Doctoral thesis from the Department of Molecular Bioscience, The
Wenner-Gren Institute, Stockholm University, Stockholm, Sweden
β-adrenergic signalling and novel effects in
skeletal muscle
Anette I. Öberg
Stockholm 2013
©Anette I. Öberg, Stockholm 2013
Cover by Jessica M. Olsen
ISBN 978-91-7447-624-8
Printed in Sweden by Universitetsservice AB, Stockholm 2013
Distributor: Stockholm University Library
ii
“Ever tried? Ever failed?
No matter. Try again.
Fail again. Fail better.”
Samuel Beckett
iii
iv
Abstract
Skeletal muscles have, due to their large mass, a big impact on the whole body
metabolism. There are many signals that can regulate the functions of skeletal
muscles and one such signal is activation of α- and β-adrenergic receptors (α- and
β-ARs) by the endogenous ligands epinephrine and norepinephrine. Such
activation leads to several effects which are examined in this thesis.
Stimulation of β-AR on muscle cells induces glucose uptake, an event that both
provides the muscle with energy and lowers the blood glucose levels. We
discovered that the β-ARs induce glucose uptake via the glucose transporter
protein GLUT4, which is known to be regulated by insulin. Interestingly, we
found that the intracellular pathway is partly different from that of insulin.
However, both pathways include mTOR, a kinase involved in several metabolic
processes but not previously known to be activated by β-ARs.
The most classical second messenger downstream of β-ARs is cAMP.
Surprisingly, we found this molecule to be only partly involved in the βadrenergic glucose uptake and we identified GRK2 as a key molecule for this
event.
A novel effect of β-AR stimulation presented in this thesis is the inhibition of
myosin II-dependent contractility in skeletal muscle cells. The intracellular
pathway regulating this event was different from the one regulating glucose
uptake, involving both classical and novel molecules in the β-AR pathway.
We also found another stimulus activating insulin-independent glucose uptake in
skeletal muscle cells: Shikonin, a natural compound used in traditional Chinese
herbal medicine. Shikonin increased glucose uptake in skeletal muscle cells via a
calcium- and GLUT4-dependent mechanism and improved glucose homeostasis
in diabetic rats.
Taken together, we have identified new key molecules in the β-adrenergic
signaling pathway as well as novel downstream effects. We conclude that glucose
uptake in muscles can be activated by β-adrenergic stimulation or by Shikonin
and that both improve glucose homeostasis in diabetic animals. This knowledge
can hopefully be used in the search for new drugs to combat type II diabetes.
v
vi
This thesis is based on the following papers, referred to in the text by their
respective Roman numerals.
I. Sato M# , Nodi Dehvari# , Öberg AI, Dallner OS, Sandström AL, Olsen JM,
Csikasz RI, Summers RJ, Hutchinson DS and Bengtsson T. (2013)
An insulin-independent pathway including β-adrenoceptors and mTORC2 that
translocates GLUT4 and increases glucose uptake in skeletal muscle
Manuscript
II. Öberg AI, Dehvari N and Bengtsson T. (2011)
β-adrenergic inhibition of contractility in L6 skeletal muscle cells
PLoS One 6(7):e22304
III. Öberg AI, Sato M, Merlin J, Csikasz RI, Sandström AL, Hutchinson DS,
Summers RJ and Bengtsson T. (2013)
Glucose uptake in skeletal muscle can be fully induced via the β2-adrenoceptor and
GRK2 without cAMP-production
Manuscript
IV. Öberg AI#, Yassin K#, Csikasz RI, Dehvari N, Shabalina IG, Hutchinson DS,
Wilcke M, Östenson C-G and Bengtsson, T. (2011)
Shikonin increases glucose uptake in skeletal muscle cells and improves plasma
glucose levels in diabetic Goto-Kakizaki rats
PLoS One 6(7):e22510
vii
viii
β-adrenergic signalling and novel effects in skeletal muscles
1. Introduction
11
2. Adrenergic signalling
2.1. Adrenoceptors
2.2. Expression of adrenoceptors in skeletal muscles
13
14
14
3. Adrenergic effects in skeletal muscles
3.1. Muscle growth
3.2. Ion-channels and force of contraction
3.3. Glycogen metabolism
3.4. Lipolysis
3.5. Glucose uptake
15
16
16
18
18
18
4. Adrenergic effects on glucose transporters
4.1. Glucose transporters
4.2. Glucose transporters in skeletal muscle cells
4.3. Mechanisms for increasing glucose uptake
4.4. mTOR
20
20
21
22
23
5. cAMP-dependent signalling
5.1. cAMP in skeletal muscles
5.2. PKA
5.3. Epac
5.4. Other GEFs
5.5. Ion-channels
25
25
27
29
31
31
6. Desensitisation
32
7. cAMP-independent effects
7.1. Coupling to Gi
7.2. Signalling via the βγ-subunit
7.3. GRKs and β-arrestins
35
35
36
37
8. Signalling bias
8.1. Signalling bias at the β2-AR
40
41
9. In vivo effects
43
10. Summary and conclusion
45
11. Sammanfattning på svenska
48
12. Acknowledgments
50
13. References
52
ix
Abbreviations
AC
ACC
AR
AKAP
AMP
AMPK
cAMP
CNG
CREB
DAG
4E-BP1
EDL
Epac
ERK 1/2
GEF
GLUT
GRK
GSV
GPCR
HCN
HSL
IP3
mTOR
mTORC1
mTORC2
NKCC
NMII
PDE
PKA
PKB
PIKK
PIP2
SGK1
SGLT
S6K1
SREBP
Raptor
Rictor
PI3K
PIP3
PKCα
PKI
PLC
PTX
ROCK
7TMR
VMH
Adenylate cyclase
Acetyl-CoA carboxylase
Adrenoceptors
A-kinase anchoring protein
Adenosine-5'-monophosphate
AMP activated kinase
3'-5'-cyclic adenosine monophosphate
Cyclic nucleotide-gated channels
cAMP response element binding protein
Diactylglycerol
Factor 4E binding protein 1
Extensor digitorum longus
Exchange factor directly activated by cAMP
Extracellular signal regulated kinases 1/2
Guanine-nucleotide-exchange factor
Facilitative glucose transporters
G-protein coupled kinase
GLUT4-storage vesicle
G-protein coupled receptors
Hyperpolarisation-activated cyclic nucleotide-gated channels
Hormone sensitive lipase
Inositol trisphosphate
Mammalian/mechanistic target of rapamycin
mTOR complex 1
mTOR complex 2
Na+-K+-2Cl--co-transporter
Non muscle myosin II
Phosphodiesterase
Protein kinase A
Protein kinase B, Akt
Phosphoinositide 3-kinase (PI3K)-related kinase family
Phosphatidylinositol 4,5-bisphosphate
Serum- and glucocorticoid-induced protein kinase 1
Sodium dependent glucose transporters
Ribosomal S6 kinase 1
Sterol regulatory element-binding protein
The regulatory associated protein of mTOR
Rapamycin-insensitive companion of mTOR
Phosphatidylinositol 3-kinase
Phosphatidylinositol (3,4,5)-triphosphate
Protein kinase Cα
Protein kinase inhibitor peptide
Phospholipase C
Pertussis toxin
Rho-activated kinase
Seven-transmembrane receptor
Ventromedial hypothalamus
x
1. Introduction
The main function of skeletal muscles is to create movement, but due to their
large mass they also have big impact on metabolic functions. Skeletal muscles
have high energy demand and are thus important for energy expenditure. They
have also a regulatory function by being a major site of glucose disposal and
thereby regulating blood glucose levels (DeFronzo et al. 1981). Skeletal muscles
are also suggested to have endocrine functions. The hormone irisin is for example
shown to be released from skeletal muscles upon exercise, leading to changed
expression of several metabolic genes in white adipocytes (Bostrom et al. 2012).
Skeletal muscles are subjected to regulation by stimuli such as internal cues,
hormones from the bloodstream and neurotransmitters from nerve-endings.
Hormones and neurotransmitters bind to receptors on the cell surface and thereby
induce intracellular signalling cascades. One receptors found on skeletal muscle
cells is the β2-adrenoceptor (β2-AR) which is activated by the hormone
epinephrine and the neurotransmitter norepinephrine. There are also synthetic
substances available that can interact with the β2-ARs. These can have different
effects on the receptor-activity, summarised in table 1.
Table 1. The definitions of common words referring to molecules which can interact with
a receptor.
Activation of the β2-ARs in skeletal muscles will induce several events, both
signals inside the cells as well as various downstream effects, which are described
in this thesis. To understand what is happening inside the cells after β2-AR
activation, and correlating these events with different endpoints, is a very
interesting field which opens up possibilities for designing new drugs.
Although the β2-AR is a well-studied receptor, its role in skeletal muscle is only
poorly understood and the aim of our research is to add new knowledge in this
field. My current view of the β-adrenergic signalling network and its downstream
11
effects in skeletal muscles is illustrated in Figure 1. Colour indicates our novel
findings with light pink indicating molecules that are previously known to be
activated by β-ARs, and for which we have found a new role, while dark pink
items mark molecules and events that prior to our studies were unknown to be
activated by β-ARs.
The main focus of my research has been on β-adrenergic stimulated glucose
uptake. It was previously not known which transporter proteins are involved in
this process, but in Paper I we revealed that the insulin-regulated glucose
transporter GLUT4 is needed also for β-adrenergic glucose uptake. By further
studying the intracellular pathway we discovered a new player in the β-adrenergic
signal: mTOR, a kinase involved in several metabolic events. We found mTOR to
be necessary both for β-adrenergic- and insulin-stimulated glucose uptake. In
Paper III we continued to examine the intracellular pathway and found this to be
only partly dependent on the classical second messenger cAMP. Instead we
identified the molecule GRK2 as a key player in the β-adrenergic signal to
glucose uptake.
We have also discovered a new effect of β-AR stimulation in muscles cells:
inhibition of cellular contraction. That β-adrenergic signalling can attenuate
contraction is a well-known phenomenon in smooth muscle cells while previous
results from skeletal muscles have been ambiguous. In our system, the skeletal
muscle cell line L6, activation of β-ARs clearly prevented contraction. We
examined the intracellular signal regulating this event and found the second
messenger cAMP and K+-channels to be involved (Figure 1).
Thus, we have discovered new intracellular signals, new mechanisms and new
endpoints due to β-AR signalling in skeletal muscle cells. In this thesis, βadrenergic signalling in skeletal muscles will be discussed with the main focus on
our novel findings.
12
Figure 1. My current view of β-AR signalling in skeletal muscles cells. Oval shape
indicates intracellular molecules while boxes indicate downstream events that are
stimulated by β-ARs. Pink colour marks our novel findings and dark pink indicates
molecules or events that prior to my studies were unknown to occur downstream of βARs.
2. Adrenergic signalling
Skeletal muscles are subjected to regulation from several endogenous signals e.g.
hormones and neurotransmitters. The primary function of muscles is contraction
which is mainly regulated by the release of acetylcholine from motorneurons,
while. Other functions can be regulated by other stimuli, for example the hormone
insulin which is released from the pancreas after a meal and regulates glucose
uptake and glycogen synthesis in skeletal muscle cells.
This thesis will however focus on the effects of sympathetic signalling, a stress
signal that both activates a systemic “fight or flight”-response and regulates
distinct organs in order to maintain everyday homeostasis. The sympathetic
system includes release of epinephrine from the adrenal medulla as well as release
of norepinephrine from sympathetic nerve endings. Epinephrine and
norepinephrine bind to adrenoceptors on the cell surface and thus induce several
effects such as increased heart rate, pupil dilatation, inhibition of digestion and
redirection of blood flow.
13
2.1. Adrenoceptors
There are nine subtypes of adrenoceptors (ARs) with distinct tissue expression
and pharmacological properties: α1A, α1B, α1D, α2A, α2B, α2C, β1, β2 and β3. These are
G-protein coupled receptors (GPCRs) meaning that they can act by activating Gproteins by exchanging the bound GDP to a GTP. The different AR subtypes
couple to distinct G-proteins leading to different intracellular pathways (Figure 2).
The α1-ARs couple to Gq/11 that activates phospholipase C (PLC). PLC hydrolyses
phosphatidylinositol 4,5-bisphosphate (PIP2), making inositol trisphosphate (IP3)
and diactylglycerol (DAG), which activate Ca2+- and protein kinase C (PKC)
respectively. The α2-ARs, couple to Gi/Go which inhibits cAMP-production. The
β-ARs couple to Gs that activates adenylate cyclase (AC) and thus increases
cAMP-production. Both norepinephrine and epinephrine show approximately the
same potency on the α1-AR and α2-AR. They have also the same potency on the
β1-AR, while epinephrine is much more potent than norepinephrine on β2-AR. β3AR exhibits opposite properties to β2-AR with norepinephrine as the more potent
agonist. Thus the AR subtypes differ both in pharmacological profiles as well as
intracellular effects, why the AR-expression pattern in a specific tissue greatly
influences the tissue’s response to adrenergic agonists.
Figure 2. Overview of the adrenoceptors and their main downstream effectors .
2.2. Expression of ARs in skeletal muscles
Skeletal muscle cells are reported to express both α- and β-ARs (Rattigan et al.
1986; Watson-Wright et al. 1986). To further scrutinise the expression of ARsubtypes one has to consider the heterogeneity amongst muscle fibers. Skeletal
muscle fibres can be divided into different fiber types depending on staining for
myosin adenosinetriphosphatase (mATPase), myosin isoform or expression of
metabolic enzymes. The latter classification is the most commonly used and
divides muscles into type I, type IIA and type IIB fibres. Type I fibres, also called
14
slow oxidative fibres, contract more slowly and are more resistant to fatigue than
type II fibres. They have a high content of myoglobin, mitochondria and
capillaries. Type IIA fibres are fast oxidative fibers which have more glycogen
and are less resistant to fatigue. Type IIB are fast glycolytic fibres with less
mitochondria, myoglobin and capillaries but high amounts of phospho-creatine
and glycogen, they contract rapidly and fatigue easily.
Skeletal muscles are shown to express α1- but not α2-ARs (Hutchinson et al. 2005;
Martin et al. 1990; Rattigan et al. 1986). One study suggested a difference
between fiber types since less α-ARs were detected in fast-twitch muscles
(Rattigan et al. 1986) while in another study (Martin et al. 1990) this difference
was not found. This discrepancy may depend on the fact that muscles containing
most fast-twitch fibers contain fewer blood vessels which also express α-ARs,
why there might be a difference in amount of α-ARs in the whole tissue but not in
the muscle fibres themselves.
The β-ARs, however, are considered to be the more abundant AR-form in skeletal
muscles since ten times more β-ARs than α-ARs were found with radioligand
binding (Rattigan et al. 1986). The major subtype of β-ARs present in skeletal
muscle cells is β2-AR, as shown with both radioligand binding assays and
pharmacological studies (Liggett et al. 1988; Roberts et al. 1998; Sillence et al.
1994; Watson-Wright et al. 1986). In fast-twitch muscles β2-AR is the only
expressed isoform while in slow-twitch muscles also β1-AR has been detected
(Kim et al. 1991; Watson-Wright et al. 1986). Slow-twitch muscles are found to
have higher total density of β-ARs with a clear correlation to the proportion of
type I (oxidative) fibers as well as an inverse correlation to the proportion of type
IIB (glycolytic) fibers (Jensen et al. 1995; Martin et al. 1989; Watson-Wright et
al. 1986; Williams et al. 1984).
The skeletal muscle-like cell lines L6 and C2C12 express increasing levels of β2ARs during differentiation, while β1 and β3 mRNA are not detected (Nagase et al.
2001; Nevzorova et al. 2002; Ngala et al. 2008).
Taken together the above described results demonstrated that skeletal muscle cells
express α1- and β2-ARs of which the latter are more abundant. In type I fibers low
levels of β1-ARs are found while in type II fibers and cell lines, including the L6cell which I´ve used in my studies, only β2-ARs are expressed.
3. Adrenergic effects in skeletal muscles
Activation of ARs is reported to cause several effects in skeletal muscles. Firstly
these effects will be described while the latter part of this thesis deals with the
intracellular signals which regulate these events.
15
3.1. Muscle growth
Feeding animals with β-adrenergic agonists leads to increased muscle mass and
decreased fat mass, as shown in several mammals including sheep (Baker et al.
1984), cattle (Ricks et al. 1984), swine (Jones et al. 1985) and rat (Emery et al.
1984). For this reason, the β-AR agonist clenbuterol has been used illegally by
bodybuilders who want to increase their muscle-to-fat ratio. Increased tissue mass
can be caused by either hyperplasia (increased number of cells) or hypertrophy
(enlargement of cells). Although both effects are suggested to occur in muscles
after β-agonist treatment, hypertrophy is likely to be the major effect as shown by
the finding that treatment with the β-AR agonist cimaterol increased the content
of both protein and RNA but decreased the content of DNA in muscles from lamb
(Beermann et al. 1987). The hypertrophy has been proposed to be caused partly
by increasing protein synthesis and partly by decreasing protein degradation: for
example clenbuterol treatment is shown both to induce expression of genes
involved in initiation of translation (Spurlock et al. 2006) and to inhibit
proteolysis (Navegantes et al. 2001) in rodent skeletal muscles. In addition to the
effect of increased muscle mass, β-adrenergic activation may also induce slow-tofast-twitch conversion in rats by altering the expression of myosin isoforms
(Beermann et al. 1987; Stevens et al. 2000).
Taken together these results state that increased muscle mass is a well established
effect of β-adrenergic activation and that there may be several molecular
mechanisms leading to this event.
3.2. Ion-channels and force of contraction
Various ion-channels in skeletal muscle cells are found to be affected by βadrenergic signalling. For example is K+-uptake through the Na+-K+-2Cl--cotransporter (NKCC) increased in rat soleus and plantaris muscles from rat due to
β-AR activation (Gosmanov et al. 2002). Altered ion-transport can also affect
contraction: Sympathetic stimulation can for example increase Ca2+ currents
conducted by the voltage sensitive channel CaV1, leading to increased muscle
contraction. This phenomenon is well-studied in heart but isoprenaline is shown
to increase Ca2+-flux also in cell cultures of embryonic muscles from chicken, in
which the effect was additive to that of depolarisation (Schmid et al. 1985).
Furthermore PKA is shown to phosphorylate the ryanodine receptor (a Ca2+channel involved in muscle contraction) in rabbit skeletal muscles, indicating this
to be a target of β-adrenergic signalling (Suko et al. 1993).
Since the ability to contract is the major feature of skeletal muscles it may not be
surprising that adrenergic signalling can modulate this parameter. In which
direction the contraction is altered may however vary between muscle types;
Bowman and Zaimis showed in the 1950´s that adrenergic agonists increase force
of contraction in fast-twitch muscles while having the opposite effect in slow
twitch muscles (Bowman et al. 1958). Other studies have however shown
increased force also in slow twitch muscles (Cairns et al. 1993).
16
A further clue in this issue can be obtained from our studies. We examined the
effect of β-AR signalling on contractility in the L6 cell line (Paper II) which we
found to contract in response to depletion of extracellular Ca2+. Such treatment
has previously been used as a model for myosin-dependent contraction in other
systems (Britch et al. 1980; Ivanov et al. 2004; Ma et al. 2000; Samarin et al.
2007). Thus, to induce contraction we exposed the cells to either Ca2+-free PBS or
media containing the Ca2+-chealator EDTA which made the cells contract and
finally detach (Figure 3B). We found that the β-AR agonist isoprenaline had a
pronounced inhibitory effect on cellular contraction and detachment (Figure 3C).
To further examine the signalling pathway regulating this effect we developed a
quantitative cell detachment assay in which we used both EDTA and trypsin
which made the cells get released from each other so that the detached cells were
possible to count. By this method we revealed the involvement of K+-channels,
thus adding these to the list of ion-channels that can be affected by β-adrenergic
signalling in skeletal muscle cells.
The discrepancy between different studies regarding β-adrenergic effect on
muscle contraction can have several reasons. For example, the L6-cell differs
from whole muscles in the sense that the contractile machinery in the L6-cells is
more similar to that found in embryonic than in fully differentiated muscle cell
(Whalen et al. 1979). There are also several different myosin isoforms which are
under distinct regulation. In addition to the muscle-specific myosins, muscle cells
also express non-muscle myosins, which are found in all eukaryotic cells and by
using the Ca2+-depletion we activate the non-muscle myosin rather than the
skeletal muscle specific isoform. Thus it is possible that the effect observed in our
study is more likely to occur during development or in satellite cells than in fully
differentiated muscles. One can however conclude that β-ARs can modulate the
contractility in muscle cells, but in which direction may depend both on model
system and on how the contraction is induced.
A
B
C
Figure 3. The effect of β-adrenergic stimulation on cellular contractility shown with
fluorescence microscopy of L6-myotubes stained with FITC-phalloidin which binds to
filamentous actin. (A) L6-mytoubes grown in culture media. (B) After 3 min exposure to
EDTA the cells contracted to finally detach. (C) When kept in EDTA in the presence
of 1 µM isoprenaline, the cells were unable to contract.
17
3.3. Glycogen metabolism
Furthermore, there are several studies dealing with the effect of β-adrenergic
agonists on glycogen metabolism. The level of glycogen in a muscle is
determined by the rate of glycogenolysis and glycogen synthesis, processes that
are catalyzed by the enzymes phosphorylase and glycogen synthase, respectively.
Norepinephrine can reduce glycogen levels by increasing the activity of
phosphorylase as well as decreasing the activity of glycogen synthase, as shown
in both prefused rat hindlimb and in biopsies from human (Dietz et al. 1980; Raz
et al. 1991). In the rat system this effect was blocked with propranolol, indicating
the involvement of β-ARs (Dietz et al. 1980).
The role of β-adrenergic signalling in the glycogenolysis occurring during
exercise has been investigated in several studies with somewhat ambiguous
conclusions; while the β-blocker propranolol was unable to block glycogen
breakdown in plantaris or soleus in rats after 55 min of tread-milling ((JuhlinDannfelt et al. 1982), Gorski et al. showed propranolol to block glycogen
breakdown in fast-twitch-muscles during low intensity exercise and in slowtwitch muscles during both high and low intensity exercise (Gorski et al. 1982).
Furthermore, biopsies from human thigh muscles show that propranolol decreases
the rate of glycogenolysis during dynamic exercise but not during static
contraction (Chasiotis et al. 1983). Thus β-adrenergic signalling can influence
glycogenolysis during exercise but this depends on fiber type as well as intensity
of exercise. In contrast to the results about glycogenolysis, results from L6myotubes show that treatment with β-agonist may lead to increased glycogen
synthesis (Yamamoto et al. 2007). Since this effect was observed after 2 h of
stimulation, it is possible that glycogenolysis is the acute effect while longer
treatment leads to glycogen synthesis.
3.4. Lipolysis
Although muscles store energy as glycogen rather than lipids, there are several
studies showing the presence of both triglycerides and free fatty acids in skeletal
muscles from several species (review by (Gorski 1992)). Lipolysis is increased by
adrenergic stimulation which is shown in human where injection of β-AR agonists
induces a concentration-dependent increase in glycerol levels (Hagstrom-Toft et
al. 1998). Furthermore the hormone sensitive lipase (HSL) is activated by
adrenergic signalling in skeletal muscles (Langfort et al. 1999). Thus one can
conclude that β-adrenergic signalling can alter also lipid metabolism in skeletal
muscle cells, although the physiological role for this remains to be proven.
3.5. Glucose uptake
For long adrenergic signalling was considered to inhibit glucose uptake since
activation of ARs is shown to antagonize insulin-mediated glucose uptake in
skeletal muscles from both rats and human (Chiasson et al. 1981; Jamerson et al.
1993; Lembo et al. 1994).
18
However, noradrenergic stimulation can by itself lead to an increase of glucose
uptake in skeletal muscles, as shown in several model systems: Electrical
stimulation of the ventromedial hypothalamus (VMH, an area in the brain
previously shown to be involved in adrenergic regulation of brown adipose tissue)
increased glucose uptake in skeletal muscles and heart without affecting insulin
levels (Minokoshi et al. 1994; Shimazu et al. 1991; Sudo et al. 1991).
A stimulatory effect was seen in both quadriceps, gastrocnemius, soleus and
extensor digitorum longus (EDL). This effect was also mimicked by stimulation
of VMH via microinjection of L-glutamate (Sudo et al. 1991). The increase in
glucose uptake upon electrical stimulation of VMH was blocked by guanethidine
which inhibits norepinephrine release from nerve ending, but was not altered by
adrenodemedullation (removal of adrenal medulla), indicating norepinephrine
from sympathetic nerves, rather than epinephrine from the adrenal gland, to be
responsible for the observed glucose uptake (Minokoshi et al. 1994). Furthermore,
injection of the β-AR agonist BRL37344 in anesthetized rats increased glucose
uptake in muscles (Abe et al. 1993; Liu et al. 1995). This injection augmented
insulin-levels, but BRL37344 can also act directly on muscles since it stimulates
glucose uptake in a dose-dependent manner in isolated muscles (Abe et al. 1993).
BRL37344 is however not a specific β2-AR agonist. It was firstly designed as a
β3-AR agonist and later found to also act on skeletal muscles and β2-ARs.
Furthermore, very low concentrations of BRL37344 can have effects which are
not dependent on any AR; BRL37344 is found to stimulate glucose uptake in rat
soleus and EDL (which does not express β3-ARs) and when used at the
concentration 10-11 M the BRL37344-effect could not be blocked by the β2antagonist ICI 118551 (Liu et al. 1996; Ngala et al. 2008; Ngala et al. 2009).
These results was confirmed in studies using soleus from KO-mice: 10-11 M
BRL37344 increased glucose uptake in soleus from both β2-KO and β-less mice
(Ngala et al. 2009). However, the effect of higher concentrations (10-6 M and 10-8
M) was blocked by ICI 118551 and the effect of 10-6 M BRL37344 was abolished
in the KO-models, showing these concentrations of BRL37344 to increase
glucose uptake in skeletal muscles by acting on the β2-AR (Ngala et al. 2009).
There are also contradicting results regarding the effect of β-adrenergic signalling
in skeletal muscle cells showing epinephrine to decrease glucose uptake in ex vivo
rat soleus (Sloan et al. 1978). This could however be a methodological issue since
another paper showed that isoprenaline treatment decreases glucose uptake in the
rat epitrochlearis muscles when Krebs-Henseleit Buffer is used as assay buffer,
but when supplemented with 1 % BSA, the glucose uptake is increased (Young et
al. 1985). Also in our studies, isoprenaline treatment increased glucose uptake in
ex vivo muscles and in these studies media containing BSA was used
(Paper I).The outcome may also differ with the ligand: while BRL37344
increased glucose uptake in mouse soleus, in the same study clenbuterol and
salbutamol had the opposite effect (Ngala et al. 2008). In any case norepinephrine
is found to increase glucose uptake in perfused rat gastrocnemius and soleus (Han
19
et al. 1998) and in our studies we found isoprenaline to stimulate glucose uptake
both ex vivo in soleus from Sprague-Dawly rats and in vivo in mice (Paper I).
A further indication of the physiological relevance of adrenergic stimulation of
glucose uptake is that injections with β-AR ligands stimulate glucose uptake in
muscles in living mice (Paper I, Paper III). In most of our studies we have used
muscle-like cell lines and treatment with β-agonists is shown to increase glucose
uptake in both C2C12 and L6 (Nevzorova et al. 2002; Nevzorova et al. 2006;
Ngala et al. 2008; Tanishita et al. 1997). The latter cell line is used as a model
system in Paper I-IV and in all our experiments we found the β-adrenergic ligand
isoprenaline to induce glucose uptake to the same extent as insulin (Figure 4,
Paper I, III, IV). Thus, there is a convincing body of evidence showing that βadrenergic stimulation leads to increased glucose uptake in skeletal muscle cells.
A
B
Figure 4. (A) L6-cells differentiated into myotubes were used as a model system in
Paper I-IV. (B) These cells responded to the β-adrenergic agonist isoprenaline by
increasing glucose uptake to the same level as induced by insulin.
4. Adrenergic effects on glucose transporters
Of the diverse β-adrenergic effects which are described in chapter three, the
increase in glucose uptake is the effect which our research has mainly focused on.
An increase in glucose uptake can be achieved through different molecular
mechanisms. The proteins involved in glucose transport, as well as the different
ways known to alter the transport trough these proteins, will be presented in the
following sections along with our novel results concerning β-adrenergic effects on
the glucose transporters.
4.1. Glucose transporters
Although it has been suggested that glucose may enter erythrocytes via a volume
activated “Cl--channel” (Kirk et al. 1992), the generally accepted mechanism for
glucose uptake is via transporter proteins belonging to either of the sodium
dependent glucose transporters (SGLT) or the facilitative glucose transporters
20
(GLUT). The SGLTs transport glucose via a secondary active, Na+-coupled
mechanism. The main function of these proteins is to transport glucose into the
cells from the small intestine, the renal proximal tubules and salivary gland ducts
(Sabino-Silva et al. 2010). However, low levels of SGTL5 are found in other
tissues, including skeletal muscles from cow (Zhao et al. 2005).
The GLUTs transport glucose along a gradient via facilitative diffusion. The
GLUT family comprises 14 isoforms divided into 3 classes dependent on
structural similarities (Augustin 2010). Class I comprises GLUT1-4 together with
GLUT14, a duplication of GLUT3 only found in testes (Wu et al. 2002). GLUT5,
GLUT7, GLUT9 and GLUT11 belong to Class II and can transport both glucose
and fructose. Class III includes GLUT6, GLUT8, GLUT10, GLUT12 and the
proton driven myoinositol transporter HMIT (Augustin 2010).
4.2. Glucose transporters in skeletal muscle cells
In skeletal muscle biopsies from humans the following glucose transporters are
found: GLUT1, GLUT3, GLUT4, GLUT5, GLUT8 and GLUT12 (Stuart et al.
2000; Stuart et al. 2006) with GLUT4, GLUT5 and GLUT12 as the most
abundantly expressed (Stuart et al. 2006). On mRNA level also GLUT11 was
detected, but this transporter was not found at protein level (Stuart et al. 2006).
Immunohistochemistry studies reveal a difference between fiber types: While
GLUT4 and GLUT12 are expressed at higher levels in type I than type II fibers,
GLUT5 have higher expression in type II fibers (Stuart et al. 2006).
The L6-cell line used in our studies (Paper I-IV) expresses decreasing levels of
GLUT1 and increasing levels of GLUT4 during differentiation (Mitsumoto et al.
1991). GLUT3 is reported to be expressed in L6-myoblasts and GLUT12 and
GLUT5 are found in both myoblasts and myotubes (Hajduch et al. 2003; Stuart et
al. 2009; Taha et al. 1995).
The function of the GLUT-expression-pattern in muscles is not fully understood.
The most well-studied of the current transporters are GLUT1 and GLUT4.
GLUT1 was the first cloned isoform (Mueckler et al. 1985) and is ubiquitously
expressed (with the exception of some hepatocytes) while GLUT4 is expressed in
skeletal muscles and adipose tissues where it mediates glucose import upon
insulin stimulation. GLUT3, which belongs to Class I as do GLUT1 and GLUT4,
is highly expressed in neurons and is thus the predominant GLUT-homolog in the
brain. In addition to glucose, GLUT3 can also transport other types of sugar like
galactose, mannose, maltose, xylose, as well as dehydroascorbic acid (Colville et
al. 1993). GLUT5 belongs to Class II and is more prone to transport fructose than
glucose and is primarily expressed in the small intestine. GLUT8 and GLUT12,
which belong to Class III, are both suggested to be involved in insulin-stimulated
glucose uptake; GLUT8 is found to be insulin-responsive in blastocysts from mice
(Carayannopoulos et al. 2000), but not in primary rat adipocytes, 3T3L1, HEK293
21
and CHO or CHO-cells (Augustin et al. 2005; Lisinski et al. 2001). Insulin
stimulation of human skeletal muscles induces an increase of GLUT12 in plasma
membrane comparable to the insulin-stimulated GLUT4 translocation in both L6myoblasts and humans, as shown by muscle biopsies taken before and after
insulin-infusion (Stuart et al. 2009).
4.3. Mechanisms for increasing glucose uptake
The rate of glucose uptake in a cell is regulated by several mechanisms. For
example, the amount of GLUT can be up-regulated via enhanced transcription or
translation or by reduced degradation. It may also be possible to increase the
activity of the already present GLUTs, via conformational changes in the proteins,
due to re-location to distinct compartments in the plasma membrane or
ologomerisation with other GLUTs. The possibility of altered intrinsic activity is
still somewhat controversial and is not proven to happen in muscle cells.
Since little was known about which end mechanism is responsible for the βadrenergic regulated glucose uptake, we chose to investigate this (Paper I). By
using different pharmacological inhibitor we narrowed down the number of
possible GLUT-isoforms and concluded that GLUT4 must be involved, a finding
that was confirmed by siRNA silencing of GLUT4. After ruling out the possible
involvement of transcription and translation, we formulated the hypothesis that βadrenergic stimulation might induce translocation of GLUT4, the same
mechanism as used by insulin. By immunohistochemistry staining we showed that
β-adrenergic stimulation promotes GLUT4-translocation to the same extent as
insulin does (Figure 5). Interestingly, this finding was repeated in human skeletal
muscle cells indicating this to be happening also in humans.
Figure 5. β-adrenergic signalling stimulates GLUT4-translocation. Confocal imaging of
L6-myoblasts revealed that GLUT4-translocation occurs not only after treatment with
insulin but also in response to stimulation with the β-adrenergic agonist isoprenaline
(Paper I).
That β-ARs can induce GLUT4-translocation is a completely novel finding, but
GLUT4-translocation due to other stimuli is a well-described event. Translocation
22
of GLUTs from intracellular storages to the plasma membrane was first described
in rat adipocytes in response insulin (Cushman et al. 1980; Suzuki et al. 1980) and
was later demonstrated also in skeletal muscles (Wilson et al. 1994). In the basal
state, the intracellular pool of GLUT4 can be found in either circulation
endosomes, the trans-Golig-network or special GLUT4-storage vesicles (GSVs)
of which the latter is believed to be responsible for the major part of insulinstimulated glucose uptake (Bryant et al. 2002). In muscle cells, translocation may
not only occur to the plasma membrane but also to the t-tubules (Ploug et al.
1998). There are also stimuli other that insulin that can induce GLUT4translocation in skeletal muscle cells such as the AMP activated kinase (AMPK)
and muscle contraction (Douen et al. 1990; Kurth-Kraczek et al. 1999).
Our next question was if the β-adrenergic signal may converge with any of other
pathways reported to cause GLUT4-translocaiton. According to previous studies
from our lab, AMPK is not activated by β-ARs (Hutchinson et al. 2006) while
there are indications of β-adrenergic signalling to share features with the insulinpathway (Nevzorova et al. 2006). Therefore we wanted to compare the βadrenergic pathway with that of insulin. Insulin signalling has been subjected to
thorough investigations since loss of insulin response in peripheral tissues is a
hallmark of type II diabetes. The insulin signal is a complex process involving
several pathways including phosphatidylinositol 3-kinase (PI3K),
phosphatidylinositol (3,4,5)-triphosphate (PIP3) and protein kinase B (PKB, also
known as Akt). How this signal is transduced to the point GLUT4-translocation is
not fully understood but one key molecule is AS160 which is phosphorylated by
Akt.
We found that the β-AR signal to glucose uptake does not involve the insulin
activated molecules PI3K, PIP3, Akt or AS160. However, the two pathways are
still not completely different since we discovered the involvement of a more
recently described player in the insulin cascade: the mammalian target of
rapamycin (mTOR, also known as the mechanistic target for rapamycin). Both
pharmacological studies and knock-down with siRNA showed dependence on
mTOR for both insulin and β-adrenergic stimulated glucose uptake in L6mytoubes. Furthermore, immuncytochemistry staining in the presence of an
mTOR-inhibitor revealed mTOR activation to be necessary for the GLUT4translocaiton. Thus, the findings in Paper I establishes that activation of β2-ARs
induces glucose uptake via mTORC2-dependent GLUT4-translocation.
4.4. mTOR
The finding of mTORC2 to be a key player in β-adrenergic regulated glucose
uptake settles a new role for this molecule since mTORC2 was not recognized to
be activated by adrenergic signalling, neither in skeletal muscle cells nor in any
other system (Figure 1). This is an exciting discovery and, from what we have
previously known about adrenergic signalling, also unexpected. mTOR is a
23
serine/threonine protein kinase belonging to the phosphoinositide 3-kinase
(PI3K)-related kinase family (PIKK) and is regulated by several extra- and
intracellular cues reflecting local and whole body nutritional homeostasis. It has
been described as a key sensor of energy and nutritional status and is involved in
several metabolic processes.
mTOR occurs as a part of either of the two distinct complexes named mTOR
complex 1 (mTORC1) and mTOR complex 2 (mTORC2). Both complexes
contain mTOR and the associated proteins mLST8 (also known as GβL),
DEPTOR, and Tti1/Tel2. In mTORC1 also PRAS140 as well as the regulatory
associated protein of mTOR (raptor) are found (Laplante et al. 2012).
Complex 1 is the better characterized of the two complexes and is shown to
integrate input from different stimuli such as growth factors, energy status,
oxygen-levels and amino acids. There are several intracellular signals that can
activate mTORC1, for example Akt, Erk, Wnt and AMPK (Laplante et al. 2012).
mTORC1 can also be activated by amino acid, probably via a different
mechanism than the other stimuli, and a sufficient amount of amino acids are
crucial for all activation of mTORC1.
There are numerous downstream effects of mTORC1 including protein and lipid
synthesis as well as inhibition of autophagy. It can affect protein synthesis via
direct phosphorylation of the eukaryotic translation initiation factor 4E binding
protein 1 (4E-BP1) and the ribosomal S6 kinase 1 (S6K1), two factors that
promote protein synthesis. Also lipid synthesis is unregulated by mTORC1 via
activation of the sterol regulatory element-binding protein (SREBP), a
transcription factor regulating various genes involved in cholesterol and fatty acid
synthesis.
However, when the involvement of mTOR in glucose uptake was investigated, it
was not mTORC1 but mTORC2 that was found to be crucial for both insulin and
β-ARs to induce glucose uptake (Paper I).
The mTORC2 contains the rapamycin-insensitive companion of mTOR (rictor)
along with mSin1 and protor1/2. The name rictor comes from the fact that the
mTOR-inhibitor rapamycin, a naturally occurring compound from the bacterium
Streptomyces hygrsoscopius, blocks the kinase activity only of mTORC1 and not
of mTORC2. However, long-term treatment can inhibit the assembly of
mTORC2, thus blocking also this complex in several, but not all, cell types
(Sarbassov et al. 2006).
In contrast to mTORC1, mTORC2 is not regulated by local energy/nutritional
homeostasis but is instead activated by external signals, such as insulin. mTORC2
regulates several downstream targets including Akt, the serum- and
glucocorticoid-induced protein kinase 1 (SGK-1) and PKCα. Downstream effects
of Akt is glucose uptake, cell survival and proliferation. mTORC2 can also
regulate cell shape due to its effect on PKCα (Sarbassov et al. 2004). SGK-1 is
implicated in several cellular processes including transport of ions and glucose.
24
Fewer signals are identified to induce activation of mTORC2 than of mTORC1,
but one major activator seems to be insulin, which is shown to cause PI3Kdependent autophosphorylation of mTOR at S2481 in both mTORC1 and
mTORC2 (Soliman et al. 2010). Furthermore mTORC2 is shown to
phosphorylate Akt (Sarbassov et al. 2005), a key hallmark of insulin signalling.
The physiological role of mTOR for the insulin signalling is further illustrated by
the finding that chronic rapamycin treatment can induce glucose intolerance,
hyperinsulinemia and hyperglycemia in rats due to impaired Akt-phosphorylation
and glucose uptake in skeletal muscles (Deblon et al. 2012).
In Paper I we used phosphorylation of mTOR as an indicator of activation. Insulin
induces phosphorylation of the two sites S2481 and S2448 while β-adrenergic
signalling only caused phosphorylation of S2481, indicating β-AR to operate via a
unique activation of mTOR. S2481 was phosphorylated also by the cAMP
analogue as 8-Br-cAMP, which suggests the involvement of cAMP, the classical
second messenger downstream of β2-AR, in the regulation of mTOR.
5. cAMP-dependent signalling
The following sections will deal with the different intracellular signalling
pathways that can be activated by β-ARs and their role in skeletal muscle cells.
There are several pathways known to be activated by β-ARs and the most well
studied of these is the second messenger cAMP. Thus, cAMP and its downstream
targets (Figure 6) will be discussed first.
5.1. cAMP in skeletal muscles
The main G-protein linked to β2-ARs is the stimulatory G-protein Gs which upon
receptor activation dissociates into its α- and βγ-subunits of which the former
stimulates production of 3'-5'-cyclic adenosine monophosphate (cAMP) via
activation of adenylate cyclase (AC). cAMP was the first second messenger to be
found and characterized as described in several papers by Sutherland and
colleagues (Rall et al. 1958; Sutherland et al. 1958), for which he was rewarded
the Nobel prize in 1971. Increased cAMP-levels can affect several downstream
targets in the cell including protein kinase A (PKA), Epac and ion-channels
(Figure 6).
Activation of β-AR is shown to increase intracellular levels of cAMP in several
model systems for skeletal muscles including skeletal muscle slices, perfused
muscle, muscle cell membranes and L6-cells (Han et al. 1998; Nagase et al. 2001;
Nevzorova et al. 2006; Roberts et al. 1998; Schubert et al. 1976; Sillence et al.
1994; Paper III). However, less is known about how cAMP regulates specific
endpoints.
25
Figure 6. The classical second messenger downstream of β2-ARs is cAMP, which can act
via PKA, Epac, ion-channels and possibly also other targets.
Several adrenergic effects in muscle cells, including gene-expression, glucose
uptake, inhibition of proteolysis, ion-transport as well as modulation of
contractility is mimicked by cAMP-analogues (Cairns et al. 1993; Nagase et al.
2001; Navegantes et al. 2001; Nevzorova et al. 2002; Schmid et al. 1985; Paper
II). According to these results, cAMP seems to be a central second messenger in
β-adrenergic signalling in skeletal muscles. However, the caveat of using cAMPanalogues is that this may not fully mimic what is happening when the receptor is
activated by an agonist. Most cAMP-analogues will not be degraded to the same
extent as endogenous cAMP and adding extracellular cAMP will also give a
different concentration gradient over the plasma membrane compared to receptor
activation. Furthermore, external addition of cAMP will not induce the
compartmentalisation inside the cell that occurs after stimulation of the receptor.
Thus, a more reliable way to show cAMP-dependence might be to inhibit AC.
The problem with this approach is that there are few pharmacological inhibitors of
AC that sufficiently block cAMP-production. In L6-myobutes the AC-blocker
ddA only inhibited a third of the isoprenaline induced cAMP-production at the
timepoint 15 minutes (Paper I). The SQ2256 is more efficient, but to keep cAMPlevels down for a longer time very high concentrations are needed (Paper III).
Because of these problems it is more common to show cAMP-involvement by
using the AC-activator forskolin or cAMP-analogues. This may however not lead
to the same conclusions, as is the case in glucose uptake: Addition of different
cAMP-analogues or regular cAMP induces glucose uptake in L6-cells (Nevzorova
et al. 2002; Nevzorova et al. 2006; Paper III), while pharmacological inhibition of
cAMP-production only partly inhibited the β-adrenergic stimulation of glucose
uptake (Nevzorova et al. 2006; Paper III).
From the above presented data it can be concluded that cAMP is indicated to play
a role in several events in skeletal muscles. However, since this was the first β-
26
adrenergic second messenger described, many papers have examined only the
cAMP-pathway and not looked for other possible effectors. In combination with
the methodological problems, this fact makes is difficult to determine the precise
role of cAMP in β-adrenergic effects in skeletal muscles.
5.2. PKA
Protein kinase A (PKA) was the first described effector downstream of cAMP
(Walsh et al. 1968) and is often considered to be the main mediator of cAMP
signalling. PKA is a serine/threonine kinase composed of two catalytic (C) and
two regulatory (R) subunits. When cAMP-levels increase within the cell, cAMPmolecules will bind to the R-subunits, causing dissociation of the holoenzyme and
leaving the C-subunits free to phosphorylate substrates. There are four isoforms of
the R-subunit (RIα, RIβ, RIIα and RIIβ) and four isoforms of the C-subunit (Cα,
Cβ, Cγ and Prkx). Cβ has also six splice variants, giving rise to even further
versions of PKA (Orstavik et al. 2001). However, the different PKA holoenzymes
are divided into PKAI and PKAII depending on the R-subunit isoform (with
PKAI containing either RIα or RIβ and PKAII containing RIIα or RIIβ). Both
PKAI and PKAII are expressed in mammalian skeletal muscles with RIIα as the
dominant R-isoform (Burton et al. 1997; Hoover et al. 2001; Imaizumi-Scherrer et
al. 1996).
The simultaneous presence of both isoforms in most mammalian cells could argue
for divergent functions, but it is not clear whether there are any functional
differences between the PKA-isoforms. There are however differences between
the isoforms regarding subcellular localisation since PKAII is more prone to
associate with membranes while PKAI is more prone to be localized in the
cytoplasm (Reviewed in (Scott 1991)). In skeletal muscle from rats, the two
subunits are found to have partly different localisation along the sarcomeres
(contractile units) (Perkins et al. 2001).
To understand PKA signalling, one has to consider its interaction and regulation
by the A kinase anchoring proteins (AKAPs), which serve as scaffolds for PKA.
By binding the regulatory subunit of PKA, the AKAPs can localize PKA to
specific sites in the cell as well as creating signalling complex by binding to other
proteins such as phosphatases and phosphodiesterases. Interestingly, one of the
AKAPs called gravin is found to associate also with the β2-AR (Lin et al. 2000).
The gravin-β2-AR- complex interacts with phosphatases in the absence of agonists
while agonist-stimulation induces association with PKA, PKC, GRK2, β-arrestin
and clathrin (Lin et al. 2000; Shih et al. 1999), indicating a role for this AKAP in
β-AR signalling.
There are remarkably few articles that show PKA-activation downstream of βARs in skeletal muscles. One reason for this may be that β-adrenergic activation
of PKA is considered evident, another is methodological problems. One of the
few findings of PKA activity in skeletal muscles is from rat muscles where
27
infusion of norepinephrine leads to increased PKA activity (Dietz et al. 1980).
This was shown by a kinase-activity-assay with extracted PKA. The risk with
using extracted enzymes is that this may not totally mimic what is happening
within the cell. Another method to detect PKA-activity is to examine
phosphorylations of known PKA-substrates in whole cell lysate. The disadvantage
of this approach is that these may be phosphorylated also by other kinases. PKA
can phosphorylate several downstream targets of which the most well-known are
the cAMP response element binding protein (CREB, a transcription faction) and
acetyl-CoA carboxylase (ACC, an enzyme regulation fatty acid oxidation). Also
other kinases can phosphorylate CREB, but PKA is considered to be the main
kinase for this phosphorylation.
To investigate PKA-activation in the L6-myotubes, we preformed western blot
using antibodies recognizing the phosphorylated form of either PKAphosphorylation consensus site or CREB. Both antibodes showed an increase in
phosphorylation in samples from cells treated with isoprenaline (Figure 7),
indicating that PKA is actually activated by β-adrenergic signalling in these cells.
Taken together these results strongly indicate that PKA can be activated in
skeletal muscle cells due to activation of β-ARs. The next question is if this
activation is important for any downstream effects. β-adrenergic regulation of
both Ca2+-channels and NKCC seems to involve PKA since they are mimicked by
the PKA-specific cAMP analogue (Sp-5,6-DCl-cBIMPS) and inhibited by the
PKA-inhibitor H-89 respectively (Gosmanov et al. 2002; Johnson et al. 1997).
Furthermore, β-adrenergic inhibition of proteolysis is likely to be mediated by
PKA since it is mimicked by a PKA-selective cAMP-analogue 6-Bnz-cAMP
(Baviera et al. 2010).
A
B
Figure 7. Monitoring PKA-activity in L6-myotubes by Western Blot. (A) Detection of
phosphorylation of general PKA-sites. (B) Detection of CREB-phosphorylation, an
indicator of PKA-activity.C = control cells, iso = 10 min treatment with 1 µM
isoprenaline.
Another endpoint of β-adrenergic activation in skeletal muscles is activation of
HSL, an effect that is more thoroughly described in adipocytes where it is clearly
mediated by PKA. Epinephrine stimulation increases HSL-activity and
28
phosphorylation of HSL at known PKA-regulated sites in L6-cells (Watt et al.
2006), suggesting PKA-dependence also in muscle cells. Furthermore, activation
of β-ARs is found to simulate the same HSL-phosphorylation in human muscles
(Jocken et al. 2008).
Taken together, these results indicate that β-adrenergic regulation of both Ca2+channels, NKCC and HSL involves PKA. One has however to be aware of the
methodological limitations. The available PKA-inhibitors are for example not as
selective as would be desired. The PKA-inhibitor H-89 (which was used in the
NKCC-study) can actually block Rho-activated kinase (ROCK) more efficiently
than it inhibits PKA (Davies et al. 2000). H-89 has also been indicated to act
directly on the β2-AR, both agonistically and antagonistically (Baker et al. 2003;
Penn et al. 1999). Also new inhibitors, such as KT 5720, can inhibit several
kinases in addition to PKA (Davies et al. 2000). PKA-selective cAMP-analogues
are also problematic to use since for example 6-Bnz-cAMP (used in the
proteolysis study) can inhibit the cAMP-regulated K+-channels bTREK-1
independent of its effect on PKA (Liu et al. 2009). The most reliable PKAinhibitor is at the moment the protein kinase inhibitor peptide (PKI). It is however
important to use more than one inhibitor in the same study in order to avoid
misleading results due to the potential side effect.
From the facts described above, we can conclude that surprisingly little is known
about PKA in β-adrenergic signalling in skeletal muscle cells. Although PKA is a
well-known signalling molecule and skeletal muscles are an important organ in
several respects, much remains to be discovered concerning this topic. The main
issues at the moment are the lack of reliable methods to show PKA-activation and
the need for more selective PKA-inhibitors.
5.3. Epac
For long PKA was believed to be the only effector downstream of cAMP.
However, in the late 90´s another cAMP-target was discovered: the exchange
factor directly activated by cAMP (Epac) (de Rooij et al. 1998; Kawasaki et al.
1998). Epac is a guanine-nucleotide-exchange factor (GEF) that can activate Rap1
and Rap2, small Ras-like G-proteins involved in cellular functions such as
proliferation, differentiation, apoptosis and adhesion. Binding of cAMP to Epac
induces a conformational change, so that the GEF-domain is exposed and thus
able to activate the downstream G-proteins. There are three Epac isoforms: Epac1,
Epac2A and Epac2B. Epac1 is ubiquitously expressed and is thus found in
skeletal muscle while Epac2 is not (Kawasaki et al. 1998). Epac1 contains one
cAMP-binding site while Epac2 contains two. The second cAMP-binding site in
Epac2 is however not needed for activation, at least not in vitro (de Rooij et al.
2000). Another difference is that Epac1 can be translocated to the plasma
membrane upon activation and although Epac2 may also be present there, this is
29
not dependent on cAMP (Ponsioen et al. 2009). Except these, there are no major
functional differences observed between the two isoforms.
Although PKA is often described as the major mediator of cAMP-dependent
signalling, there are also data showing Epac to be involved in adrenergic
signalling in skeletal muscles. For example, epinephrine enhances insulinstimulated Akt-phosphorylation in rat soleus and in EDL this is promotes also in
the absence of insulin (Baviera et al. 2010; Brennesvik et al. 2005). In both
systems, the Akt-phosphorylation seems to be independent of PKA since it was
neither blocked by the classical PKA-inhibitor H-89 nor mimicked by the PKAselective 6-Bnz-cAMP (Baviera et al. 2010; Brennesvik et al. 2005). Instead the
Epac-selective cAMP analogue 8-CPT-2Me-cAMP was able to induce Aktphosphorylation (Baviera et al. 2010; Brennesvik et al. 2005).
Thus, the only physiological β-adrenergic endpoint in which Epac is involved in
skeletal muscle cells is inhibition of proteolysis. According to our studies, neither
glucose uptake nor inhibition of contractility can be promoted by Epac-activation
(Paper III, Paper II). One has however to consider that there are much fewer
studies dealing with Epac- than PKA-signalling in muscles, why it is possible that
Epac can be involved in other β-adrenergic endpoints which have not been
examined yet. The β-adrenergic inhibition of glycogen synthase can for example
be mimicked by cAMP but is not reduced by the PKA-inhibitor H-89 (Liu et al.
2000), indicating another cAMP-effector that might be Epac.
The lack of pharmacological inhibitors is a major limitation in studies of Epacsignalling, leaving selective cAMP-analogues as the best way to show Epacdependence. There are however inhibitors for the downstream effector Rap1, but
since Rap1 can be activated also by other factors, this is not a reliable method to
prove the role of Epac.
Another circumstance that makes the role of Epac more difficult to elucidate is its
different possible relations to PKA. One can imagine Epac and PKA to occur in
the same pathway, acting either redundantly or synergistically. This seems to be
the case with inhibition of proteolysis since this is induced by both Epac- and
PKA-selective analogues (Baviera et al. 2010). It is also possible that these two
effectors act independently of each other and regulate distinct endpoints.
A third alternative is that PKA and Epac may oppose each other, so that one of
them fine-tunes the signal of the other. In skeletal muscles, there are indications of
the latter mechanism: the PKA-inhibitor H-89 is found to enhance the effect of
epinephrine or cAMP on Akt-phosphorylation (Baviera et al. 2010; Brennesvik et
al. 2005). Inhibition of PKA is also shown to left-shift the dose response curve for
glucose uptake, indicating a negative role for PKA (Nevzorova et al. 2006). Our
current results indicates glucose uptake not to be Epac-dependent, but this results
still indicates the possibility that different molecules downstream of the same
receptor may oppose each other.
30
5.4. Other GEFs
It has been suggested that also GEFs other than Epac may act downstream of βARs. This is somewhat controversial since there are no solid proofs for such
effectors. There are however some indications that it may happen: In vascular
smooth muscle cells glucose uptake is activated by β-ARs independent of PKA
and Epac (Kanda et al. 2007). In the same system, downregulation of Gs by longterm cholera toxin treatment inhibited the glucose uptake, indicating cAMP to be
involved. Pharmacological inhibition of Rap1, the major downstream target of
Epac, also blocked the glucose uptake, a finding that made the authors formulate
the hypothesis of a new GEF regulating Rap1 and by this regulating glucose
uptake. Of note is that in L6-mytoubes, Rap1-inhibition could not diminish βadrenergic stimulated glucose uptake (unpublished data) indicating differences in
signalling between smooth muscles and skeletal muscles cells in the respect of
adrenergic regulation of glucose transport.
5.5. Ion-channels
The third accepted target of cAMP, after PKA and Epac, is ion-channels which
are regulated by direct binding of cAMP. This is a well characterized
phenomenon in two type of ion-channels: the cyclic nucleotide-gated channels
(CNGs) and the hyperpolarisation-activated cyclic nucleotide-gated channels
(HCN) (reviewed in (Biel 2009)). CNGs are important for sensory signal
transduction from the eye and the olfactory bulb while HCNs are present in heart
and in neurons. Although neither CNGs nor HCNs are expressed in skeletal
muscles, a similar mechanism may still occur; In Paper II we discovered that βadrenergic signalling can inhibit cellular contractility in L6-myotubes (as
described in section 3.2) and examination of this pathway showed that it is
dependent on cAMP as it was both mimicked by 8-bromo-cAMP and the ACactivator forskolin and the effect of isoprenaline was inhibited by AC-inhibition.
The β-adrenergic effect on contractility was blocked by the K+-channel inhibitor
tetraethylammonium and mimicked by the K+-channel activator pinacidil
(Figure 8).
Interestingly the isoprenaline effect was neither blocked by PKA-inhibitors, nor
mimicked by Epac analogues, suggesting cAMP to act directly on ion-channels.
There are several papers showing cAMP to induce K+-efflux from atrial smooth
muscles (Ahn et al. 1995; Bieger et al. 2006; Fujii et al. 1999; Nakashima et al.
1995). In one of these papers (Bieger et al. 2006) the effect was shown to be
mediated by PKA, while others only showed cAMP-dependence. Since older
papers often state PKA-dependence based on the fact that cAMP is involved, the
contribution of ion-channels in β-adrenergic effects may be more common than
what is known today, simply because the question is too rarely asked.
31
A
B
+
Figure 8. K -channels are necessary for β-adrenergic inhibition of contractility in L6myotubes. A quantitative detachment assay was developed in which L6-cells were kept in
trypsin·EDTA for 20 min and then the detached cells in the supernatant were counted.
Control = unstimulated cells kept in trypsin·EDTA for 20 min, iso = cells pre-treated with
1 µM isoprenaline for 30 min before exposure to trypsin·EDTA in the continuous
presence of isoprenaline. In (A) black bars represents cells pretreated with 100 mM of the
K+-channel blocker tetraehtylammonium (TEA) for 10 min prior to stimulation with
isoprenaline of the concentration 1 µM (iso) or 1 nM (iso [10-9]). In (B), pre-treatment
+
with the K -channel opener pinacidil (100 µM) mimicked the effect of isoprenaline.
Of note is that cAMP can inhibit contraction in endothelial cells by acting
upstream of RhoA. This effect could not be blocked by PKA-inhibitors, nor was
RhoA phosphorylated, suggesting a PKA-independent mechanism (Essler et al.
2000). In the light of our results, it is possible that this mechanism is dependent on
K+-channels.
Taken together there are many cAMP-dependent effects in skeletal muscles and
these involve different downstream effectors: PKA affects inhibition of
proteolysis, lipolysis and ion-transport (Baviera et al. 2010; Watt et al. 2006;
Johnson et al. 1997) while Epac is involved in inhibition of proteolysis and direct
activation of ion-channels affects contractility (Baviera et al. 2010; Paper II).
Before proceeding to the cAMP-independent signal, the phenomenon of
desensitisation needs to be discussed. The reason for this is that there are
molecules which participate in desensitisation of cAMP-production and at the
same time start other signals.
6. Desensitisation
That a signal can be turned off or down-regulated, although the stimulus remains,
is called desensitisation. Desensitisation is a well-studied aspect of β-AR
signalling. Interestingly, molecules that are involved in ending the cAMP-signal
are at the same time found to be a part of other signalling pathways, which are
32
described in later sections. For this reason, the desensitisation of β-ARs will be
discussed before the cAMP-independent pathways.
Desensitisation of β-ARs can occur at several levels. Firstly, the receptor can,
upon ligand binding, change conformation to a low-affinity state which is less
prone to bind to the ligand. Secondly, there is a coupling switch from Gs to Gi,
which is stimulated by PKA-mediated phosphorylation of the intracellular parts of
the β2-AR. Thus the production of cAMP will decrease. There is also a constant
reduction of cAMP by phosphodiesterases (PDEs) that hydrolyses cAMP into
adenosine-5'-monophosphate (5'-AMP). cAMP can also be extruded from the cell,
either via diffusion or via transport by members of the multidrug-resistance–
associated proteins such as MRP4 (Sampath et al. 2002). Thirdly, the c-terminal
tail of β2-ARs can be phosphorylated by the G-protein coupled kinases (GRKs),
an event that leads to binding of β-arrestins followed by internalisation of the
receptor. In addition to these effects there is a more long-term aspect of
desensitisation occurring due to reduced receptor expression.
Desensitisation of β-ARs has been described in skeletal muscles: In L6-cells
β2-ARs are converted into a low-affinity conformation after only minutes of
agonist exposure (Pittman et al. 1980; Pittman et al. 1983) and after 10 min
stimulation less β2-ARs are found in the plasma membrane, indicating receptor
internalisation (Hardin et al. 1999). In the same system cAMP-levels start to
decrease after 30 min of continuous β-AR agonist exposure (Nevzorova et al.
2006). An interesting finding from L6-cells is that pre-treatment with the
PKA-inhibitor H89 blocks the decline in receptor density after 1 and 4 h of
agonist exposure (Hardin et al. 1999) suggesting a role for PKA in receptor
internalisation in this system.
In whole animal systems, long term treatment with β-AR agonists leads to
decreased levels of β2-AR in skeletal muscle cells of rat and guinea pigs (Elfellah
et al. 1990; Kim et al. 1992; Rothwell et al. 1987; Sillence et al. 1991) and the
ability to produce cAMP in response to acute β-AR agonist treatment is lost in
soleus from rats after being injected with isoprenaline by osmotic pumps for 2
weeks (Roberts et al. 1998).
Thus one can conclude that desensitisation is a common feature of β2-ARs
signalling also in muscle cells. The effect of this on different physiological
endpoints is however more difficult to evaluate, since it is possible that a β-AR
effect may decline for other reasons than receptor desensitisation. For
example, the increase in muscle mass due to clenbuterol feeding in rats is
attenuated after 2 weeks (McElligott et al. 1989) but this could either be due to
receptor desensitisation or to other factors limiting the speed of growth. The
same concern can be raised about desensitisation of β-adrenergic regulated
glucose uptake. We raised the question about how the glucose uptake in
muscle cells is affected by prolonged β-adrenergic stimulation and to test this,
33
differentiated L6-myotubes were kept for 4 days in medium with or without 1
µM isoprenaline. When glucose uptake was measured in these cells, the cells
kept in isoprenaline had lost their ability to respond to acute β-adrenergic
stimulation (Figure 9, unpublished data from Anna Sandström). This could
potentially be due to receptor desensitisation but since the basal levels of
glucose uptake were elevated by the 4-day treatment, it could just as well be
that the maximum glucose uptake is already reached. One could for example
imagine the numbers of GLUTs present in the cell or on the cell surface to be a
limiting factor. In rats, chronic treatment with BRL37344 either decrease or
totally abolish the response to acute BRL37344 injection in several muscles,
but also in this system the basal glucose uptake was more or less elevated,
complicating the possible interpretation of the results (Liu et al. 1995). Thus,
there can be a loss of acute β-adrenergic response due to other changes in the
cell than desensitisation of the receptor itself.
Figure 9. Chronic treatment with β-AR agonist decreases the acute effect of β-AR
stimulation on glucose uptake in L6-myotubes. Cells were either kept in normal
differentiation medium (basal) or in medium supplement with 1 µM isoprenaline for 4
days (4 d iso, black bars) before challenged with isoprenaline for 2 h (iso 2 h).
In conclusion, desensitisation of β-ARs in muscles is shown both in cell lines as
well as in whole animals. The present results are obtained with synthetic ligands
and the prevalence of this effect in the body during physiological circumstances is
not known. It is possible that the endogenous ligands are rarely present long
enough, in high enough concentrations, to actually induce desensitisation. In
either case, the phenomenon of desensitisation has high relevance for the potential
use of β-ARs as drugs targets, since it affects the possibility of long-term use.
34
7. cAMP-independent effects
Although cAMP for long was considered almost synonymous with β-adrenergic
signalling, there is a growing body of evidence for cAMP-independent events
downstream of β-ARs.
It is important to understand that there are several signals from the same receptor
and that these can be either totally independent or linked in one way or another. In
this thesis the term cAMP-independent will be used concerning pathways that do
not directly involve cAMP, although these pathways may still be correlated to
(being either modifying or modified by) cAMP.
The cAMP-independent signals can either involve G-proteins or be G-proteinindependent. Examples of the former are signalling via the βγ-subunits and
coupling to Gi while the latter includes phosphorylation of the receptor by GRKs
and subsequent β-arrestin binding (Figure 10). These effectors and their possible
role in skeletal muscles will be discussed in the following sections.
Figure 10. β2-ARs can activate signals other that cAMP, either via Gi or via totally Gprotein dependent pathways such as GRKs and β-arrestins.
7.1. Coupling to Gi
The major G-protein associated with β-ARs is Gs but also Gi, can couple to βARs. This is in concordance with the general insight in this field that GPCRs tend
to be somewhat promiscuous in their G-protein coupling. A switch in coupling
from Gs to Gi is promoted by PKA-meditated phosphorylation of the β2-AR cterminal tail (Daaka et al. 1997a), indicating Gi-dependent effects to occur
secondary to cAMP-signalling. Coupling to Gi will both lead to inhibition of the
cAMP-signal as well as inducing new effects. Examples of Gi mediated signals
downstream of β2-ARs are activation of MAPK-signalling (Daaka et al. 1997a)
and relaxation of endothelial cells (Ciccarelli et al. 2007). In rat cardiomyocytes
the adrenergic effect on contractility can also involve Gi to bigger or smaller
extent depending on ligand (Woo et al. 2009). In the same study, activation of Giα
35
was nicely shown by labelling active Gα with 32P-GTP followed by
immunoprecipitation of the different Gα-subtypes.
The involvement of Gi in β-AR signalling in skeletal muscles is indicated in rat
soleus where the Gi–inhibitor pertussis toxin (PTX) diminished the effect of β-AR
activation on ion-transport via NKCCs (Gosmanov et al. 2002). The PTXtreatment also inhibited phosphorylation of the extracellular signal regulated
kinases 1 and 2 (Erk1/2), indicating this molecule to be downstream of Gi. In the
same paper an interesting difference between muscle types was shown: in soleus
both 8-bromo-cAMP and the two AC-activators forskolin and cholera toxin
mimicked the isoprenaline effect while this was not the case in plantaris muscles
(Gosmanov et al. 2002), indicating the heterogeneity amongst different muscles.
Gi could possibly be involved in β-AR-mediated glucose uptake in muscle cells
since this was blocked by PTX in L6-myotubes (Nevzorova et al. 2006). The PTX
treatment did however reduce also the effect of insulin, which acts via a receptor
that does not activate G-proteins, indicating that PTX might have unspecific
effects. As most results on Gi-activation are obtained by the use of PTX, the effect
on insulin signalling is troublesome since it indicates PTX to give undesired
effects. Nevertheless, Gi is likely to be involved in the regulation of NKCC and
possibly also other effects downstream of the β2-AR in skeletal muscles.
7.2. Signalling via the βγ-subunit
Upon activation of GPCRs, the associated G-protein dissociates into one α- and
one βγ-subunit (Figure 11).
The classical idea was that only the α-subunit could signal to downstream
effectors, which in the case of αs would be AC, but there are also indications of
signalling via the βγ-subunit. The βγ-subunit from both Gs and Gi can regulate
Ca2+- and K+- ion-channels (Dascal 2001), but the major signalling effect of the
βγ-subunit is to promote receptor phosphorylation via GRKs (which will be
further discussed in the following section). Experiments preformed in purified
lipid bilayers show that the βγ-subunit can induce GRK-mediated phosphorylation
of the β2-AR. This was inhibited by the presence of the αi-subunit, suggesting that
the βγ-subunit needs to be dissociated from the α-subunit in order to interact with
GRKs (Pitcher et al. 1992).
36
Figure 11. (A) β2-ARs are, like all GPCRs, bound to a G-protein consisting of α-, β- and
γ-subunits. (B) Upon agonist binding, the receptor activates the G-protein by facilitating
the exchange of GDP to GTP. (C) This leads to dissociation of the subunits.
Direct interaction between the βγ-subunit and the c-terminal part of GRK has
been shown both by pulldown assay and binding assays with radiolabelled GRK
(Kim et al. 1993; Pitcher et al. 1992). Since the presence of the βγ-subunit is
shown to increase presence of GRKs on the plasma membrane, both in a
reconstituted lipid bilayer and in an over-expression cell system, the effect of the
βγ-subunit is most likely to localize GRK to the β2-AR (Daaka et al. 1997b;
Pitcher et al. 1992). This hypothesis is supported by the finding of GRK2
translocation to be inhibited by β-ARKct, a peptide corresponding to the cterminal tail of GRK2 that blocks the interaction between βγ and GRK (Daaka et
al. 1997b). Of note is that βγ-signalling downstream of β2-ARs is most often
associated with Gi but there can also be effects of βγ from Gs. This is for example
the case in endothelial cells where βγs is shown to induce relaxation (Ciccarelli et
al. 2007).
There is little or no evidence of βγ-dependent signalling downstream of β2-ARs in
skeletal muscle cells. There are however indications that Gi and GRK, molecules
often associated with βγ, could be involved in β-adrenergic signalling in muscle
cells (Gosmanov et al. 2002; Nevzorova et al. 2006; Paper III). The lack of results
about βγ-signalling, in skeletal muscles as well as in other tissues, can have
several reasons. One is methodological issues; The only accepted βγ-inhibitor is
the β-ark-ct peptide which corresponds to the GRK2 and can sequester βγ (Koch
et al. 1994). Since this peptide needs to be transfected into the cells, it is not
possible to use it in cells that are too difficult to transfect. In the GPCR-field,
there has been more focus on G-protein independent pathways than on atypical Gprotein signalling. If this mirrors what is actually happening inside the cells, or
rather depends on the current interest of scientists in the field, is hard to
determine.
37
7.3. GRKs and β-arrestins
There are signals downstream of β-ARs that do not involve any G-protein. This
phenomenon occurs at several GPCRs which has led to the use of other names for
this class of receptors like seven-transmembrane receptors (7TMRs) or
heptahelical receptors. G-protein independent signalling can be induced by
phosphorylation of the receptor by the GRKs. There are seven identified GRK
isoforms named GRK1-7. Of these GRK1 and GRK7 are expressed only in the
retina while GRK4 has the highest expression in testes and the other isoforms are
more or less ubiquitously expressed. GRK1 and GRK7 are selective rhodopsin
kinases while the others can act on different receptors including on β2-ARs.
Several GRK isoforms are detected in muscles: GRK2 and GRK5 (but not GRK3,
GRK4 or GRK6) were detected with western blot in rat soleus and EDL (Jones et
al. 2003) and in L6-cells we found expression of GRK4 and GRK6 as well when
analysing this with PCR (Paper III).
The GRKs are considered to be constitutively active and are able to phosphorylate
several sites on the β2-AR upon the conformational change of the receptor which
occurs during agonist binding. GRK4, GRK5 and GRK6 are constitutively
associated with the plasma membrane while GRK2 and GRK3 can be recruited by
Gβγ-subunits. Thus GRK2 and GRK3 are linked to G-protein activity while the
other GRK-isoforms are considered to be totally G-protein independent. Although
PKA may increase GRK2 activity (Murthy et al. 2008) it is doubtful if
cAMP/PKA has any significant role for GRK-mediated phosphorylation of βARs. Of the different G-protein dependent mechanisms, the βγ-subunit is the most
likely to stimulate GRK-signalling, as described in the previous section.
GRK-mediated phosphorylation of the β2-AR facilitates binding of β-arrestins.
This event leads to internalisation of the receptor and desensitisation of the cAMP
response. Although PKA is involved in some parts of receptor desensitisation, it is
only when the β2-AR is phosphorylated by GRK that β-arrestins are able to bind
and inhibit the GTP:ase activity of the receptor (Lohse et al. 1992).
There are four arrestin isoforms of which two are found to interact with the βARs. These are named arrestin2 and arrestin3 or, more frequently used, βarrestin1 and β-arrestin2. Both are considered to be ubiquitously expressed and
consequently both were detected in the L6-cells by western blot (our unpublished
data). Binding of β-arrestin to the receptor can in addition to the effect on
internalisation act as a scaffold for other proteins, for example Src (Luttrell et al.
1999).
ERK1 and 2 are found to be regulated by β2-ARs via two mechanisms in HEK293-cells: one fast G-protein dependent and one slower dependent of GRKs and
β-arrestins (Shenoy et al. 2006). Furthermore isoprenaline treatment can induce
translocation of both β-arrestin and c-Src to the β2-ARs and an interaction
between β-arrestin and c-Src is necessary for phosphorylation of Erk1/2 (Luttrell
et al. 1999).
38
The role of the different GRK isoforms for β2-AR phosphorylation and β-arrestin
binding has been investigated in several papers. Silencing GRKs with siRNA in
HEK-293 shows that both GRK2, GRK3, GRK5 and GRK6 can be responsible
for β2-AR phosphorylation and the biggest effect was seen with GRK2 (Violin et
al. 2006). In concordance with these results, Nobles et al. showed that GRK2 is
responsible for phosphorylation of six amino acid residues of the β2-AR while
GRK6 phosphorylates only two residues (Nobles et al. 2011).
The different phosphorylation-patterns of GRK2 and GRK6 reflect further
events: silencing of GRK2 altered the conformational change of β-arrestin upon
receptor interaction while GRK6 silencing did not affect this. Consequently,
GRK2 had bigger effect than GRK6 on Erk-phosphorylation (Nobles et al. 2011).
The GRK isoforms differ as to what extent they enhance the speed of β-arrestin
recruitment. The relative contribution of the different isoforms varies between cell
types: while GRK2 and GRK6 has significant effects on the speed of β-arrestin
recruitment in HEK-293 cells, in U2-OS cells (a human osteosarcoma cell line)
this was instead dependent on GKR2, GRK3 and GRK5, with the biggest effect of
GRK5 (Violin et al. 2006). Taken together, these results show that the GRK
isoforms vary in their effects on β2-AR, that this system is partly redundant and
that the relative contribution of the different isoforms may differ between cell
types.
When looking into skeletal muscle systems there are fewer findings of the GRK
or β-arrestin signalling. Results from C2C12-cells tell however that association
between β2-AR and β-arrestin occurs after 5 min of agonist stimulation and
remains for over 1 h (Carter et al. 2005). Erk1/2 is phosphorylated upon βadrenergic stimulation in both soleus and plantaris muscles from rat (Gosmanov et
al. 2002; Wong et al. 2001), but whether this is due to β-arrestins is not certain
since also other signals, including cAMP-dependent ones, can affect Erkphosphorylation.
In Paper II we examine the role of GKR2 in β-adrenergic glucose uptake. When
we by pharmacological inhibition of AC discovered that cAMP is not responsible
for more than a part of the β-adrenergic stimulated glucose uptake, we chose to
investigate the possible involvement of GRK2 since this molecule previously has
been shown to regulate GLUT4 trafficking in CHO-cells (Dehvari et al. 2012).
Excitingly, knocking down GRK2 in L6-myotubes significantly blocked the
response to isoprenaline, indicating GRK2 to be necessary for the β-adrenergic
stimulation of glucose uptake (Figure 12). There are few data concerning the
physiological role of GRK2 in skeletal muscles and our current data suggest a
novel role for this kinase by showing it to be involved in the regulation of glucose
uptake.
39
Figure 12. GRK2 is necessary for β-adrenergic stimulation of glucose uptake.
Measurement of glucose uptake in L6-myotubes transfected with siRNA against GRK2
(black bars) revealed that removal of GRK2 abolishes the β-adrenergic effect on glucose
uptake (n=3).
In conclusion, skeletal muscle expresses both β-arrestins and several GRKisoforms, but there are few results about their function in this system. While βarrestins have not been associated with any β-adrenergic endpoint in skeletal
muscle cells, our current results show that GRK2 is necessary for β-adrenergic
stimulation of glucose uptake in myotubes.
8. Signalling bias
As described in previous chapters, there is a diversity of both intracellular signals
and endpoints downstream of the β2-AR in skeletal muscle cells. The traditional
view of GPCR signalling was that the receptor could switch between only two
states, being either active or inactive. According to this model, the active form,
which is stabilized or induced by any agonist, could promote all the downstream
effects equally. There is however emerging evidence that this model is
incomplete. Instead, the β2-AR is found to occur in multiple conformations and
these are differently stabilized by different ligands (Kahsai et al. 2011).
Consequently, ligands can differ in their ability to regulate specific pathways and
endpoints. There are several expressions used for this phenomenon and the
terminology is unfortunately somewhat inconsistent, making a complex matter
even more confusing. It has been referred to as ligand-directed signalling,
stimulus trafficking, functional selectivity, collateral efficacy, ligand bias, biased
40
agonism, ligand-directed signalling bias or signalling bias. In this thesis however,
the term signalling bias will be used.
A key expression in the field of signalling bias is reversal of efficacy. This is
when the effects of two different ligands are tested for two distinct responses and
one drug is found to have the highest efficacy for the first endpoint while the other
drug has the highest efficacy for the second endpoint. Though reversal of efficacy
is a clear version of signalling bias, there are also other types. For example, a
receptor may not couple totally equally to different pathways. This means that a
partial agonist may only activate the strongest coupled pathway while a full
agonist can activate all pathways. How well the receptor is coupled to different
downstream effectors can differ between cell types, giving rise to what is called
cell-based functional selectivity or conditional efficacy. Strength of coupling may
also depend on further parameters, such as levels of confluence or if the cells are
grown on a surface or in suspension (Kaya et al. 2012).
A ligand may thus act as a full agonist on one pathway and at the same time be a
partial agonist or even an antagonist for another. In this perspective it gets
difficult to categorize ligands as agonists, antagonists etc. for the receptor. Instead
these expressions should rather describe the effect of a ligand on a specific
endpoint.
The phenomenon of signalling bias is of great interest from a pharmaceutical
perspective. Today almost half of all prescription drugs target GPCRs. Using
biased ligands which exclusively activate the intracellular pathway that gives the
beneficial effect could be used to increase efficacy and reduce side effects of such
drugs.
8.1. Signalling bias at the β2-AR
Signalling bias is shown to occur on the β2-AR at the level of G-protein coupling
(Wenzel-Seifert et al. 2000; Woo et al. 2009) and, what is more well documented,
on cAMP versus the β-arrestin pathway (Wisler et al. 2007). Furthermore
different ligands can induce distinct changes in β-arrestin conformation (Shukla et
al. 2008).
When comparing the effects of β-AR ligands on cAMP-accumulation and Erk1/2phosphorylation, of which the latter is used as marker for β-arrestin signalling,
some ligands are found to affect both pathways equally: isoprenaline and labetolol
are agonists for both pathways while metoprolol, biosprolol and atenolol are
inverse agonists (Galandrin et al. 2006). Other ligands exerted different or even
opposing effects, being agonistic for MAPK but antagonistic (carvedilol and
buncindolol) or even inversely agonistic (propanolol) for cAMP-formation
(Galandrin et al. 2006). Because of its promising effects on heart failure,
carvedilol is the most well studied compound with regard to biased agonism on
the β2-AR. Carvedilol has been described not to affect cAMP-production while it
at the same time stimulates receptor binding to β-arrestins, thus inducing receptor
41
internalisation and Erk1/2 phosphorylation in both HEK-cells and cardiomyocytes
(Galandrin et al. 2006; Wisler et al. 2007). This feature of carvedilol is believed to
contribute to its efficacy in the treatment of myocardial dysfunction.
Interestingly, the two endogenous AR-ligands, epinephrine and norephinephrine,
are shown to have dispersed actions on the β2-AR: in cardiomyocytes the response
to epinephrine is found to be PTX-sensitive while the response to norepinephrine
is not, indicating only epinephrine to promote the Gi-dependent pathway (Wang et
al. 2008). In this system the β2-ARs were also found to be recycled faster after
epinephrine than after norepinephrine treatment (Wang et al. 2008).
Most data concerning signalling bias at the β2-AR is obtained from non-muscle
cell lines or from cardiomyocytes, but there are also data from skeletal muscle
cells. In C2C12-cells, terbutaline and salbutamol stimulate cAMP-production to
similar levels as isoprenaline, whereas it induces β2-AR/β-arrestin-interactions to
a markedly lesser extent (Carter et al. 2005). Furthermore, salmeterol is shown to
act as a full agonist for cAMP-accumulation while having antagonistic effects on
the β2-AR/β-arrestin-interaction (Carter et al. 2005). In mouse soleus signalling
bias is found on the physiological endpoint of glucose uptake: Treatment with
10-8 M BRL37344 leads to increased glucose uptake while 10-7 M clenbuterol has
the opposite effect (Ngala et al. 2009). Both effects are blocked by β2-antagonists
and are abolished in β2-KO why they indeed are confirmed to be mediated by the
β2-receptor (Ngala et al. 2009).
We found BRL37344 to be of further interest since it can increase glucose uptake
in rat soleus (Abe et al. 1993; Liu et al. 1995; Liu et al. 1996; Ngala et al. 2008;
Ngala et al. 2009) while it in the same model system has been reported not to
affect cAMP-levels (Roberts et al. 1998). Thus, the effect of BRL37344 on the
L6-cells was examined in Paper III. Also in this system BRL37344 was found to
be a very poor agonist for cAMP and yet being able to induce glucose uptake
significantly (Figure 13).
Furthermore, in contrast to isoprenaline, BRL37344 did not phosphorylate the
PKA-substrate CREB and its effect on glucose uptake could not be blocked by the
AC-inhibitor SQ22563 (Paper III). By pharmacological inhibition we show
BRL37344 to indeed act via β2-ARs. Furthermore, transfection with siRNA
against GRK2 inhibited the effect of BRL37344. Thus, while isoprenaline
activated both cAMP and GRK2-depednent signals, BLR37344 only activates the
latter. From these data we concluded that β2-ARs can activate a different subset of
signals dependent on which ligand is activating the receptor.
42
A
B
Figure 13. Isoprenaline and BRL37344 exert the same effect on glucose uptake while
inducing different cAMP-responses. Measurement of cAMP (A) and glucose uptake (B) in
L6-myotubes in response to increasing concentrations of isoprenaline or BRL37344 at the
timepoint 2 h (n=3).
9. In vivo effects
There are two main questions concerning the effects of β-AR stimulated glucose
uptake in skeletal muscles in a whole animal system: What is the physiological
role of this and can this be used pharmacologically?
Firstly, little is known when, how or even if, β-adrenergic regulation of glucose
uptake occurs in the body. Epinephrine, which is more potent on the β2-AR than
norepinephrine (MacGregor et al. 1996), is released during acute physical or
psychological stress. During such circumstances it is logical that the muscles need
fuel for “fight or flight”. Norepinephrine is on the other hand released during
every-day life to control homeostasis. As previously mentioned, glucose uptake is
increased in skeletal muscles after artificial activation of sympathetic nerves
(Shimazu et al. 1991; Sudo et al. 1991), showing that such nerves can indeed
regulate glucose uptake in this tissue. Norepinephrine can be released with high
resolution in time and space – an example of this is noradrenergic regulation of
the heart that is very precise. It could thus be possible that sympathetic nerves
may fine-tune glucose uptake in muscles according to their need for fuel. This
hypothesis is consistent with the fact that norepinephrine output is increased in
muscles during exercise, independent of increased blood flow (Savard et al.
1989). It is well known that muscle contraction in itself leads to increased glucose
uptake independent of AMPK and it might be possible that norepinephrine release
is a part of this phenomenon.
The second question is if β-AR agonists could be used to regulate blood glucose
levels, an effect that could be used in the treatment of type II diabetes. The role of
skeletal muscles as a blood glucose clearing organ leads to the possibility of
improving blood glucose levels in diabetic patients by inducing glucose uptake in
43
skeletal muscles. This idea is utilized by anti-diabetic drugs targeting AMPK
which regulates several events including insulin-independent glucose uptake in
skeletal muscles. Furthermore, in paper IV we show that the Shikonin, a
component used in Chinese herbal medicine, can increase glucose uptake in
skeletal muscle cells via GLUT4-translocation. This compound was also able to
improve blood glucose levels in diabetic Goto-Kakizaki rats both in basal state
and after insulin injection. These findings illustrate the role for insulinindependent glucose uptake as an adequate mechanism for glucose clearance.
Concerning the possible medical use of activating β-ARs, one need however to
consider that there are effects, other than glucose uptake, that are regulated by βARs and that could have an impact on glucose metabolism. One such effect is
increased glucose output from the liver which is affected by both β- and αadrenergic stimulation (Vardanega-Peicher et al. 2000). Consequently, injection
of epinephrine leads to elevated blood glucose levels in rat (Kansal et al. 1967).
There may also be an adrenergic effect on insulin-release, although the results
concerning this are somewhat in consistent. In rats α- but not β-adrenergic,
stimulation increases insulin secretion (Kansal et al. 1967; Lacey et al. 1991)
while results from mice instead show that activation of β2-ARs induces insulin
release (Ahren et al. 1981). Also the hormone glucagon, which acts
antagonistically to insulin, is suggested to be released upon β-AR stimulation of
pancreatic islets from mice (Lacey et al. 1991).
Despite that short term treatment with β-adrenergic agonist may increase blood
glucose levels, long-term treatment has been found to have beneficial effects on
glucose homeostasis. Oral administration of clenbuterol for 5 to 6 weeks resulted
in increased glucose uptake in both muscles and fat and higher insulin and glucose
tolerance in obese Zucker rats (Castle et al. 2001; Pan et al. 2001). In these studies
it was not concluded if these effects were merely attributed to the change in body
composition with increased muscle mass and decreased abdominal fat caused by
clenbuterol treatment (Castle et al. 2001; Pan et al. 2001). However, a recent
study in old C57BL/6 mice with high-fat induced diabetes showed that the βagonist salbutamol improves glucose sensitivity (Elayan et al. 2012). In
combination with exercise this treatment also leads to improved insulin sensitivity
(Elayan et al. 2012). This is probably not only depending on changed body
composition since exercise alone could not change glucose metabolism.
Furthermore, in Paper III we injected β3-KO mice with BRL37344 twice a day for
four days and found this treatment to significantly improve glucose clearance
without changing body composition.
Activation of β-AR will affect several parameters in the body, which could cause
negative side effects of systemic β-AR administration. Not only do the β-ARs
induce growth of muscles but also of the heart which could lead to hypertrophic
cardiomyopathy, meaning that the walls of the heart are thickened. Furthermore,
44
body builders that use β-AR agonists illegally often experience tachycardia
(uncontrolled heart beating) and muscle tremor.
To overcome these problems more facts are needed about the intracellular
pathway regulating the different endpoints in the distinct tissue. Such
knowledge could make it possible to design drugs that activate one
physiological endpoint exclusively, or at least give less undesired side effects.
This could be done either by using biased ligands activating a specific function
of the β2-AR or by using drugs acting on targets further down in the signalling
cascade.
10. Summary and conclusions
Although skeletal muscles constitute the major part of the body and that activation
of β-ARs due to sympathetic signalling occurs on daily basis, the effects of β2-AR
activation in muscles have previously not been very well examined. The results
presented in this thesis have added new information to this field, both concerning
intracellular pathways as well as downstream effects. The main findings are
illustrated in figure 14.
One important consequence of β-AR stimulation of muscle cells is glucose
uptake. It has until now not been known which transporter-protein is mediating
this uptake. In Paper I we discovered that β-adrenergic stimulation induces
translocation of the insulin-sensitive transporter GLUT4. Furthermore, both
pharmacological inhibitors and siRNA-technique showed GLUT4 to be vital for
the β-AR stimulated glucose uptake. Thus the β-ARs utilize the same end
mechanism for glucose transport as does insulin. Further examination of the
upstream signal revealed that it shares some, but not all, features of the insulin
signalling pathway. In contrast to insulin the β-ARs do not activate PI3K, PIP3,
Akt or AS160. Instead we discovered that mTORC2 is a key regulator of both
insulin- and β-adrenergic stimulated glucose uptake. This is very interesting since
mTORC2 was previously not known to be activated by β-ARs.
In Paper II we discovered a new endpoint downstream of β2-ARs in L6-cells:
inhibition of contractility. Further investigations of the signalling pathway showed
that this is dependent on cAMP, but surprisingly not on the classical downstream
effectors PKA or Epac. Nor was it dependent on PKG, PI3K or PKC. Instead, this
specific signal involves K+-channels. There are other ion-channels that can be
regulated by β-ARs, but this is the first time K+-channels are shown to be
necessary for a β-AR effect in skeletal muscle cells.
Previous results concerning the possible role of cAMP in β2-AR stimulation of
glucose uptake in muscle cells are somewhat ambiguous, why we aimed to
investigate this further. Surprisingly, in Paper III we found that cAMP is only
partly involved in β-adrenergic stimulation of glucose uptake. Instead we
45
discovered that the GRK isoform GRK2 is necessary for this process. This is the
first time GRKs have been shown to regulate a physiological endpoint in
differentiated muscle cells, why this is a very interesting finding. Furthermore, it
opens up for the possibility to activate glucose uptake without simultaneously
stimulating cAMP. Excitingly, we found that the β-AR ligand BRL37344 can
activate glucose uptake via GRK2 without affecting cAMP. We showed that
BRL37344 increases glucose uptake also in vivo in mice. Furthermore, injection
of BRL37344 improved glucose clearance in glucose intolerant mice. Based on
these data, it is possible that ligands with similar properties could be used to
improve glucose levels in type II diabetes.
Paper IV shows that Shikonin, a compound found in the dried root of gromwell
(Lithospermum erythrorhizon), used in traditional Chinese medicine, can induce
glucose uptake in skeletal muscle cells. This occurred via an insulin-independent
pathway involving increased levels of intracellular Ca2+ and GLUT4translocation. Furthermore, we showed that Shikonin has beneficial effects in vivo
by lowering blood glucose levels in diabetic Goto-Kakizaki-rats.
Figure 14. In Paper I we show that activation of β2-ARs can induce glucose uptake via a
series of events involving cAMP, mTOR and GLUT4 (this signal is marked in green).
Paper II describes how activation of β2-ARs can inhibit contractility via cAMP and K+channels (pink). The work presented in Paper III shows that activation of β2-ARs can
activate glucose uptake also independently of cAMP and that this pathway involves GRK2
(yellow). In Paper IV we show that the compound Shikonin can increase glucose uptake in
muscle cells via Ca2+ and GLUT4-translocation (blue).
46
Taken together, we have revealed new features of β-AR signalling by identifying
both key molecules in the intracellular pathway as well as novel downstream
effects. We have also shown that activation of glucose uptake in muscles, either
via β-AR activation or via Shikonin, improves glucose homeostasis in diabetic
animals. To understand what happens inside the cells and correlate this with the
downstream effects is important basic knowledge, which also potentially could be
used to develop new and more selective drugs. Our studies have contributed to a
better understanding of this and our current knowledge is summarised in Table 2
with pink boxes indicating our novel findings.
In conclusion, β-AR signalling in muscle cells is a multifaceted phenomenon with
important effects on both cellular and physiological levels. Our result gives new
clues to understanding these complex events and could potentially be used in the
search for new drugs to treat metabolic disorders such as type II diabetes.
Tabel 2. The correlation between different β-adrenergic effects and intracellular signalling
pathways in skeletal muscle cells. Pink boxes indicate new findings presented in this
thesis.
47
11. Sammanfattning på Svenska
Skelettmusklernas viktigaste uppgift är att ge kroppen rörelseförmåga, men de är
även viktiga för metabolismen. Till exempel har musklerna relativt stor påverkan
på kroppens totala energiförbrukning. En annan viktig funktion är upptag av
glukos (socker), vilket sänker glukoshalten i blodet. Eftersom skelettmuskulaturen
är en så stor del av kroppen (viktmässigt är det kroppens största organ) kan även
en liten ökning av glukosupptag leda till en märkbar sänkning av blodsockret.
Det finns en rad olika signaler, inre såväl som yttre, som påverkar muskelcellerna.
Flera olika hormoner och neurotransmittorer binder till receptorer på
muskelcellernas yta. Den här avhandlingen fokuserar på vad som händer när man
aktiverar en slags receptorer som kallas β2-adrenerga receptorer (förkortas här β2AR). Det finns flera olika typer av adrenerga receptorer på samtliga organ i
kroppen där de reglerar olika fysiologiska funktioner. De adrenerga receptorerna
kan aktiveras av de kroppsegna signalsubstanserna adrenalin och noradrenalin
samt en rad syntetiskt framtagna substanser. Jag har undersökt vad som händer
inuti cellen när muskelcellernas β2-AR aktiveras.
När muskelcellernas β2-AR aktiveras påverkar detta glukosupptaget. Vi har
undersökt vilka signalmolekyler inuti cellerna som är viktiga för denna process.
Avhandlingens första artikel (Paper I) handlar om hur vi upptäckte att denna
signalväg både har likheter och skillnader gentemot den signalväg som aktiveras
av insulin. Vi visade att den adrenerga signalen till glukosupptag är beroende av
en molekyl som heter mTOR. Detta är väldigt spännande eftersom man tidigare
inte visste att mTOR kan regleras av β-ARs.
Det har inte heller varit känt tidigare vilket transportprotein som tar upp glukosen
vid adrenerg stimulering. För att undersöka detta testade vi olika alternativ och
kom fram till att glukosupptaget är beroende av transportproteinet GLUT4,
samma transportör som är ansvarig för det insulin-reglerade glukosupptaget.
Eftersom insulin-signaleringen inte fungerar vid typ II diabetes är det av stort
intresse att hitta nya, insulin-oberoende, sätt att reglera GLUT4. Vi har nu visat att
aktivering av β-ARs ger glukosupptag via GLUT4 oberoende av insulinsignalering.
I följande artiklar (Paper II och III) tittar vi närmare på vad som händer inuti en
muskelcell efter aktivering av β2-AR. Vi upptäckte att β-adrenerg aktivering
hämmar kontraktionen av muskelcellerna och undersökte vilken signalväg som
reglerar detta. Den vanligaste signalvägen som aktiveras av β2-AR innehåller
molekylerna cAMP och PKA. Vi upptäckte att den signalväg som påverkar
kontraktionen är beroende av cAMP, men inte PKA. Istället är en viss typ av
jonkanaler viktiga för just den här signalen (Paper II).
I Paper III fortsatte vi att undersöka den β-adrenerga signalvägen till
glukosupptag. Vi upptäckte att cAMP enbart är en liten del av signalvägen och att
48
PKA inte är med alls. Istället fann vi att molekylen GRK2 är en nyckelspelare i
den adrenerga regleringen av glukosupptag. Sammantaget visar dessa resultat att
den klassiska cAMP-PKA vägen inte alls är så viktig i muskelceller som man
tidigare trott. Att β-AR kan ge glukosupptag utan inblandning av cAMP är mycket
intressant eftersom cAMP påverkar många olika funktioner, både i muskler och i
andra organ. Till exempel kan frisättning av glukos från levern stimuleras av
cAMP. Vi fann att en substans som heter BRL37344 kan aktivera β-AR på ett sätt
som ger glukosupptag utan att påverka nivåerna av cAMP. Denna, eller liknande
substanser, skulle alltså potentiellt kunna öka glukosupptaget i muskler utan att ge
så många biverkningar. Spännande nog fann vi också att injektioner av
BRL37344 förbättrade glukoshomeostasen i diabetiska möss.
I avhandlingens sista artikel (Paper IV) lämnar vi β2-AR och tittar istället på ett
annat sätt att stimulera glukosupptag. Vi undersökte effekten av Shikonin, ett
ämne som använts inom traditionell kinesisk medicin. Tidigare studier har visat
att Shikonin kan påverka glukosupptag i fettceller, men vi upptäckte att detsamma
gäller även muskelceller och att dessa celler till och med är känsligare för
Shikonin än vad fettceller är. Ytterligare försök visade att injektioner av Shikonin
leder till förbättrade blodglukoshalter i diabetiska råttor.
Sammanfattningsvis har jag hittat nyckelmolekyler i den adrenerga signalvägen
och har upptäckt att aktivering av β2-AR leder till flera olika effekter, exempelvis
glukosupptag och inhibering av kontraktion, via distinkta signalvägar i
muskelcellerna. Att förstå vad som händer inuti muskelcellerna efter β2-adrenerg
aktivering och korrelera detta med olika cellulära och fysiologiska funktioner är
viktig kunskap som kan vara till nytta vid exempelvis utvecklandet av nya
behandlingar mot sjukdomar såsom typ II diabetes.
49
12. Acknowledgements
I would like to thank all people that in one way or another have contributed in the
work of making this thesis:
Tore and the GLUT-group
My supervisor Tore Bengtsson, thanks for your never ceasing enthusiasm,
support and encouragement. Our scientific discussions have been one of the best
and most inspiring parts of my entire education. I am happy to have you as my
supervisor – you are a true source of inspiration for me, both as a scientist and as
a leader.
Former and present members of the GLUT-group
Robert Csikasz, our fantastic lab manager who has helped me immensely with
everything from injecting mice to doing computer-stuff and, not at least, provided
the whole group with fantastic food. Daniel Yamamoto for making me laugh and
Olof Dallner for believing in me. Masa Sato, for your admirable lab-skills and
good mood and Nodi Dehvari, for being that skilled and encouraging PostDoc
that any stressed PhD-student would need. Things that seem impossible suddenly
get possible when you are around. Our newest members Martina Hackl and
Svetlanda Pavlova, welcome to the group! Many thanks to my wonderful
support-crew Anna Sandström, Mia Rinde and Jessica Olsen, for all your help,
practical as well as emotional, during the final making of the thesis. Thanks Anna
for fantastic bakeries, for being my extra hands in the lab and not at least for being
such a warm and down-to-earth person. Mia, for always being so kind and caring.
Thanks also for your keen proof-reading and for understanding the immense
importance of matching colors. Jessica, for your positive and encouraging attitude
– it is very fun and inspiring to work with someone who is so motivated, helpful
and loves science as much as you do. Good luck to all of you, I’m sure you will
do great!
Everyone at “Zoofys”
Professors Barbara Cannon and Jan Nedergaard, thanks for interesting
discussions and for sharing your vast knowledge and experience with me. Thanks
for all questions that have challenged me to become a better scientist. Thanks
Anders Jacobsson for your warm attitude and your ability to always see things
from a different point of view. Ida Hansen for helping me out in the lab when I
was new and utterly confused and for teaching me so much, both concerning labwork as well as other important things in life. Gabriella von Essen for your
relaxed and sensible view of life and Yan-Ling Wang for being so helpful,
Amanda González Bengtsson for your bubbling enthusiasm and Jasper de Jong
for always helping out. Anna Pauter for providing us with Polish candy and
Gustavo de Abreu Vieira for teaching me how to do glucose tolerance test and
for making me giggle. Our Western-queen Natasa Petrovic for good advices.
Everyone at the department including our present members Irina Shabalina,
50
Anastasia Kalinovich, Abbe Asadi and Ineke Luijten as well as the oldies who
took care of me in the beginning of my studies: Tomas Waldén, Annelie
Brolinson, Charlotte Mattson, Damir Zadravec and, not at least, Birgitta
Leksell and Eva Nygren who have always been there for me with warmth and
encouragement. Thanks also to the rest of the Zoofys-crew including Gelana
Yadeta, Magdalena Hernow and Irina Sabanova.
The Drug Discovery Laboratory, Monash University in Melbourne
Thanks Roger Summers, Dana Hutchinson, Jon Merlin and Bronwyn Evans
for fruitful collaborations and for making my stay in Melbourne so enjoyable. I
had a great time and learned so much during those three months!
Thanks also to the animal house staff for practical help with the animals and
Stina Höglund at IFSU for help with fluorescence imaging.
Friends and family
Thanks to “Blackanbrudarna” Elin Robertsson, Kristina “Stina” Moberg,
Johanna Gustafson Stjernborg, Natalie “Natti” Beitner, Nahal Ghanbari and
Tove Åberg for all the fun we have together and for always reminding me of
whom I want to be. Thanks scouts and other members of Immanuel Church and
Mälarscouterna that I have had the pleasure to interact with. My mentor
Kenneth Hagsten for interesting and inspiring conversations and for your
valuable support and advice. You make me see opportunities instead of problems.
Christine Öberg, my fantastic sister who is always there for me, thanks for all
encouragements and for all fun moments that we share. Without you my life
would be empty and I would have very limited knowledge about politics,
feminism and literature. My parents, Catrine and Hasse Öberg, for your endless
love and patience. Both of you are wonderful persons and will always be my
biggest source of inspiration. Thanks for your support and encouragement, in this
work as in everything else!
51
13. References
Abe H, Minokoshi Y, Shimazu T (1993) Effect of a beta 3-adrenergic agonist,
BRL35135A, on glucose uptake in rat skeletal muscle in vivo and in vitro. J
Endocrinol 139:479-486
Ahn DS, Jeong YK, Lee YH, Kang BS (1995) Activation of Ca(2+)-activated K+
channels by beta agonist in rabbit coronary smooth muscle cells. Yonsei Med J
36:232-242
Ahren B, Lundquist I (1981) Effects of selective and non-selective beta-adrenergic
agents on insulin secretion in vivo. Eur J Pharmacol 71:93-104
Augustin R, Riley J, Moley KH (2005) GLUT8 contains a [DE]XXXL[LI] sorting motif
and localizes to a late endosomal/lysosomal compartment. Traffic 6:1196-1212
Augustin R (2010) The protein family of glucose transport facilitators: It's not only
about glucose after all. IUBMB Life 62:315-333
Baker JG, Hall IP, Hill SJ (2003) Agonist and inverse agonist actions of beta-blockers at
the human beta 2-adrenoceptor provide evidence for agonist-directed signaling.
Mol Pharmacol 64:1357-1369
Baker PK, Dalrymple RH, Ingle DL, Ricks CA (1984) Use of a β-adrenergic agonist to
alter muscle and fat deposition in lambs. J Anim Sci 59:1256-1261
Baviera AM, Zanon NM, Navegantes LC, Kettelhut IC (2010) Involvement of
cAMP/Epac/PI3K-dependent pathway in the antiproteolytic effect of epinephrine
on rat skeletal muscle. Mol Cell Endocrinol 315:104-112
Beermann DH, Butler WR, Hogue DE, Fishell VK, Dalrymple RH, Ricks CA, Scanes
CG (1987) Cimaterol-induced muscle hypertrophy and altered endocrine status in
lambs. J Anim Sci 65:1514-1524
Bieger D, Parai K, Ford CA, Tabrizchi R (2006) beta-adrenoceptor mediated responses
in rat pulmonary artery: putative role of TASK-1 related K channels. Naunyn
Schmiedebergs Arch Pharmacol 373:186-196
Biel M (2009) Cyclic nucleotide-regulated cation channels. J Biol Chem 284:9017-9021
Bostrom P, Wu J, Jedrychowski MP, Korde A, Ye L, Lo JC, Rasbach KA, Bostrom EA,
Choi JH, Long JZ, Kajimura S, Zingaretti MC, Vind BF, Tu H, Cinti S, Hojlund
K, Gygi SP, Spiegelman BM (2012) A PGC1-alpha-dependent myokine that
drives brown-fat-like development of white fat and thermogenesis. Nature
481:463-468
Bowman WC, Zaimis E (1958) The effects of adrenaline, noradrenaline and
isoprenaline on skeletal muscle contractions in the cat. J Physiol 144:92-107
Brennesvik EO, Ktori C, Ruzzin J, Jebens E, Shepherd PR, Jensen J (2005) Adrenaline
potentiates insulin-stimulated PKB activation via cAMP and Epac: implications
for cross talk between insulin and adrenaline. Cell Signal 17:1551-1559
Britch M, Allen TD (1980) The modulation of cellular contractility and adhesion by
trypsin and EGTA. Exp Cell Res 125:221-231
Bryant NJ, Govers R, James DE (2002) Regulated transport of the glucose transporter
GLUT4. Nat Rev Mol Cell Biol 3:267-277
Burton KA, Johnson BD, Hausken ZE, Westenbroek RE, Idzerda RL, Scheuer T, Scott
JD, Catterall WA, McKnight GS (1997) Type II regulatory subunits are not
required for the anchoring-dependent modulation of Ca2+ channel activity by
cAMP-dependent protein kinase. Proc Natl Acad Sci U S A 94:11067-11072
Cairns SP, Dulhunty AF (1993) The effects of beta-adrenoceptor activation on
contraction in isolated fast- and slow-twitch skeletal muscle fibres of the rat. Br J
Pharmacol 110:1133-1141
52
Carayannopoulos MO, Chi MM, Cui Y, Pingsterhaus JM, McKnight RA, Mueckler M,
Devaskar SU, Moley KH (2000) GLUT8 is a glucose transporter responsible for
insulin-stimulated glucose uptake in the blastocyst. Proc Natl Acad Sci U S A
97:7313-7318
Carter AA, Hill SJ (2005) Characterization of isoprenaline- and salmeterol-stimulated
interactions between beta2-adrenoceptors and beta-arrestin 2 using betagalactosidase complementation in C2C12 cells. J Pharmacol Exp Ther 315:839848
Castle A, Yaspelkis BB,3rd, Kuo CH, Ivy JL (2001) Attenuation of insulin resistance by
chronic beta2-adrenergic agonist treatment possible muscle specific contributions.
Life Sci 69:599-611
Chasiotis D, Brandt R, Harris RC, Hultman E (1983) Effects of beta-blockade on
glycogen metabolism in human subjects during exercise. Am J Physiol 245:E16670
Chiasson JL, Shikama H, Chu DT, Exton JH (1981) Inhibitory effect of epinephrine on
insulin-stimulated glucose uptake by rat skeletal muscle. J Clin Invest 68:706-713
Ciccarelli M, Cipolletta E, Santulli G, Campanile A, Pumiglia K, Cervero P, Pastore L,
Astone D, Trimarco B, Iaccarino G (2007) Endothelial beta2 adrenergic signaling
to AKT: role of Gi and SRC. Cell Signal 19:1949-1955
Colville CA, Seatter MJ, Jess TJ, Gould GW, Thomas HM (1993) Kinetic analysis of
the liver-type (GLUT2) and brain-type (GLUT3) glucose transporters in Xenopus
oocytes: substrate specificities and effects of transport inhibitors. Biochem J 290 (
Pt 3):701-706
Cushman SW, Wardzala LJ (1980) Potential mechanism of insulin action on glucose
transport in the isolated rat adipose cell. Apparent translocation of intracellular
transport systems to the plasma membrane. J Biol Chem 255:4758-4762
Daaka Y, Luttrell LM, Lefkowitz RJ (1997a) Switching of the coupling of the beta2adrenergic receptor to different G proteins by protein kinase A. Nature 390:88-91
Daaka Y, Pitcher JA, Richardson M, Stoffel RH, Robishaw JD, Lefkowitz RJ (1997b)
Receptor and G betagamma isoform-specific interactions with G protein-coupled
receptor kinases. Proc Natl Acad Sci U S A 94:2180-2185
Dascal N (2001) Ion-channel regulation by G proteins. Trends Endocrinol Metab
12:391-398
Davies SP, Reddy H, Caivano M, Cohen P (2000) Specificity and mechanism of action
of some commonly used protein kinase inhibitors. Biochem J 351:95-105
de Rooij J, Zwartkruis FJ, Verheijen MH, Cool RH, Nijman SM, Wittinghofer A, Bos
JL (1998) Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by
cyclic AMP. Nature 396:474-477
de Rooij J, Rehmann H, van Triest M, Cool RH, Wittinghofer A, Bos JL (2000)
Mechanism of regulation of the Epac family of cAMP-dependent RapGEFs. J Biol
Chem 275:20829-20836
Deblon N, Bourgoin L, Veyrat-Durebex C, Peyrou M, Vinciguerra M, Caillon A,
Maeder C, Fournier M, Montet X, Rohner-Jeanrenaud F, Foti M (2012) Chronic
mTOR inhibition by rapamycin induces muscle insulin resistance despite weight
loss in rats. Br J Pharmacol 165:2325-2340
DeFronzo RA, Jacot E, Jequier E, Maeder E, Wahren J, Felber JP (1981) The effect of
insulin on the disposal of intravenous glucose. Results from indirect calorimetry
and hepatic and femoral venous catheterization. Diabetes 30:1000-1007
Dehvari N, Hutchinson DS, Nevzorova J, Dallner OS, Sato M, Kocan M, Merlin J,
Evans BA, Summers RJ, Bengtsson T (2012) beta(2)-Adrenoceptors increase
53
translocation of GLUT4 via GPCR kinase sites in the receptor C-terminal tail. Br J
Pharmacol 165:1442-1456
Dietz MR, Chiasson JL, Soderling TR, Exton JH (1980) Epinephrine regulation of
skeletal muscle glycogen metabolism. Studies utilizing the perfused rat hindlimb
preparation. J Biol Chem 255:2301-2307
Douen AG, Ramlal T, Rastogi S, Bilan PJ, Cartee GD, Vranic M, Holloszy JO, Klip A
(1990) Exercise induces recruitment of the "insulin-responsive glucose
transporter". Evidence for distinct intracellular insulin- and exercise-recruitable
transporter pools in skeletal muscle. J Biol Chem 265:13427-13430
Elayan H, Milic M, Sun P, Gharaibeh M, Ziegler MG (2012) Chronic beta2 adrenergic
agonist, but not exercise, improves glucose handling in older type 2 diabetic mice.
Cell Mol Neurobiol 32:871-877
Elfellah MS, Reid JL (1990) Regulation of beta-adrenoceptors in the guinea pig left
ventricle and skeletal muscle following chronic agonist treatment. Eur J Pharmacol
182:387-392
Emery PW, Rothwell NJ, Stock MJ, Winter PD (1984) Chronic effects of beta 2adrenergic agonists on body composition and protein synthesis in the rat. Biosci
Rep 4:83-91
Essler M, Staddon JM, Weber PC, Aepfelbacher M (2000) Cyclic AMP blocks bacterial
lipopolysaccharide-induced myosin light chain phosphorylation in endothelial
cells through inhibition of Rho/Rho kinase signaling. J Immunol 164:6543-6549
Fujii K, Onaka U, Goto K, Abe I, Fujishima M (1999) Impaired isoproterenol-induced
hyperpolarization in isolated mesenteric arteries of aged rats. Hypertension
34:222-228
Galandrin S, Bouvier M (2006) Distinct signaling profiles of beta1 and beta2 adrenergic
receptor ligands toward adenylyl cyclase and mitogen-activated protein kinase
reveals the pluridimensionality of efficacy. Mol Pharmacol 70:1575-1584
Gorski J, Pietrzyk K (1982) The effect of beta-adrenergic receptor blockade on
intramuscular glycogen mobilization during exercise in the rat. Eur J Appl Physiol
Occup Physiol 48:201-205
Gorski J (1992) Muscle triglyceride metabolism during exercise. Can J Physiol
Pharmacol 70:123-131
Gosmanov AR, Wong JA, Thomason DB (2002) Duality of G protein-coupled
mechanisms for beta-adrenergic activation of NKCC activity in skeletal muscle.
Am J Physiol Cell Physiol 283:C1025-32
Hagstrom-Toft E, Enoksson S, Moberg E, Bolinder J, Arner P (1998) beta-Adrenergic
regulation of lipolysis and blood flow in human skeletal muscle in vivo. Am J
Physiol 275:E909-16
Hajduch E, Litherland GJ, Turban S, Brot-Laroche E, Hundal HS (2003) Insulin
regulates the expression of the GLUT5 transporter in L6 skeletal muscle cells.
FEBS Lett 549:77-82
Han XX, Bonen A (1998) Epinephrine translocates GLUT-4 but inhibits insulinstimulated glucose transport in rat muscle. Am J Physiol 274:E700-7
Hardin AO, Lima JJ (1999) Beta 2-adrenoceptor agonist-induced down-regulation after
short-term exposure. J Recept Signal Transduct Res 19:835-852
Hoover F, Mathiesen I, Skalhegg BS, Lomo T, Tasken K (2001) Differential expression
and regulation of the PKA signalling pathway in fast and slow skeletal muscle.
Anat Embryol (Berl) 203:193-201
54
Hutchinson DS, Bengtsson T (2005) alpha1A-adrenoceptors activate glucose uptake in
L6 muscle cells through a phospholipase C-, phosphatidylinositol-3 kinase-, and
atypical protein kinase C-dependent pathway. Endocrinology 146:901-912
Hutchinson DS, Bengtsson T (2006) AMP-activated protein kinase activation by
adrenoceptors in L6 skeletal muscle cells: mediation by alpha1-adrenoceptors
causing glucose uptake. Diabetes 55:682-690
Imaizumi-Scherrer T, Faust DM, Benichou JC, Hellio R, Weiss MC (1996)
Accumulation in fetal muscle and localization to the neuromuscular junction of
cAMP-dependent protein kinase A regulatory and catalytic subunits RI alpha and
C alpha. J Cell Biol 134:1241-1254
Ivanov AI, McCall IC, Parkos CA, Nusrat A (2004) Role for actin filament turnover and
a myosin II motor in cytoskeleton-driven disassembly of the epithelial apical
junctional complex. Mol Biol Cell 15:2639-2651
Jamerson KA, Julius S, Gudbrandsson T, Andersson O, Brant DO (1993) Reflex
sympathetic activation induces acute insulin resistance in the human forearm.
Hypertension 21:618-623
Jensen J, Brors O, Dahl HA (1995) Different beta-adrenergic receptor density in
different rat skeletal muscle fibre types. Pharmacol Toxicol 76:380-385
Jocken JW, Roepstorff C, Goossens GH, van der Baan P, van Baak M, Saris WH, Kiens
B, Blaak EE (2008) Hormone-sensitive lipase serine phosphorylation and glycerol
exchange across skeletal muscle in lean and obese subjects: effect of betaadrenergic stimulation. Diabetes 57:1834-1841
Johnson BD, Brousal JP, Peterson BZ, Gallombardo PA, Hockerman GH, Lai Y,
Scheuer T, Catterall WA (1997) Modulation of the cloned skeletal muscle L-type
Ca2+ channel by anchored cAMP-dependent protein kinase. J Neurosci 17:12431255
Jones RW, Easter RA, McKeith FK, Dalrymple RH, Maddock HM, Bechtel PJ (1985)
Effect of the beta-adrenergic agonist cimaterol (CL 263,780) on the growth and
carcass characteristics of finishing swine. J Anim Sci 61:905-913
Jones SW, Baker DJ, Greenhaff PL (2003) G protein-coupled receptor kinases 2 and 5
are differentially expressed in rat skeletal muscle and remain unchanged following
beta2-agonist administration. Exp Physiol 88:277-284
Juhlin-Dannfelt AC, Terblanche SE, Fell RD, Young JC, Holloszy JO (1982) Effects of
beta-adrenergic receptor blockade on glycogenolysis during exercise. J Appl
Physiol 53:549-554
Kahsai AW, Xiao K, Rajagopal S, Ahn S, Shukla AK, Sun J, Oas TG, Lefkowitz RJ
(2011) Multiple ligand-specific conformations of the beta2-adrenergic receptor.
Nat Chem Biol 7:692-700
Kanda Y, Watanabe Y (2007) Adrenaline increases glucose transport via a Rap1p38MAPK pathway in rat vascular smooth muscle cells. Br J Pharmacol 151:476482
Kansal PC, Buse MG (1967) The effect of adrenergic blocking agents on plasma insulin
and blood glucose during urethan or epinephrine induced hyperglycemia.
Metabolism 16:548-556
Kawasaki H, Springett GM, Mochizuki N, Toki S, Nakaya M, Matsuda M, Housman
DE, Graybiel AM (1998) A family of cAMP-binding proteins that directly activate
Rap1. Science 282:2275-2279
Kaya AI, Onaran HO, Ozcan G, Ambrosio C, Costa T, Balli S, Ugur O (2012) Cell
contact-dependent functional selectivity of beta2-adrenergic receptor ligands in
55
stimulating cAMP accumulation and extracellular signal-regulated kinase
phosphorylation. J Biol Chem 287:6362-6374
Kim CM, Dion SB, Benovic JL (1993) Mechanism of beta-adrenergic receptor kinase
activation by G proteins. J Biol Chem 268:15412-15418
Kim YS, Sainz RD, Molenaar P, Summers RJ (1991) Characterization of beta 1- and
beta 2-adrenoceptors in rat skeletal muscles. Biochem Pharmacol 42:1783-1789
Kim YS, Sainz RD, Summers RJ, Molenaar P (1992) Cimaterol reduces beta-adrenergic
receptor density in rat skeletal muscles. J Anim Sci 70:115-122
Kirk K, Ellory JC, Young JD (1992) Transport of organic substrates via a volumeactivated channel. J Biol Chem 267:23475-23478
Koch WJ, Hawes BE, Inglese J, Luttrell LM, Lefkowitz RJ (1994) Cellular expression
of the carboxyl terminus of a G protein-coupled receptor kinase attenuates G beta
gamma-mediated signaling. J Biol Chem 269:6193-6197
Kurth-Kraczek EJ, Hirshman MF, Goodyear LJ, Winder WW (1999) 5' AMP-activated
protein kinase activation causes GLUT4 translocation in skeletal muscle. Diabetes
48:1667-1671
Lacey RJ, Berrow NS, Scarpello JH, Morgan NG (1991) Selective stimulation of
glucagon secretion by beta 2-adrenoceptors in isolated islets of Langerhans of the
rat. Br J Pharmacol 103:1824-1828
Langfort J, Ploug T, Ihlemann J, Saldo M, Holm C, Galbo H (1999) Expression of
hormone-sensitive lipase and its regulation by adrenaline in skeletal muscle.
Biochem J 340 ( Pt 2):459-465
Laplante M, Sabatini DM (2012) mTOR Signaling in Growth Control and Disease. Cell
149:274-293
Lembo G, Capaldo B, Rendina V, Iaccarino G, Napoli R, Guida R, Trimarco B, Sacca L
(1994) Acute noradrenergic activation induces insulin resistance in human skeletal
muscle. Am J Physiol 266:E242-7
Liggett SB, Shah SD, Cryer PE (1988) Characterization of beta-adrenergic receptors of
human skeletal muscle obtained by needle biopsy. Am J Physiol 254:E795-8
Lin F, Wang H, Malbon CC (2000) Gravin-mediated formation of signaling complexes
in beta 2-adrenergic receptor desensitization and resensitization. J Biol Chem
275:19025-19034
Lisinski I, Schurmann A, Joost HG, Cushman SW, Al-Hasani H (2001) Targeting of
GLUT6 (formerly GLUT9) and GLUT8 in rat adipose cells. Biochem J 358:517522
Liu H, Enyeart JA, Enyeart JJ (2009) N6-substituted cAMP analogs inhibit bTREK-1
K+ channels and stimulate cortisol secretion by a protein kinase A-independent
mechanism. Mol Pharmacol 76:1290-1301
Liu J, Brautigan DL (2000) Glycogen synthase association with the striated muscle
glycogen-targeting subunit of protein phosphatase-1. Synthase activation involves
scaffolding regulated by beta-adrenergic signaling. J Biol Chem 275:26074-26081
Liu YL, Stock MJ (1995) Acute effects of the beta 3-adrenoceptor agonist, BRL 35135,
on tissue glucose utilisation. Br J Pharmacol 114:888-894
Liu YL, Cawthorne MA, Stock MJ (1996) Biphasic effects of the beta-adrenoceptor
agonist, BRL 37344, on glucose utilization in rat isolated skeletal muscle. Br J
Pharmacol 117:1355-1361
Lohse MJ, Andexinger S, Pitcher J, Trukawinski S, Codina J, Faure JP, Caron MG,
Lefkowitz RJ (1992) Receptor-specific desensitization with purified proteins.
Kinase dependence and receptor specificity of beta-arrestin and arrestin in the beta
2-adrenergic receptor and rhodopsin systems. J Biol Chem 267:8558-8564
56
Luttrell LM, Ferguson SS, Daaka Y, Miller WE, Maudsley S, Della Rocca GJ, Lin F,
Kawakatsu H, Owada K, Luttrell DK, Caron MG, Lefkowitz RJ (1999) Betaarrestin-dependent formation of beta2 adrenergic receptor-Src protein kinase
complexes. Science 283:655-661
Ma TY, Tran D, Hoa N, Nguyen D, Merryfield M, Tarnawski A (2000) Mechanism of
extracellular calcium regulation of intestinal epithelial tight junction permeability:
role of cytoskeletal involvement. Microsc Res Tech 51:156-168
MacGregor DA, Prielipp RC, Butterworth JF,4th, James RL, Royster RL (1996)
Relative efficacy and potency of beta-adrenoceptor agonists for generating cAMP
in human lymphocytes. Chest 109:194-200
Martin WH,3rd, Murphree SS, Saffitz JE (1989) Beta-adrenergic receptor distribution
among muscle fiber types and resistance arterioles of white, red, and intermediate
skeletal muscle. Circ Res 64:1096-1105
Martin WH,3rd, Tolley TK, Saffitz JE (1990) Autoradiographic delineation of skeletal
muscle alpha 1-adrenergic receptor distribution. Am J Physiol 259:H1402-8
McElligott MA, Barreto A,Jr, Chaung LY (1989) Effect of continuous and intermittent
clenbuterol feeding on rat growth rate and muscle. Comp Biochem Physiol C
92:135-138
Minokoshi Y, Okano Y, Shimazu T (1994) Regulatory mechanism of the ventromedial
hypothalamus in enhancing glucose uptake in skeletal muscles. Brain Res
649:343-347
Mitsumoto Y, Burdett E, Grant A, Klip A (1991) Differential expression of the GLUT1
and GLUT4 glucose transporters during differentiation of L6 muscle cells.
Biochem Biophys Res Commun 175:652-659
Mueckler M, Caruso C, Baldwin SA, Panico M, Blench I, Morris HR, Allard WJ,
Lienhard GE, Lodish HF (1985) Sequence and structure of a human glucose
transporter. Science 229:941-945
Murthy KS, Mahavadi S, Huang J, Zhou H, Sriwai W (2008) Phosphorylation of GRK2
by PKA augments GRK2-mediated phosphorylation, internalization, and
desensitization of VPAC2 receptors in smooth muscle. Am J Physiol Cell Physiol
294:C477-87
Nagase I, Yoshida T, Saito M (2001) Up-regulation of uncoupling proteins by betaadrenergic stimulation in L6 myotubes. FEBS Lett 494:175-180
Nakashima M, Vanhoutte PM (1995) Isoproterenol causes hyperpolarization through
opening of ATP-sensitive potassium channels in vascular smooth muscle of the
canine saphenous vein. J Pharmacol Exp Ther 272:379-384
Navegantes LC, Resano NM, Migliorini RH, Kettelhut IC (2001) Catecholamines
inhibit Ca(2+)-dependent proteolysis in rat skeletal muscle through beta(2)adrenoceptors and cAMP. Am J Physiol Endocrinol Metab 281:E449-54
Nevzorova J, Bengtsson T, Evans BA, Summers RJ (2002) Characterization of the betaadrenoceptor subtype involved in mediation of glucose transport in L6 cells. Br J
Pharmacol 137:9-18
Nevzorova J, Evans BA, Bengtsson T, Summers RJ (2006) Multiple signalling
pathways involved in beta2-adrenoceptor-mediated glucose uptake in rat skeletal
muscle cells. Br J Pharmacol 147:446-454
Ngala RA, O'Dowd J, Wang SJ, Agarwal A, Stocker C, Cawthorne MA, Arch JR
(2008) Metabolic responses to BRL37344 and clenbuterol in soleus muscle and
C2C12 cells via different atypical pharmacologies and beta2-adrenoceptor
mechanisms. Br J Pharmacol 155:395-406
57
Ngala RA, O'Dowd J, Wang SJ, Stocker C, Cawthorne MA, Arch JR (2009) Beta2adrenoceptors and non-beta-adrenoceptors mediate effects of BRL37344 and
clenbuterol on glucose uptake in soleus muscle: studies using knockout mice. Br J
Pharmacol 158:1676-1682
Nobles KN, Xiao K, Ahn S, Shukla AK, Lam CM, Rajagopal S, Strachan RT, Huang
TY, Bressler EA, Hara MR, Shenoy SK, Gygi SP, Lefkowitz RJ (2011) Distinct
phosphorylation sites on the beta(2)-adrenergic receptor establish a barcode that
encodes differential functions of beta-arrestin. Sci Signal 4:ra51
Orstavik S, Reinton N, Frengen E, Langeland BT, Jahnsen T, Skalhegg BS (2001)
Identification of novel splice variants of the human catalytic subunit Cbeta of
cAMP-dependent protein kinase. Eur J Biochem 268:5066-5073
Pan SJ, Hancock J, Ding Z, Fogt D, Lee M, Ivy JL (2001) Effects of clenbuterol on
insulin resistance in conscious obese Zucker rats. Am J Physiol Endocrinol Metab
280:E554-61
Penn RB, Parent JL, Pronin AN, Panettieri RA,Jr, Benovic JL (1999) Pharmacological
inhibition of protein kinases in intact cells: antagonism of beta adrenergic receptor
ligand binding by H-89 reveals limitations of usefulness. J Pharmacol Exp Ther
288:428-437
Perkins GA, Wang L, Huang LJ, Humphries K, Yao VJ, Martone M, Deerinck TJ,
Barraclough DM, Violin JD, Smith D, Newton A, Scott JD, Taylor SS, Ellisman
MH (2001) PKA, PKC, and AKAP localization in and around the neuromuscular
junction. BMC Neurosci 2:17
Pitcher JA, Inglese J, Higgins JB, Arriza JL, Casey PJ, Kim C, Benovic JL, Kwatra
MM, Caron MG, Lefkowitz RJ (1992) Role of beta gamma subunits of G proteins
in targeting the beta-adrenergic receptor kinase to membrane-bound receptors.
Science 257:1264-1267
Pittman RN, Molinoff PB (1980) Interactions of agonists and antagonists with betaadrenergic receptors on intact L6 muscle cells. J Cyclic Nucleotide Res 6:421-435
Pittman RN, Molinoff PB (1983) Interactions of full and partial agonists with betaadrenergic receptors on intact L6 muscle cells. Mol Pharmacol 24:398-408
Ploug T, van Deurs B, Ai H, Cushman SW, Ralston E (1998) Analysis of GLUT4
distribution in whole skeletal muscle fibers: identification of distinct storage
compartments that are recruited by insulin and muscle contractions. J Cell Biol
142:1429-1446
Ponsioen B, Gloerich M, Ritsma L, Rehmann H, Bos JL, Jalink K (2009) Direct spatial
control of Epac1 by cyclic AMP. Mol Cell Biol 29:2521-2531
Rall TW, Sutherland EW (1958) Formation of a cyclic adenine ribonucleotide by tissue
particles. J Biol Chem 232:1065-1076
Rattigan S, Appleby GJ, Edwards SJ, McKinstry WJ, Colquhoun EQ, Clark MG,
Richter EA (1986) Alpha-adrenergic receptors in rat skeletal muscle. Biochem
Biophys Res Commun 136:1071-1077
Raz I, Katz A, Spencer MK (1991) Epinephrine inhibits insulin-mediated glycogenesis
but enhances glycolysis in human skeletal muscle. Am J Physiol 260:E430-5
Ricks CA, Dalrymple RH, Baker PK, Ingle D, (1984) Use of a β-adrenergic agonist to
alter muscle deposition in steers. J Anim Sci 59:1247-1255
Roberts SJ, Summers RJ (1998) Cyclic AMP accumulation in rat soleus muscle:
stimulation by beta2- but not beta3-adrenoceptors. Eur J Pharmacol 348:53-60
Rothwell NJ, Stock MJ, Sudera DK (1987) Changes in tissue blood flow and betareceptor density of skeletal muscle in rats treated with the beta2-adrenoceptor
agonist clenbuterol. Br J Pharmacol 90:601-607
58
Sabino-Silva R, Mori RC, David-Silva A, Okamoto MM, Freitas HS, Machado UF
(2010) The Na(+)/glucose cotransporters: from genes to therapy. Braz J Med Biol
Res 43:1019-1026
Samarin SN, Ivanov AI, Flatau G, Parkos CA, Nusrat A (2007) Rho/Rho-associated
kinase-II signaling mediates disassembly of epithelial apical junctions. Mol Biol
Cell 18:3429-3439
Sampath J, Adachi M, Hatse S, Naesens L, Balzarini J, Flatley RM, Matherly LH,
Schuetz JD (2002) Role of MRP4 and MRP5 in biology and chemotherapy. AAPS
PharmSci 4:E14
Sarbassov DD, Ali SM, Kim DH, Guertin DA, Latek RR, Erdjument-Bromage H,
Tempst P, Sabatini DM (2004) Rictor, a novel binding partner of mTOR, defines a
rapamycin-insensitive and raptor-independent pathway that regulates the
cytoskeleton. Curr Biol 14:1296-1302
Sarbassov DD, Guertin DA, Ali SM, Sabatini DM (2005) Phosphorylation and
regulation of Akt/PKB by the rictor-mTOR complex. Science 307:1098-1101
Sarbassov DD, Ali SM, Sengupta S, Sheen JH, Hsu PP, Bagley AF, Markhard AL,
Sabatini DM (2006) Prolonged rapamycin treatment inhibits mTORC2 assembly
and Akt/PKB. Mol Cell 22:159-168
Savard GK, Richter EA, Strange S, Kiens B, Christensen NJ, Saltin B (1989)
Norepinephrine spillover from skeletal muscle during exercise in humans: role of
muscle mass. Am J Physiol 257:H1812-8
Schmid A, Renaud JF, Lazdunski M (1985) Short term and long term effects of betaadrenergic effectors and cyclic AMP on nitrendipine-sensitive voltage-dependent
Ca2+ channels of skeletal muscle. J Biol Chem 260:13041-13046
Schubert D, Tarikas H, LaCorbiere M (1976) Neurotransmitter regulation of adenosine
3',5'-monophosphate in clonal nerve, glia, and muscle cell lines. Science 192:471473
Scott JD (1991) Cyclic nucleotide-dependent protein kinases. Pharmacol Ther 50:123145
Shenoy SK, Drake MT, Nelson CD, Houtz DA, Xiao K, Madabushi S, Reiter E,
Premont RT, Lichtarge O, Lefkowitz RJ (2006) beta-arrestin-dependent, G
protein-independent ERK1/2 activation by the beta2 adrenergic receptor. J Biol
Chem 281:1261-1273
Shih M, Lin F, Scott JD, Wang HY, Malbon CC (1999) Dynamic complexes of beta2adrenergic receptors with protein kinases and phosphatases and the role of gravin.
J Biol Chem 274:1588-1595
Shimazu T, Sudo M, Minokoshi Y, Takahashi A (1991) Role of the hypothalamus in
insulin-independent glucose uptake in peripheral tissues. Brain Res Bull 27:501504
Shukla AK, Violin JD, Whalen EJ, Gesty-Palmer D, Shenoy SK, Lefkowitz RJ (2008)
Distinct conformational changes in beta-arrestin report biased agonism at seventransmembrane receptors. Proc Natl Acad Sci U S A 105:9988-9993
Sillence MN, Matthews ML, Spiers WG, Pegg GG, Lindsay DB (1991) Effects of
clenbuterol, ICI118551 and sotalol on the growth of cardiac and skeletal muscle
and on beta 2-adrenoceptor density in female rats. Naunyn Schmiedebergs Arch
Pharmacol 344:449-453
Sillence MN, Matthews ML (1994) Classical and atypical binding sites for betaadrenoceptor ligands and activation of adenylyl cyclase in bovine skeletal muscle
and adipose tissue membranes. Br J Pharmacol 111:866-872
59
Sloan IG, Sawh PC, Bihler I (1978) Influence of adrenalin on sugar transport in soleus,
a red skeletal muscle. Mol Cell Endocrinol 10:3-12
Soliman GA, Acosta-Jaquez HA, Dunlop EA, Ekim B, Maj NE, Tee AR, Fingar DC
(2010) mTOR Ser-2481 autophosphorylation monitors mTORC-specific catalytic
activity and clarifies rapamycin mechanism of action. J Biol Chem 285:7866-7879
Spurlock DM, McDaneld TG, McIntyre LM (2006) Changes in skeletal muscle gene
expression following clenbuterol administration. BMC Genomics 7:320
Stevens L, Firinga C, Gohlsch B, Bastide B, Mounier Y, Pette D (2000) Effects of
unweighting and clenbuterol on myosin light and heavy chains in fast and slow
muscles of rat. Am J Physiol Cell Physiol 279:C1558-63
Stuart CA, Wen G, Gustafson WC, Thompson EA (2000) Comparison of GLUT1,
GLUT3, and GLUT4 mRNA and the subcellular distribution of their proteins in
normal human muscle. Metabolism 49:1604-1609
Stuart CA, Yin D, Howell ME, Dykes RJ, Laffan JJ, Ferrando AA (2006) Hexose
transporter mRNAs for GLUT4, GLUT5, and GLUT12 predominate in human
muscle. Am J Physiol Endocrinol Metab 291:E1067-73
Stuart CA, Howell ME, Zhang Y, Yin D (2009) Insulin-stimulated translocation of
glucose transporter (GLUT) 12 parallels that of GLUT4 in normal muscle. J Clin
Endocrinol Metab 94:3535-3542
Sudo M, Minokoshi Y, Shimazu T (1991) Ventromedial hypothalamic stimulation
enhances peripheral glucose uptake in anesthetized rats. Am J Physiol 261:E298303
Suko J, Maurer-Fogy I, Plank B, Bertel O, Wyskovsky W, Hohenegger M, Hellmann G
(1993) Phosphorylation of serine 2843 in ryanodine receptor-calcium release
channel of skeletal muscle by cAMP-, cGMP- and CaM-dependent protein kinase.
Biochim Biophys Acta 1175:193-206
Sutherland EW, Rall TW (1958) Fractionation and characterization of a cyclic adenine
ribonucleotide formed by tissue particles. J Biol Chem 232:1077-1091
Suzuki K, Kono T (1980) Evidence that insulin causes translocation of glucose transport
activity to the plasma membrane from an intracellular storage site. Proc Natl Acad
Sci U S A 77:2542-2545
Taha C, Mitsumoto Y, Liu Z, Skolnik EY, Klip A (1995) The insulin-dependent
biosynthesis of GLUT1 and GLUT3 glucose transporters in L6 muscle cells is
mediated by distinct pathways. Roles of p21ras and pp70 S6 kinase. J Biol Chem
270:24678-24681
Tanishita T, Shimizu Y, Minokoshi Y, Shimazu T (1997) The beta3-adrenergic agonist
BRL37344 increases glucose transport into L6 myocytes through a mechanism
different from that of insulin. J Biochem 122:90-95
Vardanega-Peicher M, Lopes G, Lima FB, Curi R, Nakano LC, Bazotte RB (2000)
Time sequence of changes in the responsiveness of glycogen breakdown to
adrenergic agonists in perfused liver of rats with insulin-induced hypoglycemia.
Braz J Med Biol Res 33:805-813
Violin JD, Ren XR, Lefkowitz RJ (2006) G-protein-coupled receptor kinase specificity
for beta-arrestin recruitment to the beta2-adrenergic receptor revealed by
fluorescence resonance energy transfer. J Biol Chem 281:20577-20588
Walsh DA, Perkins JP, Krebs EG (1968) An adenosine 3',5'-monophosphate-dependant
protein kinase from rabbit skeletal muscle. J Biol Chem 243:3763-3765
Wang Y, De Arcangelis V, Gao X, Ramani B, Jung YS, Xiang Y (2008)
Norepinephrine- and epinephrine-induced distinct beta2-adrenoceptor signaling is
60
dictated by GRK2 phosphorylation in cardiomyocytes. J Biol Chem 283:17991807
Watson-Wright WM, Wilkinson M (1986) The muscle slice--a new preparation for the
characterization of beta-adrenergic binding in fast- and slow-twitch skeletal
muscle. Muscle Nerve 9:416-422
Watt MJ, Holmes AG, Pinnamaneni SK, Garnham AP, Steinberg GR, Kemp BE,
Febbraio MA (2006) Regulation of HSL serine phosphorylation in skeletal muscle
and adipose tissue. Am J Physiol Endocrinol Metab 290:E500-8
Wenzel-Seifert K, Seifert R (2000) Molecular analysis of beta(2)-adrenoceptor coupling
to G(s)-, G(i)-, and G(q)-proteins. Mol Pharmacol 58:954-966
Whalen RG, Butler-Browne GS, Sell S, Gros F (1979) Transitions in contractile protein
isozymes during muscle cell differentiation. Biochimie 61:625-632
Williams RS, Caron MG, Daniel K (1984) Skeletal muscle beta-adrenergic receptors:
variations due to fiber type and training. Am J Physiol 246:E160-7
Wilson CM, Cushman SW (1994) Insulin stimulation of glucose transport activity in rat
skeletal muscle: increase in cell surface GLUT4 as assessed by photolabelling.
Biochem J 299 ( Pt 3):755-759
Wisler JW, DeWire SM, Whalen EJ, Violin JD, Drake MT, Ahn S, Shenoy SK,
Lefkowitz RJ (2007) A unique mechanism of beta-blocker action: carvedilol
stimulates beta-arrestin signaling. Proc Natl Acad Sci U S A 104:16657-16662
Wong JA, Gosmanov AR, Schneider EG, Thomason DB (2001) Insulin-independent,
MAPK-dependent stimulation of NKCC activity in skeletal muscle. Am J Physiol
Regul Integr Comp Physiol 281:R561-71
Woo AY, Wang TB, Zeng X, Zhu W, Abernethy DR, Wainer IW, Xiao RP (2009)
Stereochemistry of an agonist determines coupling preference of beta2adrenoceptor to different G proteins in cardiomyocytes. Mol Pharmacol 75:158165
Wu X, Freeze HH (2002) GLUT14, a duplicon of GLUT3, is specifically expressed in
testis as alternative splice forms. Genomics 80:553-557
Yamamoto DL, Hutchinson DS, Bengtsson T (2007) Beta(2)-Adrenergic activation
increases glycogen synthesis in L6 skeletal muscle cells through a signalling
pathway independent of cyclic AMP. Diabetologia 50:158-167
Young DA, Wallberg-Henriksson H, Cranshaw J, Chen M, Holloszy JO (1985) Effect
of catecholamines on glucose uptake and glycogenolysis in rat skeletal muscle.
Am J Physiol 248:C406-9
Zhao FQ, Zheng YC, Wall EH, McFadden TB (2005) Cloning and expression of bovine
sodium/glucose cotransporters. J Dairy Sci 88:182-194
61