes of PKC
(53-4). Currently, seven highly-homologous isozymes of PKC have
been isolated and characterized. Four are single polypeptide chains
with four constant and five variable regions, while three subspecies
differ slightly. The isozyme distributions differ with respect to one

another (55-6) and with respect to the development of the organism
(53,57), however, the kinetic properties of the isozymes are very
similar. For a review see Nishizuka (54).
Tumor-promoting phorbol esters, such as 12-O-tetradecanoyl-
phorbol-13-acetate (TPA), act as exogenous stimulators of PKC (58).
Upon activation by TPA or calcium, there is a rapid decrease in
cytoplasmic PKC and a corresponding increase in membrane PKC (59-
60, Figure 1-1). Following this translocation, the membrane-bound
PKC catalyzes the phosphorylation of specific proteins (61-2).
Translocation appears to mediate other activities of PKC such as
neuronal potentiation (63) as well. Prevention of PKC
redistribution, as with concanavalin A, has been shown to block
activation of PKC (64). Once activated, PKC acts in many types of
cells to block hormone-stimulated phosphoinositide hydrolysis, and
thus exerts negative feedback over its own activation (65-8).
PKC has different effects in different tissues (48). In the brain
PKC is involved in the regulation of neuronal ion channels (69-71),
synaptic plasticity (72-3), neurite outgrowth (74) and
neurotransmitter release (75-7), as well as the differentiation of
astrocytes and oligodendrocytes (78-9) and changes in membrane
conductance in astrocytes (80). Many of the varied effects of PKC
appear to result from its interactions with growth factor receptors
with tyrosine kinase activities. PKC-induced receptor
phosphorylations alter the affinities, activities and effects of some
of these receptors. These growth factors, in turn, have been shown
to regulate either content or activity of PKC in certain cells.

Ligand-Receptor Interactions
PKC regulates the specific binding of insulin and insulin-like
growth factor I (IGF I). Phorbol esters regulate the insulin receptor
in lymphocytes, adipocytes and monocytes (81-3). In each of these
types of cells, phorbol esters inhibit the binding of insulin by
increasing the Km of the high affinity receptor. The endogenous
analogues of TPA, the DAGs, also reversibly inhibit the binding of
insulin to its receptor (84) by altering the affinity of the receptor.
The calcium ionophore A23187 potentiates the effect of TPA on
insulin binding in monocytes (85). TPA does not, however, decrease
insulin binding in all cell types. TPA has no effect on insulin binding
in either 3T3 cells or in hepatoma cells (86-7) although the
response to insulin is impaired in the latter. Thus, PKC-induced
decreases in insulin binding may result from either decreased
affinity of the receptors or increased internalization of the insulin-
receptor complex, depending on the type of tissue or cell.
The case for regulation of the IGF I receptor by PKC is similar to
that for the insulin receptor. DAGs inhibit IGF I binding to the IGF I
receptor in monocytes. TPA also inhibits IGF I binding in
lymphocytes, monocytes and adipocytes by altering the high-affinity
binding site without altering the number of receptors (84). This
differs from insulin or IGF I down-regulation of the IGF I receptor,
which results from a decrease in the number of receptors (88).

Phosohorvlation of Receotors
The mechanism for PKC-stimulated alterations in the insulin and
IGF I receptors appears to involve serine/threonine phosphorylation

of those receptors. In 1983, TPA was first shown to stimulate
phosphorylation of both insulin and IGF I receptors in IM-9 cells that
had been preincubated with H332p04 (61). Insulin- and TPA-
stimulated phosphorylation appeared to be additive, suggesting that
there was no interaction between the sites. In 1984, TPA was
shown to enhance serine/threonine phosphorylation of the insulin
receptor in rat hepatoma cells at nine sites (87). Insulin was shown
to stimulate phosphorylation of tyrosine and serine residues at six
sites, three of which were similar to the TPA-phosphorylated sites.
In addition, the phorbol ester decreased insulin-stimulated
phosphorylation, suggesting that there was, in fact, an interaction
between the sites of action of the two agents. In later studies on
IM-9 and HepG2 cells, TPA was found to phosphorylate four major
serine residues, which were not phosphorylated in untreated cells
and to increase the phosphorylation of one threonine residue on the
insulin receptor. These serine residues were not phosphorylated by
insulin, which, however, did phosphorylate three tyrosine residues
(89). PKC acts directly on the insulin receptor as it phosphorylates
the insulin receptor in vitro (90). Similar results were seen with
the IGF I receptor. These very different profiles of phosphorylation
induced by insulin and phorbol esters give strong evidence that
insulin and IGF I were not acting through PKC. In 1988, TPA was
found to enhance predominantly the phosphorylation of one serine
residue on the insulin receptor in hepatoma cells (91). TPA-
treatment of cells inhibits insulin-stimulated receptor
phosphorylation of exogenous substrates by 50 percent. These
changes in the receptor are maintained when the receptors are

isolated and are reversed by incubation with alkaline phosphatase,
suggesting that PKC decreases the tyrosine kinase activity of the
insulin receptor and that this decrease is due to the phosphorylative
changes induced in the receptor. Studies on rat adipocytes have also
shown that TPA increases the Km of the insulin receptor for ATP,
thus suggesting a mechanism for insulin resistance in adipocytes

Receptor-Induced Effects
The tyrosine kinase activity of the insulin receptor is necessary
for normal receptor function and down-regulation (92-3). This has
been demonstrated by studies in which kinase-defective mutant
insulin receptors were used to transfect cells. The mutant
receptors demonstrate normal binding of insulin but do not possess
tyrosine kinase activity, are not internalized and do not possess
biological activity. Treatment of endogenous, biologically-active
insulin receptors with monoclonal antibodies against the receptor
kinase inhibits insulin-stimulated effects as well (94). Inhibition
of receptor tyrosine kinase activity by TPA leads to the same types
of defects. Treatment with TPA is associated with inhibition of
insulin-mediated DNA synthesis (95), phosphorylation of metabolic
enzymes (87), glycogen synthesis (96) and glucose uptake (97),
among others. Thus, an impaired PKC pathway can have dire
consequences for the cell or organism. This is demonstrated by
genetically obese (fa/fa) rats, in whose hearts and hepatocytes both
the basal distribution and the translocation of PKC are abnormal.
The resultant insulin-insensitivity can be duplicated by treating

lean rats with TPA to down-regulate PKC (98). The same is not true
for all tissues, however, as phorbol esters have only minimal
effects on insulin sensitivity in rat skeletal muscle (99). The
interaction of PKC and the insulin receptor is somewhat complicated
by reports of synergism in mitotic stimulation (100). In addition,
there are proteins which have phosphorylation sites for both PKC and
receptor kinases (101) and are stimulated by both types of mitogens
(102). This last effect would account for the paradoxical way in
which PKC both inhibits the insulin receptor and mimics many of
insulin's effects within the cell.

Effects of Growth Factors on PKC
While the majority of studies on PKC/growth factor interactions
in the literature focus on regulation of growth factor receptors by
PKC, those receptors frequently regulate PKC as well. Treatment
with insulin in the presence of glucose has been shown to increase
both the binding capacity of PKC and enzymatic activity in
adipocytes (103). As the insulin effect is eliminated in the presence
of high glucose, the effect may be secondary to increased glucose
uptake in the insulin-treated cells. Insulin also increases the level
of cytosolic calcium ion in adipocytes (104), which could account
for the increased PKC activity. Studies also demonstrate
enhancement of PKC activity in myocytes and mammary tumor cells
by insulin (105-6). In the former the increase occurs in both the
cytosolic and membrane fractions and is not inhibited by
cycloheximide. This increase in the activity of PKC is reportedly
mediated via increased DAG generated by phospholipid hydrolysis and

phospholipid synthesis. There is some controversy on this point,
with other work suggesting that insulin does not increase the
activity of PKC in myocytes (107). It is suggested that the
increases in phosphorylation observed after the administration of
insulin are mediated by S6 kinase which is activated by both insulin
and TPA. Growth factors, then, may increase PKC indirectly by
increasing either cytosolic free calcium ion or DAG or by acting at a
point in the pathway beyond the PKC molecule itself. A direct effect
of insulin on PKC must necessarily be demonstrated on the isolated

PKC in the Brain : Distribution
Although PKC is present in many tissues, it occurs at highest
concentration in the brain. The seven subspecies of PKC have
different distributions in the brain (54) and these change with brain
development. The gamma subspecies, which occurs only in the
central nervous system in the rat and monkey, has a
developmentally-regulated distribution, with expression increasing
from birth until it reaches a maximum at about three weeks of age.
Total PKC is also developmentally regulated in brain in studies in
cultured neurons and in vivo (57). Interestingly, insulin and IGF 1
receptors are developmentally regulated in the rat brain with
increases occurring in the first weeks of life, followed by a decline
(108-9). Immunohistochemical analyses show different antibody
staining patterns. There appear to be subspecies which are present
almost exclusively in neurons (55,110), in astrocyte-glial cells and
in oligodendrocytes (55,111). The enzymatic activity is also

unevenly distributed, with the left cerebral hemisphere expressing
more than the right in the rat (112). Binding studies using
radiolabeled phorbol esters show two to three times more PKC in
neurons than in glial cells cultured from the same rat brains (113).
The subcellular distribution of PKC has been the subject of many
investigations. It is localized in dendrites, axons, perikarya and
nuclei (110) of neurons with particularly high concentrations in
presynaptic terminals (111) and in growth cones (114). This is not
unexpected as PKC mediates both neurotransmission and neurite
outgrowth (115). Fractionation of glial cells demonstrated that the
majority of the PKC was cytoplasmic (116). The same study showed
that the majority of PKC in whole brain tissue is associated with
the membrane, suggesting that the majority of neuronal PKC is
membrane-bound. Seventy-five percent of the glial cell cytoplasmic
PKC can be translocated to the membrane within 30-60 minutes of
TPA-treatment. This is similar to the situation seen in peripheral

PKC Regulation of Glucose Uptake and Insulin Receptors in the Brain
PKC has been shown to regulate glucose uptake in many tissues.
Phorbol ester-induced decreases in the binding of insulin to its
receptor are not associated with decreased glucose uptake as, in
fact, phorbol esters stimulate glucose uptake in adipocytes,
myocytes, fibroblasts and astrocytes (97,117-9).
As discussed, in the brain there are distinctions between neurons
and glia with regard to PKC: PKC is present in higher concentrations
in neurons than in glia from the same brains as demonstrated by the

binding of phorbol esters (113); different isotypes of PKC are
present in neurons and glia (55,110,111) and phorbol esters
stimulate glucose uptake in glia but not in neurons (119).
The joint observations that neurons and glia differ with respect
to the physiological activities of the insulin receptor and PKC and
that phorbol esters regulate the insulin receptor and glucose uptake
peripherally, led us to investigate the role of PKC in the regulation
of insulin receptors in neurons and glia from the central nervous
system. Because activation of PKC by TPA has been shown to involve
translocation of PKC from the cytosolic to the membranous fraction
in many types of cells, we chose to examine TPA's effects on
neurons and glia to determine the relative concentrations of
immunoactive PKC in neurons and glia, whether TPA stimulates the
translocation of PKC in both neurons and glia, and whether the
translocation in glia, if it occurs, precedes TPA's effects on glucose
uptake and the insulin receptor and, thus, might be involved in a
mechanism. Differences in the ability of TPA to induce
translocation or the time-course of the translocation might
presumably explain TPA's differential effects on neurons and glia.
Lastly, in order to determine whether insulin regulates its own
receptor by this pathway, we chose to study insulin's effects on PKC
in neurons and glia from the brain.

Cultured Brain Cells: A Model for the Study of Neurotrophic/
Neuroactive Substances
The importance of neurons in brain function is unquestioned.
Proper neuronal growth, development and maintenance are essential

for every aspect of normal brain function. The importance of growth
factors, such as nerve growth factor, insulin and the insulin-like
growth factors, in neuronal development and activity has been well
documented although the mechanisms by which these agents act are
still objects of intense study (115,120).
In contrast, although glial cells are the predominant cell type in
the mature nervous system, their involvement in the growth,
development, differentiation and function of the brain has only
recently become a subject of investigation. Glial cells have recently
been implicated in processes involving the growth, development and
function of the nervous system. Glia are not only responsive to
trophic factors but may produce them as well. They facilitate
neuronal migration in fetal life but induce scarring to inhibit
regeneration in the mature nervous system. In addition, they alter
the levels of neurotransmitters available at the synapse, thus
altering neuronal excitability and they may even be excitable
Glia respond to trophic factors such as insulin and insulin-like
growth factor I (IGF I) with an increase in glucose uptake in contrast
to neurons (119,121). This is of interest developmentally as both
insulin and IGF I receptors in brain increase to a maximal level
during brain development and then show a gradual decline (108-9).
Glia also exhibit an increase in macromolecular synthesis in
response to both insulin (33) and IGF I (121). Epidermal growth
factor receptors in the brain appear predominantly in glia, as well

In addition to being responsive to trophic factors, glial cells may
produce them as well. A substance which is immunologically
identical to nerve growth factor is present in glia (123) and glia,
which develop at the same time as neurons, provide other
extracellular molecules which enhance neuronal migration (124).
This glial stimulus to neuronal growth is lost with age as astrocytic
glial cells block axonal regeneration (125) and synapse formation
(126) in the mature nervous system. Inflammation of the mature
nervous system causes a reactive gliosis which prevents neuronal
repair as in multiple sclerosis (124).
Finally, glia contribute to the regulation of nervous system
excitability. Glia take up, and thus inactivate, glutamate, GABA,
aspartate and serotonin (127) at the synapse. Insulin decreases
levels of alpha2-adrenergic receptors in glia; this would tend to
regulate the amount of norepinephrine in the synaptic cleft (128).
Glia do not only take up neurotransmitters but also act on several
via specific enzymes such as glutamine synthetase and GABA
transaminase. In addition, they have specific receptors for alpha-
and beta-adrenergic agonists, dopamine, prostaglandin El, secretin,
somatostatin and vasoactive intestinal peptide (127), among others.
Lastly, glia have themselves been shown to possess some voltage-
gated channels and, thus, may act as excitable cells (129).
Because both neurons and glia are important for appropriate
development and function of the nervous system and because of the
afore-mentioned differences in neurons and glia with regard to both
the insulin receptor and PKC, we chose to study the interaction of


insulin and PKC in differential cultures of neurons and glia from the
same rat brains.

(n n u

C- 2 --o ; = -

o c-.o o -- E CD
0 E0c
>S(-D CAo- iE

E --= L 0 .cE 0

>- Co a o c

0N0 CL .

CLs C. m a )





Preparation of Primary Neuronal Cultures from Rat Brains
Brains from 8-10 one-day-old rats were removed in a sterile
manner from the cranium at the level of the medulla oblongata and
placed in an isotonic buffer solution containing 100 units penicillin,
100ltg streptomycin and 2.5jg Fungizone/ml. All pia mater and
blood vessels were removed and the brains were minced into 1-
2mm3 pieces. Brain cells were dissociated by trypsin and
deoxyribose treatment as described previously (18). Dissociated
cells were suspended in 50 ml of Dulbecco's modified Eagle's
medium (DMEM) containing 10 percent plasma-derived horse serum
(PDHS) and sedimented at 1000 x g for 10 minutes at 240C. The cell
pellet was suspended in the same medium at a concentration of 1.5 x
106 cells/ml. Fifteen million cells were placed in each Falcon
tissue culture dish pretreated with poly-L-lysine and the cells were
placed in a humidified incubator at 370C with 6 percent C02/94
percent air. The cells began to attach immediately and 90-96
percent were attached to culture dishes within 30-60 min. After
three days in culture, the cells were treated with 10.M cytosine

arabinoside in DMEM containing 10 percent PDHS. This treatment
resulted in the death of rapidly dividing cells. After 48 hours
cytosine arabinoside was removed and the cells were refed with 10
percent PDHS in DMEM and the cells were grown in culture for
another 7-10 days prior to use.
These cultures contain 80-85 percent neurons as demonstrated
by light microscopy and immunocytochemical markers (18, Figure 2-
1). Since insulin and insulin receptors were widely distributed
throughout the brain, in contrast with the localized distribution of
other neuropeptides and their receptors, cultures from the whole
brain were used. We considered growing the cells in a chemically
defined medium. However, since one of the essential components of
this defined medium is insulin in relatively high concentration, this
possibility was abandoned. It was felt that the presence of insulin
in the medium would complicate studies related to the expression
and action of insulin receptors. Thus, we proposed to continue to
culture cells in DMEM containing either PDHS or fetal bovine serum
(FBS). PDHS and FBS were purchased from Hyclone and Gibco
Laboratories, respectively, with an insulin concentration of 1-4
ng/ml. At this concentration, insulin did not cause either down-
regulation or other biological effects on insulin receptors of either
neurons or glia (130).

Preparation of Primary Astrocytic Glial Cultures from Rat Brains
The procedure for removing the brains and dissociating the cells
was the same as that described above. After the cells were
centrifuged, they were suspended in medium containing 10 percent

FBS and plated onto 100mm culture dishes at a density of 10 million
cells/dish. After three days the cells were refed with 10 percent
FBS in DMEM. After an additional three days, cultures were rinsed
once with an isotonic buffer and dissociated by treatment with
trypsin. The cells were centrifuged at 1000 x g for 10 min at room
temperature and the pellet was resuspended in DMEM containing 10
percent FBS. Five hundred thousand cells were placed on each
100mm culture dish and the cells were returned to the incubator
until they were confluent, at which time they were used. Phase
contrast microscopic examination revealed a confluent monolayer of
large flat cells by day 6 or 7 after transfer. These cells have
previously been demonstrated to be of glial origin (131). Greater
than 98 percent of these cells have been identified as astrocytic
glial cells. Neuronal cells did not survive the transfer (Figure 2-2).

Immunocvtochemistry of Neuron-Specific Enolase/Glial Fibrillarv
Acidic Protein
Either cultured neurons or glia were washed with PBS and the
cells were fixed in a solution of 4 percent paraformaldehyde and 10
percent picric acid in PBS (pH 7.4) for 30 minutes at 40C. Cultures
were again washed 3 times with PBS and permeabilized in 0.1
percent Triton-X 100, 5 percent low-fat, dry milk in PBS for 30
minutes at room temperature. After being washed again with PBS,
the cells were exposed to a polyclonal antibody (either 1:100
neuron-specific enolase or 1:20 glial fibrillary acidic protein)
diluted in 0.1 percent NaN3/5 percent non-fat dry milk in PBS for 24
hours at 4C. PBS was used to wash the excess primary antibodies

from the cultured cells and a rhodamine-linked, goat anti-rabbit Ig
was diluted 1:100 in 5 percent non-fat dry milk in PBS and applied
to the cells for 30 minutes at room temperature. Excess secondary
antibody was washed from the cultured cells with PBS and the cells
were photographed under fluorescent light at 400x magnification on
a Zeiss D-7082 Axiophot photomicroscope.

Neuronal Deoolarization
Medium was aspirated from 14 day old neuronal cultures and
replaced by 8 ml of various solutions: the control solution contained
140 mM NaCI and 1.2 mM MgSO4; the depolarizing solution contained
78 mM NaCI, 60 mM KCI, 2 mM CaCl2 and 1.2 mM MgSO4; a potassium
solution contained 80 mM NaCI, 60 mM KCI and 1.2 mM MgSO4; a
calcium solution contained 138 mM NaCI, 2 mM CaCl2 and 1.2 mM
MgSO4, and a veratridine solution contained 10 uM veratridine, 138
mM NaCI, 2 mM CaCl2 and 1.2 mM MgSO4. Cultures were incubated
for 30 minutes at 37C. The solutions were aspirated, lyophilized,
reconstituted in distilled water and subjected to radioimmunoassay
for insulin.

Insulin Radioimmunoassav
A 100 ul aliquot of a sample of human insulin standard (0-300
uU/ml) was mixed with 5000 cpm of [1251]-insulin. One hundred ul
of guinea pig anti-human insulin (Serono) was added and the samples
and standards were vortexed and incubated at room temperature for
2 hours. A secondary antibody (sheep anti-guinea pig Ig, Serono) was

added and the sample and standard tubes were vortexed and
incubated for a further 30 minutes at room temperature. Following
the incubation, all tubes were centrifuged at 2500 g for 30 minutes
at 4C. Supernatants were discarded. All tubes were swabbed dry of
supernatant and the radioactivity in the pellets was counted.
Samples were compared to a standard curve and the cpms were
expressed as international units of insulin/ml.

Characterization of Immunoorecipitable Insulin by HPLC
Neuronal cultures grown in 100mm culture dishes were washed
twice with a solution of 138mM NaCI, 1.2mM MgSO4 and 2mM CaC12
(pH 7.4) at room temperature to remove the growth medium. Eight
ml of a depolarizing solution (78mM NaCI, 1.2mM MgSO4, 2mM CaCl2
and 60mM KCI, adjusted to pH 7.4) were placed on each culture dish
and the dishes were incubated at 370C for 30 min. in a 94 percent
air/6 percent C02 incubator. After incubation the solution was
aspirated from the plates.
High-pressure liquid chromatography (HPLC) was used to
determine whether the released peptide had the same
chromatographic properties as commercial rat insulin (Eli Lilly).
Two buffers were prepared. Buffer A consisted of 0.1 percent
trifluoroacetic acid (TFA) and 0.1 percent triethylamine (TEA) in
water. Buffer B consisted of 0.1 percent TFA and 0.1 percent TEA in
acetonitrile. The sample was dissolved in Buffer A and injected
onto a BioRad HiPore C4 column. A buffer system of 1:9, A:B was
graduated to 5:5, A:B over the course of 60 min. The column was run
at room temperature with a flow rate of 2 ml/min. Fractions were

collected every 0.2 min. and absorbance was monitored at 210 and
280 nm. Fractions were subjected to radioimmunoassay (Serono) to
determine whether the fraction which bound the insulin antibody
corresponded to insulin chromatographically.

Labelling of Immunoreactive Insulin in Neuronal Cultures
Neuronal cultures grown in 100mm culture dishes were incubated
in leucine-free DMEM, 10 percent dialyzed PDHS and 50p.Ci [3H]-
leucine (146.5 Ci/mmole) for 24 hours at 370C. Medium was
aspirated from the culture dishes; cultures were washed 4 times
with PBS and used for experiments as described above. A 100 pl
aliquot of each solution was exposed to a guinea pig anti-human
insulin antibody (Serono) for 2 hours at room temperature. Sheep
anti guinea pig Ig was added to each sample for 30 minutes to
precipitate the primary antibody and all samples were centrifuged
at 2500g to pellet the antibody-antigen complexes. Supernatants
were decanted and the radioactivity in each pellet was counted
(Liquiscint, National Diagnostics, LKB1217 Rack Beta Counter).

Characterization of the Regulation of Neuronal Insulin Release by
Neuronal cultures prepared in 100mm culture dishes were washed
twice with a control buffer [25mM NaHCO3, 1.2mM NaH2PO4, 122mM
NaCI, 1.2mM MgSO4 and 2.5mM CaCl2 (pH 7.4)] at room temperature
to remove media. Glucose buffers were prepared such that they
were isotonic to the control buffer (eg. 25mM NaHCO3, 5.5mM
glucose, 119mM NaCI, 2.5mM CaCl2 and 1.2mM MgSO4). Eight ml of

either the control or a glucose buffer were placed on each culture
dish. Cells were incubated at 370C for 30 min. in 6 percent CO2/94
percent air with 90 percent relative humidity. After the incubation
the buffers were aspirated from the dishes, lyophilized, and insulin
in the buffers was quantitated by radioimmunoassay.

2-Deoxv-D-Glucose (2-dGlc) Uptake
Medium was aspirated from the culture dishes and the cells were
washed three times with phosphate-buffered saline (PBS) at pH 7.4.
The cells were then incubated in PBS containing 1mM CaCl2, 0.5 mM
MgSO4 and 0.5 mM 2-dGIc (1 pCi/plate). After a five min. incubation
at 37C, the cells were washed three times with ice-cold PBS to
remove excess radioactivity. Cells were dissolved in 0.2 N NaOH;
they were then scraped from the culture dishes and radioactivity
was counted (Ecoscint, National Diagnostics, LKB 1217 Rack Beta
scintillation counter). The cpm were converted to dpm by the
counter (efficiency about 30 percent) and normalized for protein
values. The specific activity was then used to convert results to
nmoles of 2-dGlc/mg protein.

lodination of Insulin
One hundred ml of a pH 6.7 phosphate buffer was prepared from
55 ml of 0.3M KH2PO4 and 45 ml of 0.3M Na2HPO4. Chloramine T was
prepared as follows: the surface was scraped and chloramine T was
weighed and placed in a foil-covered tube; the salt was diluted to 4
mg/ml in the phosphate buffer and again diluted 1:100 (i.e. 25 l in
2.5 ml) in phosphate buffer just before use; all dilutions were in

foil-covered tubes as the compound is light-sensitive. One mg of
porcine insulin was weighed and dissolved in 2 ml of 0.01 N HCI.
Sodium metabisulfite was weighed and diluted similarly to the
chloramine T except that it was made to 8 mg/ml in the first
dilution such that the concentration after 1:100 dilution was 80
pg/ml. The decay chart for 1251 was checked to determine what
volume of 1251 contained 1 mCi. Fifty ml of a solution of 1 mg/ml
of insulin-free BSA (eg Sigma A-7030) in phosphate buffer was
prepared and about 20 ml was used to wash a sephadex-G25 column
(PD 10 column from Pharmacia). The following were added in
sequence to a disposable plastic test tube in a fume hood: 1) 25 pI
phosphate buffer, 2) 5 Ig insulin (10 I1), 3) 1 mCi 1251 and 4) 10 pl
chloramine T. The tube was capped, vortexed and contents were
incubated at room temperature for 5 min. Ten ul of sodium
metabisulfite was added to stop the reaction and the tube was
capped and vortexed once again. Two hundred Il of phosphate buffer
was added to increase the volume and the solution was added to the
top of the column. After the reaction mixture was absorbed by the
column, the column was washed with the BSA/phosphate buffer and
fractions were collected. [1251]-insulin usually eluted in about 3-5
ml. Ten pl of each fraction was removed and placed in a test tube,
capped and counted on a scintillation counter. To calculate percent
incorporation, 50 p. of 3 percent BSA and 1 ml of ice-cold 10
percent trichloroacetic acid were added to the test tube containing
the highest counts. Then the sample was incubated at 4C for 5 min.
and centrifuged at 1000 x g for 1 min. Radioactivity in the pellet
represented incorporated insulin. This should be greater than 90

percent. To calculate specific activity the number of cpm per ml
was determined. The counting efficiency was used to convert to
dpm/ml (75 percent for the LKB Rack Beta). Two million two
hundred thousand dpm/pCi was used to convert to .Curies. The
number of ig/ml was divided by the number of piCi/ml. Insulin
specific activity was generally about 40 I.Ci/g. Labelled insulin
was aliquoted into microfuge tubes, capped and stored at -700C until
use. The labelled material was tested for bioactivity by
displacement with a high concentration of unlabelled insulin (100
g.M) in a binding assay.

Insulin Binding
Medium was aspirated from culture dishes and the cells were
washed three times with PBS. Total binding was determined by
incubating triplicate dishes with a binding buffer (100mM Hepes,
30mM NaCI, 10mM glucose, 1mM CaCl2, 0.5 mM MgSO4, 0.1 percent
bovine serum albumin (BSA) at pH 7.4) containing [1251]-insulin
(100,000 cpm/plate). Non-specific binding was determined with a
similar buffer containing, in addition, 100p.M unlabelled insulin.
After a 1 hour incubation (2 hours for competition experiments) at
room temperature, cells were washed three times with ice-cold PBS
to remove excess, unbound insulin. The cells were then dissolved in
0.2 N NaOH and scraped from the culture dishes. Radioactivity was
counted on a Beckman Gamma 5500 counter. All values were
normalized for protein content and specific binding was determined
by subtracting non-specific binding from total binding.

Protein Determinations
One hundred to two hundred ul of a solution of protein in 0.1 N
NaOH was used for protein determinations. Bovine serum albumin
standards (10-100 ug of protein) were prepared in the same volume
of 0.1 N. NaOH. Samples and standards were made to 500 ul with
deionized, distilled water and protein determinations were made by
the method of Lowry (132).

Western Blot
Treated or untreated cultures of neurons and glia were washed 3
times with PBS, scraped from the culture dishes and centrifuged at
1000 x g for 5 min. to pellet cells. The supernatant was poured off
and the cells were resuspended in a homogenizing buffer consisting
of 20mM Tris HCI, 2mM EDTA, 0.5mM EGTA, 0.1mM PMSF and 1
percent 28-mercaptoethanol at pH 7.5. The suspension was
homogenized with 15 strokes of a glass homogenizer and centrifuged
at 1500 x g at 40C for 8 min. to remove nuclei and large particles.
The supernatant was recentrifuged at 100,000 x g at 2-40C for 30
min. to isolate cytosolic and membranous fractions. The membrane
was resuspended in homogenizing buffer containing 0.1 percent
Triton X-100. Samples of homogenates from whole cell were used
after the homogenization step. The protein content of samples was
determined and samples were made to a final concentration of 10mM
Tris base, 2 percent sodium dodecyl sulfate (SDS), 15 percent
sucrose, 0.002 percent bromophenol blue and 10 percent 28-
mercaptoethanol at pH 8.3. Samples were boiled for one min. and
stored at 40C before proteins were separated on discontinuous

polyacrylamide gels (5 percent stacking gel, 7.5 percent separating
gel, 30mA for 6 hours) and transferred to a nitrocellulose membrane
in a buffer consisting of 25mM Tris base, 150mM glycine and 20%
v/v methanol at pH 8.3 for three hours at 150mA. Total transfer
was demonstrated by transfer of prestained standards and by the
lack of a band as demonstrated by Coommassie staining of the gel
after the transfer. The nitrocellulose membrane was stored in 3
percent BSA in PBS overnight to decrease non-specific binding of the
The nitrocellulose membrane was incubated with a monoclonal
antibody against protein kinase C (Amersham, diluted 1:100 in PBS
with 0.1 percent BSA) for 4 hours at room temperature. This
antibody recognized the alpha and beta subtypes (Types II and III) of
PKC, which are the majority of PKC in the brain (159). It was then
washed 4 times for 5 min. with 0.1% Tween-20 in PBS (pH 7.4) and
incubated with horseradish-peroxidase-linked anti-mouse Ig diluted
1:100 with 0.1 percent BSA/0.1 percent Tween-20 in PBS for 30 min.
at room temperature. The membrane was washed as before and
incubated in 0.03 percent hydrogen peroxide/0.5 mg/ml 3,3'-
diaminobenzidine (DAB) in PBS prepared immediately prior to use,
until bands appeared. It was then washed and allowed to air dry.
Bands were quantitated via densitometry. Immunological PKC will
be referred to as iPKC throughout this dissertation, while PKC
activity will be designated aPKC.

PKC Immunocvtochemistrv
Cells were grown on sterile glass coverslips in culture dishes.
Prior to staining, they were washed three times with PBS and fixed
in 3.5 percent paraformaldehyde/0.25 percent glutaraldehyde in PBS
(pH 7.4) on ice for 30 min. The cultured cells were then
permeabilized with 0.1 percent Triton X-100 in PBS for 30 min. at
room temperature. Following the fixing and permeabilizing steps,
the cells were rinsed three times with PBS and a 1:10 dilution of a
monoclonal anti-PKC antibody in 1 percent BSA, 0.1 percent sodium
azide in PBS was applied. After a 24 hour exposure to this antibody
at 40C and they were again rinsed with PBS. Control cells were
treated with the same solution without the primary antibody. A
1:100 dilution of an anti-mouse Ig-peroxidase conjugate in 0.1
percent BSA in PBS was applied for 30 min. at room temperature and
then the excess was removed by washing with PBS. Finally, the
cells were incubated in a solution of 0.5 mg/ml DAB/0.03 percent
hydrogen peroxide in PBS prepared immediately prior to use. After
10 min. they were washed and a drop of 9:1 glycerol:PBS was applied
to the coverslip. The coverslips were inverted, placed on a glass
microscope slide and the edges were sealed with nail polish.
Photographs were taken at 400 and 1000 x magnification with a
Zeiss D-7082 Axiophot photomicroscope.

Statistical Analysis
Statistical analysis was by analysis of variance (ANOVA)
followed by Duncan's post hoc test when the means of several groups
were to be compared or Dunnett's post hoc test when the means of


several groups were to be compared to that of one control group.
Significance was determined for p<0.05. Experiments whose results
were expressed as a percent of control were converted to arcsin
prior to ANOVA if all values were equal to or less than 100 percent
or to the log if any of the values exceeded 100 percent. The
particular test used was specified in the legend of each figure. The
one exception to these rules was Figure 3-1, for which a two-tailed
Wilcoxan Rank Sum nonparametric test was employed because of one
outlying value in the depolarized group.

CD y

o w

( 0E

o t*.-
Sc 0

S0. o

j o CU C

.3 (D) 0


c _

co o,
.- 0

T5 t


= 3=) s
07 a

- E tm
= u Cu

) S a

D(D1 |

a 2 = 0
C> ^ p5




Insulin alters the content of several neurotransmitters in the
brain (36-8), stimulates the release of others (39) and changes
neuronal electrical activity in specific regions of the brain (42-3).
Thus, it definitely acts as a neuromodulator. As insulin is suggested
to act via specific insulin receptors (44), it may act as a
neurotransmitter as well. In order to be classified as a
neurotransmitter, however, synthesis and release of insulin under
depolarizing conditions must be demonstrated in neurons.
Two separate bodies of evidence suggest that the insulin present
in the brain is also synthesized there: 1) preproinsulin mRNA is
present in brain (20-23) and 2) the level of insulin in the brain is
independent of the level of insulin in the periphery (19,23). Several
investigators have identified an mRNA in brain tissue that
hybridizes to a cDNA for insulin. Insulin mRNA is found in neurons,
but not in glial cells in cultures from the brains of both rats and
rabbits (22-3). Only 3-5 percent of cultured neurons from rabbits
contain the mRNA. This is in agreement with an earlier study

localizing insulin immunoreactivity to 3-5 percent of cultured
neurons from rat brain (18). The mRNA species is larger than that
observed in the human pancreas when the two are compared by
Northern blotting techniques (24). In situ hybridization studies
localize neurons containing the insulin mRNA to the periventricular
hypothalamus and cerebral cortex in rat brain (21) and rat, mouse
and hamster anterior pituitary cells (20). Other areas of the brain,
including the olfactory bulb and choroid plexus do not contain the an
mRNA for insulin (21). In the anterior pituitary, only 5-10 percent
of the cells are positive for the mRNA; those cells are epithelial
and the immunoreactive insulin that they contain is localized in
secretary granules (20).
The second body of evidence relates to the independent regulation
of brain and peripheral insulin levels. Many investigators have
reported brain insulin concentrations that are higher than those
observed in plasma (14,23). It is unlikely that this represents
sequestration and concentration of plasma insulin for two reasons.
First, brain insulin concentrations are not altered by disease states
which raise or lower plasma insulin concentrations, although the
concentration in CSF is lowered in response to lowered plasma
insulin. Secondly, the capillaries of the blood-brain barrier do not
transport active insulin into the brain.
As discussed previously, insulin acts in the brain to alter
neuronal electrical activity. This, in combination with the evidence
that insulin may be synthesized in the brain, led us to investigate
whether pulse-labelled immunoreactive insulin could be synthesized
in the brain and whether this immunoreactive insulin could be

released under depolarizing conditions. The chromatographic
properties of the released immunoreactive insulin were then
compared to those of an insulin purified from rat pancreas by HPLC.

In this study, primary neuronal cultures from rat brains were
treated with depolarizing solutions containing a high concentration
of potassium (60mM) with or without calcium (2mM). After 30 min.
the solutions were aspirated and lyophilized and insulin was
quantitated by radioimmunoassay. Depolarized neuronal cultures
released more than three times as much insulin as saline controls
(103.21U/ml vs 31.7pU/ml) in the presence of calcium. Potassium-
stimulated release was calcium-dependent as in the absence of
calcium, insulin release was negligible (34.0iU/ml, Figure 3-1).
The sodium ionophore, veratridine, similarly stimulated the release
of insulin from neurons by 379 percent. In contrast, release of
insulin from glial cultures was not stimulated by depolarization.
This, in combination with the evidence that no mRNA for insulin has
been demonstrated in glial cells and that glial cells are not
immunoreactive for insulin (18), suggested that the insulin released
from neuronal cultures was not due to glial contamination of those
cultures. The very low level of insulin released by glial cells may,
in fact, have represented insulin taken up from the medium. The
immunoprecipitable insulin that was released from neurons under
depolarizing conditions coeluted with an insulin purified from rat

pancreas (supplied by Eli Lilly) when subjected to reverse phase
HPLC. Both the released material and the standard showed peaks for
both rat 1 insulin and rat 2 insulin (Figure 3-2).
When primary neuronal cultures were treated with D-glucose
(0.1-0.6 percent) in the presence of 2mM calcium, a dose-dependent
stimulation of insulin release was observed (Figure 3-3). Neuronal
cultures were exposed to [3H]-leucine for 24 hours prior to a timed-
release experiment. These cultures were washed and incubated at
37C with control and depolarizing solutions which were aspirated
from the cultures at time intervals from 1 to 60 min. A biphasic
pattern of insulin release was observed (Figure 3-4); this pattern of
release was similar to that observed with stimulation of pancreatic
insulin release by glucose. The fact that exogenous [3H]-leucine was
incorporated into the immunoreactive insulin suggested that this
insulin was synthesized within the neurons.

There is some controversy as to the origin of insulin in the brain.
Some investigators have shown that the concentration of insulin in
the brain is a fraction of that found in the plasma when quantitated
by different antibodies than the one first used by Havrankova, et al.
(16, 14). These same investigators have also found that the
concentration of insulin in the CSF is dependent on the plasma
concentration. They have suggested that these two pieces of
information, in conjunction with the lack of a demonstrable
proinsulin in the brain and the inability of one investigator to
demonstrate an mRNA for insulin in brain prove that insulin is not

synthesized in the brain (16). They propose that insulin passes from
the plasma into the CSF and is retained in brain to the extent that it
binds to local insulin receptors. They further propose that this
accounts for insulin immunocytochemistry in the brain.
Both an mRNA for insulin and concentrations of insulin which are
independent of those in the plasma and CSF have been demonstrated
in the brain in several other studies, as described previously. The
results of this study suggest that insulin is synthesized in neurons
from the brain and released under depolarizing conditions. The
insulin in the brain has immunological and chromatographic
properties that resemble those of pancreatic insulin. While this
evidence suggests that the material is, in fact, insulin, it is not
definitive proof. Rat proinsulin is not commercially available and
would be likely to have similar properties. Sequencing with an
amino acid analyzer would be the most appropriate method of
identifying the peptide, but it requires a larger sample of the
material than is available.
Depolarized release was measured using the solutions described
in methods in order to duplicate the methods used by Yalow to study
cholecystokinin release from synaptosomes. These unbuffered
solutions had an acidic pH after 30 min. This low pH may have
altered the release and/or the viability of the cells, although the
cells were still attached to the culture dish and appeared normal
under the microscope. Acidity was observed in all groups, but
onlythe depolarized groups showed increased release of insulin,
suggesting that it was the depolarization, and not the acidity that
induced release. In any case, glucose-induced release was measured

in a Krebs buffered solution, in order to eliminate this problem.
Insulin degradation in the release solutions was probably not
significant as degradation of insulin in binding studies, in which the
insulin is exposed to the cells for an hour or more, is generally less
than 10 percent.
It is likely that at least some of the insulin released from
neuronal cultures is also synthesized there: The cultured cells have
been removed from any peripheral insulin for at least 10 days prior
to use; the serum in which the cells are grown is plasma-derived
horse serum, which contains only one insulin, not the two observed
in the rat, and exogenous leucine is incorporated into the
immunoprecipitable insulin. The inability of some researchers to
demonstrate synthesis of insulin in the brain may be related either
to the small percentage of cells producing insulin or to their very
specific localization. In addition, insulin acts to promote survival
of brain cells in culture (115) and, thus, insulin-producing neurons
may survive preferentially in culture and may represent a larger
percentage of neurons than are present in whole brain. That
immunocytochemical evidence localizing insulin to neurons is the
result of insulin bound to surface receptors, is unlikely as glial
cells have specific, high-affinity insulin receptors as well (28) The
evidence presented here that insulin is synthesized in neurons from
the brain and is released under depolarizing conditions, in addition
to the evidence that insulin in the brain binds to specific, high-
affinity receptors and has electrical and physiological effects in the
brain suggest that insulin is a neurotransmitter in the brain.

CZ5 E E )

-.520 =
w Co c i o > 0 -.u
0 .00 EE
r- Z0> C s 0 a

cta Soo E SN E c
iEErE28! E

. Cgl l ili o -
ml 8l0 ca cis .-

E 00 I Cd 01D
oQ.0g- Cecnu -Ncuc0E

D 2 00N- !E 0 'D
0-0 (DoCM>

7-0 0 'l ) Cu E




0 0 n

uelold owu/posseOle ullnsul nr"


C M .(
Do ~~ 5 5

I E C L ci

0 0" 0 0 0)
o '0 = 0 c 0.
o ca E o.

-wIc o, c a -.

w -w._ r CC a -o '0 0
0- 0 0 0
r g .- 0-
S 0 a

E C C 2 0 E -- 1 )
0 C
$ c is (1-

oo O a


s 80


1 40

c 20

20 30 40
Elution Time (min)

x I 0i I

E 'D )
E~ cu -o(D -

5 0 o
|o 0 o0 E

,- g g .__E

SZD =2 o) 0 c '
"- E c S .- wo
0 O 0"U

-- cu v r E

: 1B E x 0 .
S00 f3 0=I 0

0- 0Z 0 cc

t: w og n 6c sI




o o .0 0
o 0 o 0

(104JUOO %) peseeleU uIlnsul





D .g) 0 =o .
a) r --- 2 u 0o
a E u E a
oJ 0 o (a, E

00 -
l -76 8 0S

a0 C a Z
o 0 E > .
L cr CL-6 '5 -- t"
S. a E M E r
a) 0)D
" rC C- 0 W ) o 0 01 .
co ( c 0 c
(D ( -OCa 2 0
w0 (U 00 OEC

iloE..2 V(
o Ec 0a OV w

0CO 8 a) .-
0 o0 (0 C (0 1 3

m o E ",- a g |
iS'^-s gg ,ol pl.S-




I )


(4s!p wwOOL/SelOWl)
peseeaeuE u!lnsul-[He]


PKC is present in both neurons and glial cells from the brain
(55,110-11). It is present in dendrites, axons, perikarya and nuclei
(110) of neurons with particularly high concentrations in
presynaptic terminals (111) and in growth cones (114). The last two
are predictable, as PKC mediates both neurotransmission and neurite
outgrowth (115). In glial cells, PKC is involved in the regulation of
differentiation (78) and membrane conductance (80). Thus, PKC is
also physiologically active in both types of cells, although its
activity takes different forms in neurons and glial cells.
Because of these differences and the existence of different
isotypes of PKC we elected to study neurons and glia in culture to
determine the relative quantities of PKC, the relative molecular
weight (Mr) of PKC and the distribution of PKC within the cells. As
TPA-activation of PKC involves translocation of the enzyme, we
chose to observe the time-course of the translocation to determine
whether this differed in the different types of cells.

Immunoreactive PKC was present in both neurons and glia.
Immunocytochemistry showed neurons staining darker than glial
cells in the same cultures with the stain distributed throughout both
types of cells. In glial cells the staining was unevenly distributed
with particularly dark staining in nuclear and perinuclear areas. In
neurons, both perikarya and processes stained. No further
differences in staining could be seen in neurons at this
magnification. No unstained cells were observed in either the
neuronal or glial cultures (Figure 4-1,2). Control cells, which
received the same treatment except for the primary antibody, had no
staining whatsoever. The concentration of iPKC was 4.60.5 (mean
standard error) times higher in neuronal than in glial cultures from
the same brains when measured by Western blot and densitometry
(Figure 4-3) of whole cell homogenates. Immunoactive PKC in both
neurons and glia had a relative molecular weight of about 80 kD as
determined by polyacrylamide gel electrophoresis followed by
Western blotting and comparison with commercial molecular weight
The cytosol contained 63 9 percent of the iPKC in glial cells
(Figure 4-4). In contrast, the neuronal iPKC resided primarily in the
membrane with only 12 2.3 percent in the cytosolic fraction when
quantitated by Western blot and densitometry (Figure 4-5).
Treatment of both neurons and glia with 100 nM TPA induced a
time-dependent translocation of iPKC from the cytosolic fraction.
In glial cells treatment with TPA decreased the cytosolic level of
iPKC to 33 7 percent of the basal level within 5 min. Within 15

min. the cytosolic iPKC was barely detectable by Western blot (6 4
percent of the control level). At 24 hours after treatment with
TPA, the concentration of iPKC in the cytosol remained low. In
contrast, the iPKC in glial membranes was increased to 150 39
percent of the control within 5 min. of the administration of TPA,
after which it declined. Within 24 hours the membrane-bound iPKC
had downregulated such that it was barely detectable as well
(Figures 4-4,6).
Treatment with 100 nM TPA induced a downregulation of iPKC in
neuronal cells. The cytosolic iPKC decreased to 59 percent of the
control concentration within 1 hour, to 16 percent within 2 hours
and did not increase again over the course of a 24 hour treatment
with TPA. The level of iPKC bound to the membrane did not change
within 15 min. after treatment with TPA but decreased to 54
percent within 1 hour and continued to decline such that the iPKC
was only 17 percent of the control iPKC concentration after 24
hours (Figures 4-5,7).

An iPKC of 80 kD is present in both neurons and glia, although
neurons express several times the cconcentration of iPKC as that
seen in glia. This is seen with both the immunocytochemistry and
the Western blot experiments. This is in agreement with studies
showing that the binding of phorbol esters is 2-3 times higher in
cultured neurons than in glial cells from the same rat brains (113).
No differences with regard to immunostaining of different regions
of the neurons were observed. As the vast majority of the neuronal

enzyme is associated with the membranous fraction the membrane
staining is very dark and differences within the cell cannot be seen.
In contrast, in glia, the majority of the iPKC is cytoplasmic. This
was not unexpected, as an earlier study reported that whereas glial
PKC was predominantly cytosolic, PKC from homogenized whole
brain tissue was predominantly membrane-bound (116). The nuclei
in glial cells appear to stain darkly for iPKC, but nuclear PKC would
not contribute to either the cytosolic or membranous fractions as
the nuclei were removed in a centrifugation step prior to the
separation of membrane and cytosol. The aPKC in the membrane-
bound fraction is latent in the liver (52,133). The concentration of
PKC associated with neuronal cytosol was, in fact, low enough to be
the result of glial cell contamination. Neuronal cultures contain 15-
20 percent glial cells, as described in methods.
Complementary studies on neuronal and glial aPKC demonstrate
that glial cells have far greater aPKC than neurons (134). The
different results obtained by immunological and bioassay suggest
that either our antibody does not recognize a large percentage of
glial aPKC or that a large percentage of the neuronal iPKC is
inactive. The first could result from the presence of different
isozymes of PKC in different types of cells, one or several of which
are not recognized by an antibody directed against the alpha and beta
(beta-1 and -2) forms of the enzyme, although these Type II and III
subtypes of PKC do represent the majority of PKC in the brain, there
is another subtype, Type I, which is found only in the brain, and
which is not recognized by our antibody (159). The existence of this
subtype, which is not immunoactive could lead to a false negative

result in studies in which iPKC was measured. In the brain,
different subspecies of PKC have been identified at different stages
of development, as discussed previously (54,57). This cannot
explain the disparity between the immunoreactive levels and enzyme
activities in neurons and glia, as experiments were done on cultures
of the same age. The increase in neuronal aPKC when stimulated
was much smaller than the increase in glial activity under the same
circumstances. This suggests that a subset of the neuronal PKC
enzymes may be physiologically inactive or less active than their
glial counterparts. The second situation could result from inactivity
of the neuronal, predominantly membrane-bound iPKC. As discussed
previously, there is evidence to support either of these hypotheses;
that is, neurons and glial cells express different isozymes of PKC
and the membrane-bound enzyme may be latent. TPA stimulates
translocation of the glial aPKC over a similar time-course and to a
similar extent as that observed immunologically. This is similar to
the situation in the periphery, in which PKC is primarily localized in
the soluble fraction and is translocated to the membrane in a time-
dependent fashion when stimulated (48). It is unlikely that the
enzyme is translocated during the process by which the cells are
prepared for electrophoresis as they are washed extensively to
remove TPA and the procedure is carried out in the presence of
calcium chelators, which prevent endogenous stimulation of
translocation as well as inhibiting proteases.
Other investigators have found that only 25-35% of the PKC in
whole brain is in the soluble fraction (51,116). It is likely that this
is due to the presence of the predominantly membrane-bound enzyme


in neurons. Our failure to observe a consistent increase in the
membranous fraction at 5-15 min. after TPA stimulation may be due
to the very small percentage of cytosolic iPKC available to be
translocated. It may be that the change in the concentration of iPKC
in the membrane is too small to be observed consistently.

o -


o aE ) oU
6 M0 c -

a)O E


c 0- =
o m o m

- E CO U) E
o oXX x

L 0 : 0'




r- C N
| ca, g

o a8 0)>
a a 0

Eo )

z 0 E= o = E =


_- 0


C -
0) co

o m
0 *~ a ( o O

-r ) C1
.2 c0

)le 0 D co

a| -C:

I -
(D"o (D -l

0m o ii -( 0 01

0) c 000

- co cV





.o E So
2 1 .E -
0- )-- -

0E oo
"g o o E
(U) (D

c CL- o U )> -

= ) -

010 a) C

CI- r L. o L E
. S, o r g w

~ ~a C.- O o'04
ilI'm l 3 EC
-0 O (DO C

C.) 0 lcoi2g.o
lm= = ) ( Cl:


o J

o o



: 0 0

(D E

US2 o Mc .c 0 "
. 'n. g 0

00 E x: -o

.'- a 0 ~ ,

0o 0 CD
a. c o M o
.o -Dw a 0
io 0.EaQ(B o -
0) C CD ECoo) O

.. cao (D 4
P^ x:5 .O**; *
m n* 8 C E -O




m I

3 (D fno

"- "E -
= o Co
3c .

E o C >- c- C

co '0 m o(D
c_ wC U)( m
- (a iEn) -0

S 80

0L E 0

icmZ a 0 C0
LL t a C~F;Q |

o 'B o
t o 10

(IOJiuo 10o %) ONd eAlloeoUnUwwUI




C 0 0


o 0

< s


S o

"- >

0 0
a 0

- -
o 68
. a
to0) i

.,- o.

o rB C

5L 0


J4 a
Irl i


(10JIOJI %)
3)ld 9AIJBounu Iw





Insulin has been shown to stimulate sugar uptake in astrocytic
glial cells from rat brain. This effect was time- and dose-
dependent with a maximal stimulation observed at 18 nM and a half-
maximal effect at 0.1 nM insulin (28). The latter value has been
shown to be well within the concentrations found in the brain (15)
indicating that the endogenous insulin is sufficient to induce this
stimulatory effect. This effect was unique to glial cells as neurons
prepared from the same brains failed to express similar properties.
Because of these and other observations it has been proposed that
insulin's lack of effect on sugar uptake in neurons may be due to the
absence of an intracellular pool of glucose transporters and/or an
inability of insulin to translocate intracellular transporters or to
activate membrane-bound transporters, or to a combination of all
three of these.
In this study we utilized TPA to determine its effects on
neuronal and glial glucose uptake as TPA has been shown to
stimulate glucose uptake in other tissues (97, 117-9). The

activation of PKC will provide us with an additional parameter to
study the differences in the regulation of neuronal and glial glucose

TPA stimulated [3H]2-dGlc uptake in glial cells in a time-
dependent manner. Glucose uptake was increased as early as 20 min.
after the administration of TPA with maximal increases occurring
after 4 hours of treatment with 100 nM TPA. The maximal level of
increase was 204.5 12.5 percent (Figure 5-1). In contrast, TPA
failed to influence 2-dGlc uptake in neuronal cultures under similar
conditions. The stimulatory effect of TPA on glial glucose uptake
was selective and was due to an increase in the number of glucose
transporters rather than a change in the Km of the transporter (119).

The differences in PKC-stimulated glucose uptake between
neurons and glial cells may be due to the different isozymes of PKC
present in the two types of cells, or to the presence of a smaller
pool of PKC available to be translocated in neurons. In other related
work in this area, our group found that phorbol ester-induced
stimulation of glucose uptake in glial cells is also concentration-
dependent with a maximal effect at 100 nM TPA. It is likely that
TPA is acting through PKC as the potency of five phorbol esters
paralleled their abilities to bind and activate PKC (119). However,
PKC has many effects on cells within the brain; even if these
effects occur by way of activastion of PKC, they may be indirect


effects that occur as a result of PKC's other actions in the brain,
such as neurotransmitter release (75-77), or alterations in
neuronal ion channels (69-710 or membrane conductance in
astrocytes (80)

(D E
a, mEo0- +l

ca o = m o -
3 0_
0 0 c0 a 0D

(D .|-@.- Wc
ca co i: l4

z o-o U o
- ca = 0 t E

o a, aO
.) : z = o o "
(=C > 0)

'0 0 E 0. 0 CL

E E 0o U-6 o 6


S~ %0)
< E

(IJ %) edn pO[H
(IOJ4uO3 %) e)IedN oicp-z[He]


Neuronal and glial insulin receptors are structurally and
physiologically distinct in several respects: neurons have a receptor
of lower molecular weight (29,32); insulin stimulates the uptake of
glucose in glia but not in neurons (28), and insulin inhibits the
uptake of norepinephrine in neurons, but not in glia (30). In addition,
there are distinctions between neurons and glia with regard to PKC:
PKC is present in higher concentration in neurons than in glia from
the same brains as demonstrated immunologically and by binding
studies with phorbol esters (113) although the activity is higher in
glial cells; different isotypes of PKC are present in neurons and glia
(55,110,111), and phorbol esters stimulate glucose uptake in glia,
but not in neurons (119). As neurons and glia differ with respect to
the physiological activities of both insulin receptors and PKC and
because PKC regulates the binding, autophosphorylation, tyrosine
kinase activity and some cellular responses of the insulin receptor
peripherally (81-3,87,91,95-7), we chose to investigate the role of
PKC in the regulation of the insulin receptor in neurons and glia from

the central nervous system to determine whether PKC might
differentially regulate these two receptors.

TPA induced a dose-dependent increase in the binding of insulin
in glial cells with no effect on binding in neurons over the same
range of concentrations (Figure 6-1). TPA treatment of glial cells
for two hours did not alter the binding of insulin at a concentration
of 1 nM TPA, began to increase binding at 10 nM TPA and caused a
maximal increase at a dose of 50 nM TPA. The ED50 was 15 nM and
50 nM caused a maximal increase of 77 percent. The effect of TPA
on the binding of insulin was time-dependent as well (Figure 6-2).
Treatment with 100 nM TPA induced an increase in the binding of
insulin in glial cells within 30 min. with a maximal increase at two
hours, followed by a decline. The amount of bound insulin stabilized
at four hours after treatment with TPA. Translocation of PKC in
glial cells from the cytosol to the membrane began within 5 min. and
was virtually complete within 15 min. The effects of TPA on insulin
binding in glial cells and 2-dGlc uptake followed TPA-stimulated
translocation of PKC and, thus, PKC may be involved in TPA's effect
on insulin binding (Figure 6-3). Treatment of neurons with 10 and
100 nM doses of TPA had no effect on the binding of insulin (Figure
Glial cells were treated with 100 nM TPA for 2 hours. Following
this treatment, competitive inhibition experiments for untreated
and TPA-treated glial cultures were conducted by adding a constant
amount of [1251]insulin (100,000 cpm/dish) and increasing amounts

of unlabelled insulin (0.167 nM-133 nM) to binding buffers. An IC50
of 40 nM insulin was observed for both curves (Figure 6-4). The data
were used for Scatchard analysis. The Kds for the high and low
affinity insulin receptors were 18.5 and 131.6 nM, respectively, in
control cells. Treatment with TPA induced a 184 percent increase in
the number of high affinity binding sites on glial cells without
altering their affinity. TPA increased the number of low affinity
sites by 74 percent while decreasing the affinity of those sites only
minimally (Table 6-1). Glial cells were exposed to three different
phorbol esters for 2 hours to evaluate the specificity of the
stimulation of insulin binding by TPA. TPA stimulated the binding of
insulin by 109 percent at a concentration of 50 nM. The same dose
which caused maximal stimulation by TPA had no effect when cells
were treated with 48-phorbol 128,13a-didecanoate (B-PDD). B-PDD
exhibited an increase of 82 percent only at a concentration of 500
nM while 4a-phorbol 12B,13a-didecanoate (a-PDD) had no effect at
any concentration (Figure 6-5). Thus, the potencies of these drugs
corresponded to their abilities to bind and activate PKC, that is, TPA
> B-PDD > the inactive analog, a-PDD (147). Pretreatment of glial
cultures for two hours with 100 nM TPA followed by its removal
resulted in a time-dependent recovery of insulin binding. The
binding recovered by 59 percent within 2 hours. Recovery was
complete within 6 hours (Figure 6-6).
The next question concerned the TPA-stimulated increase in the
number of glial insulin receptors and whether it corresponded to an
increase in responsiveness to insulin. Insulin and/or TPA was added
directly to the glial medium and the dishes were returned to the

incubator for the appropriate preincubation time prior to
quantitation of 2-dGlc uptake. Insulin induced a 34 percent increase
in glial 2-dGlc uptake when cells were treated with a dose of 167
nM for 15 min. Treatment of cells for 2 hours with 100 nM TPA
stimulated 2-dGlc uptake by 112 percent. However, TPA eliminated
the response to insulin. Administration of insulin in combination
with TPA resulted in a stimulation of 144 percent (Figure 6-7).
Thus, the binding of insulin increased, but the response to insulin did
not when glial cells were treated with TPA.

TPA stimulates the binding of insulin in glial cells from the brain
without altering neuronal binding of insulin. This stimulation in glia
occurs without a parallel increase in the responsiveness to insulin,
as demonstrated by the lack of effect of TPA on insulin stimulated
2-dGlc uptake. Thus, it represents an increase in the number of
binding sites, not necessarily an increase in the number of
receptors. The potency of the phorbol esters in effecting a
stimulation of glial insulin binding corresponds to their respective
abilities to activate PKC (147), suggesting that TPA is acting
through PKC. As with TPA stimulation of glial glucose uptake,
however, caution must be used in translating an effect of phorbol
esters to an effect of PKC. Again also, the effect may well be an
indirect effect as PKC has other effects in the brain, as described in
the introduction. TPA's effects on glial insulin binding occur more
slowly than PKC translocation, but over the same range of doses.
Thus, translocation of PKC may be involved in TPA's stimulation of

insulin binding. This is to be expected as translocation precedes and
is necessary for other PKC-induced effects, both centrally and
peripherally (61-4). This increase in the binding of insulin could
also be interpreted as an increase in ionternalization of insulin.
Further studies using an acid wash to separate bound versus
internalized insulin will be necessary to resolve this question.
As was demonstrated, treatment with TPA for 24 hours
downregulates PKC in the brain. As the binding of insulin in glial
cells recovered to the level of the control, but not below, 24 hours
after TPA administration and removal, it appears that the binding of
insulin in glial cells is not under chronic control by PKC.
PKC-induced inactivation of receptors has generally been
associated with a decrease in binding of the appropriate ligand in
studies in peripheral tissues. This example of differential
regulation of brain and peripheral insulin receptors as well as
distinct regulation of insulin receptors in neuronal and glial cells is
by no means unprecedented. Insulin, which generally downregulates
its receptor peripherally, downregulates its receptor in glia but
upregulates the receptor in neurons (148). Glucocorticoids, which
increase the binding of insulin in hepatocytes (149) and
lymphocytes(150), have no effect in adipocytes (151) and decrease
binding in an astrocytic cell line (152). As insulin has different
actions in different types of cells, it is to be expected that
regulation of insulin receptors might vary among cell and tissue
TPA induces insensitivity to insulin insofar as glial 2-dGlc
uptake is concerned. This is true even at doses of TPA which elicit a

less than maximal increase in binding. This suggests that these
glial insulin receptors, like their peripheral counterparts, may be
inactivated by PKC. Inactivation of tyrosine kinase prevents
internalization of insulin receptors in Chinese hamster ovary cells
(93). A similar proposal could be made for the glial insulin receptor.
The most intriguing aspect of this study concerned TPA's lack of
effect on neuronal insulin binding, although this does not necessarily
indicate that PKC does not inactivate the receptors. Both neurons
and glia contain a PKC which is capable of binding phorbol esters,
and neurons and glia both respond physiologically when stimulated
by phorbol esters (69-80). PKC is certainly involved in neuronal
receptor regulation as phorbol esters induce an increase in the
binding of angiotensin II by a calcium-dependent mechanism (153).
Thus, TPA's failure to alter the binding of insulin in neurons can
neither be attributed to a lack of PKC, nor to a non-functional PKC.
As PKC's effects on the insulin receptor are direct effects (90), the
different effects of TPA on the binding of insulin must necessarily
be due to differences in neuronal and glial PKC or to structural
differences within the insulin receptors themselves.

5 0
E5 '

CD9 > C

o-- w


0 o IE+

2 o c

(E .c g

p CLuC
~ ^ :
S? 3 8 s-



(IOJIUOO j0 %)
Bu!puie u!lnsul-[li


ac a
,_C- Q. +1

E c^ 0o
0) V )

EO m(

. ._ 0 c

LE c a)-
w E o

E c aC

c6.0o *D
V a li= V z C
V V 0 r_ x
S' < U. cm E a



co 0D C) Co

(IOJIuoo to %)
Bu!pu!8 u!lnsul-[Igl]

c e .
C a,

(D ) E a' E
> o e
. Eu -

73 co C 0) *)

a < 0 a-)Co

o (DO
*ii- x S

o, E 0
(6 :3 'a

o w a, C 0 C
oE o) a M V a
E E w1 LX




0 '

CO .

S _- S. -

E o" c

0( 0 --
.0 a ^- .s C( 0C
r- P c -

m -o C ato

(0 (o,
i~ ruio
.I'li x
u- 0 ? E ,-


U 4-
--U-- -


punoB u!inSUl[l/L] %

Table 6-1: Data from

the Scatchard plot of TPA effect on glial cell
[1251]insulin binding


High Affinity
Receptor number

Low Affinity
Receptor number

1.130.13 pmol/mg
0.0540.008 nM-1

5.020.81 pmol/mg
0.00760.0005 nM-1

3.210.65 pmol/mg
0.0430.006 nM-1

8.771.69 pmol/mg
0.00660.002 nM-1

-oo E
3 CL
I Ci

0 E w .u
o 1coo uj c
- a2 0 0

.0 Wa --

a. E
i.- c0. .2
a-Sc o E'


flUOOL .
zz wuoso


up!8 ul luonsu-[0gZjs
c.u .
BujpuiN o01 ui|su-|g.l

2 o m -(D
c -0 o a +-D

0 00
Z5g ca+ 7. EEa0

- o u> ... ..
. CL )
O U- C-- -n
0) 00 CM C

. cao o" -
8o D 0 a
0 ci-s w rii

cE lD~i vn
, Lu_ ol cco
>^$2.S?%* ?


up -l[l

Buipuw ujlnsu[|igg9] %

wD CD '

) -. eM(
Oc o .
. 0'- oL 0

B-8 ag
o (D 0)

< l C a, O
= = 1-, (.

S .. '8 ) 5, .o

.is ild
(D- 'S = -- a
0co. 0-0

0 E- ,
> r 0 U) C

^p-i,--- ow"
(D C- 4

cm 75 0 coQ w
E 0S 0m -r AI)CE

Full Text
xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EGZJNC6PK_2T609F INGEST_TIME 2012-09-24T14:41:54Z PACKAGE AA00011810_00001