Download Glucose induces an autocrine activation of the Wnt/β

Biochem. J. (2008) 416, 211–218 (Printed in Great Britain)
211
doi:10.1042/BJ20081426
Glucose induces an autocrine activation of the Wnt/β-catenin pathway
in macrophage cell lines
Sasha H. ANAGNOSTOU and Peter R. SHEPHERD1
Department of Molecular Medicine and Pathology and Maurice Wilkins Centre for Molecular Biodiscovery, University of Auckland, Auckland 1142, New Zealand
The canonical Wnt signalling pathway acts by slowing the rate
of ubiquitin-mediated β-catenin degradation. This results in the
accumulation and subsequent nuclear translocation of β-catenin,
which induces the expression of a number of genes involved
in growth, differentiation and metabolism. The mechanisms
regulating the Wnt signalling pathway in the physiological
context is still not fully understood. In the present study we
provide evidence that changes in glucose levels within the
physiological range can acutely regulate the levels of β-catenin
in two macrophage cell lines (J774.2 and RAW264.7 cells). In
particular we find that glucose induces these effects by promoting
an autocrine activation of Wnt signalling that is mediated by the
hexosamine pathway and changes in N-linked glycosylation of
proteins. These studies reveal that the Wnt/β-catenin system is
a glucose-responsive signalling system and as such is likely to
play a role in pathways involved in sensing changes in metabolic
status.
INTRODUCTION
The binding of Wnts to cell-surface Frizzled receptors and LRP5/6
(low-density lipoprotein receptor-related protein 5/6) co-receptors
results in a functional change in this complex such that GSK3β no
longer phosphorylates β-catenin. The resultant accumulation of
β-catenin in the cytoplasm leads to nuclear accumulation
and binding to TCF (T-cell-specific transcription factor)/LEF
(lymphoid enhancer-binding factor) transcription factors to
induce the expression of specific target genes [10,11]. Many of
these genes are involved in regulating cell growth and differentiation (see http://www.stanford.edu//∼rnusse/wntwindow.html)
and thus the Wnt signalling pathway plays an important role in
development and cancer [10,12]. It is also becoming clear that the
Wnt/β-catenin pathway is involved in the mechanisms regulating
energy metabolism [13]. Previous studies have provided strong
evidence that the Wnt/β-catenin pathway plays roles in the release
of incretins from the gut [14], β-cell development [15,16] and
glucose-induced insulin secretion [17]. It is also clear that Wnt/βcatenin signalling must be blocked for fat cell differentiation to
occur [18–20]. Furthermore, in mice overexpressing β-catenin
in liver, a number of the genes implicated in switching from
gluconeogeneis to glycolysis are up-regulated [21]. Finally,
compelling evidence that the Wnt/β-catenin pathway is likely
to be involved in processes regulating glucose metabolism
comes from genome-wide SNP (single nucleotide polymorphism)
analysis studies which have identified a strong association
between risk for developing Type 2 diabetes and a polymorphism
in the β-catenin effector gene TCF7L2 [22,23]. Taken together
these findings raise the possibility that factors that regulate
signalling via the Wnt/β-catenin pathway will have an impact
on metabolic pathways.
In the present study we show that physiologically relevant
changes in glucose levels can regulate the canonical Wnt
signalling pathway. This is mediated by the hexosamine pathway
and N-linked glycosylation. This is the first evidence that
One important component of the glucose homoeostatic process
is the ability of key tissues to detect changes in glucose
levels and to respond appropriately. There are a number of
mechanisms described for how cells can sense glucose. These
include ATP-sensitive potassium channels [1], AMP-activated
protein kinase [2,3], activation of PKC (protein kinase C) [4]
and flux through the hexosamine pathway [5]. The hexosamine
pathway uses fructose-6-phosphate to generate UDP-GlcNAc
(UDP-N-acetylglucosamine), which is used for both N-linked
and O-linked glycosylation of proteins [4,5]. Flux through the
hexosamine pathway leads to modification of various intracellular
proteins with O-linked GlcNAc (N-acetylglucosamine) [6]. This
modification mediates a range of functions including protein
stabilization, nuclear localization and transcriptional potential
[6,7]. In addition, it has recently been demonstrated that flux
through the hexosamine pathway can acutely regulate the
dynamics of N-linked glycosylation of receptors and secreted
proteins in a functional manner [8].
We have previously shown that Id2 (inhibitor of DNA binding
2), a downstream target of the Wnt/β-catenin pathway, is regulated
by changes in glucose levels [9], so in the present study we sought
to investigate whether the canonical Wnt signalling pathway could
be regulated by glucose. Key to the transmission of canonical
Wnt signals is the intracellular protein β-catenin. β-catenin is a
transcriptional co-activator, and it also binds to cadherin proteins
to form part of adherens junctions. The membrane-associated,
cadherin-bound pool of β-catenin is highly stable. Cytosolic
β-catenin is usually found in a protein complex with GSK3
(glycogen synthase kinase 3), Axin and APC (adenomatous
polyposis coli). This leads to the phosphorylation of β-catenin by
GSK3, which targets it for rapid ubiquitin-mediated degradation,
thereby maintaining low levels of cytosolic β-catenin [10,11].
Key words: β-catenin, glucose metabolism, hexosamine pathway,
N-linked glycosylation, nutrient sensing.
Abbreviations used: CREB, cAMP-response-element-binding protein; 2DOG, 2-deoxy-D-glucose; Dkk, dickkopf; DMEM, Dulbecco’s modified Eagle’s
medium; DON, 6-diazo-5-oxonorleucine; GFAT, glutamine:fructose-6-phosphate amido transferase; GlcN, glucosamine; GSK3, glycogen synthase kinase
3; Id2, inhibitor of DNA binding 2; LEF, lymphoid enhancer-binding factor; LRP5/6, low-density lipoprotein receptor-related protein 5/6; O -GlcNAc, O-linked
N -acetylglucosamine; PUGNAc, O -(2-acetamido-2-deoxy-D-glucopyranosylidene) amino N -phenylcarbamate; sFRP2, secreted frizzled-related protein 2;
TCF, T-cell-specific transcription factor; UDP-GlcNAc, UDP-N -acetylglucosamine.
1
To whom correspondence should be addressed (email [email protected]).
c The Authors Journal compilation c 2008 Biochemical Society
212
S. H. Anagnostou and P. R. Shepherd
the canonical Wnt signalling pathway is glucose-responsive,
indicating that this is a previously undescribed mechanism for
sensing and responding to changes in glucose concentrations.
TBP (TATA box-binding protein) and HPRT1 (hypoxanthine
phosphoribosyltransferase), which were internally consistent in
all samples analysed. Primer sequences used in the present study
are shown below.
EXPERIMENTAL
Unless otherwise stated, reagents were purchased from Sigma–
Aldrich. The β-catenin antibody was from Symansis. Antibodies
directed at phospho-β-catenin (Ser33 /Ser37 / Thr41 ), GSK3β,
phospho-GSK3β (Ser9 ), phospho-LRP6 (Ser1490 ) and CREB
(cAMP-response-element-binding protein) were from Cell Signaling Technologies. An O-GlcNAc detection kit was from Pierce.
TOPFLASH and FOPFLASH constructs were kindly provided
by Dr Andrew Grey (Department of Medicine, University of
Auckland, New Zealand). PUGNAc [O-(2-acetamido-2-deoxy-Dglucopyranosylidene) amino N-phenylcarbamate] was purchased
from Toronto Research Chemicals. Recombinant mouse Dkk
(dickkopf) and sFRP2 (secreted frizzled-related protein 2) were
from R&D Systems.
Cell culture and transfections
RAW264.7 and J774.2 cells were maintained in RPMI 1640
medium (Gibco) supplemented with 10 % (v/v) newborn calf
serum, 100 units/ml penicillin and 100 μg/ml streptomycin
(Gibco). HepG2 cells were maintained in DMEM (Dulbecco’s
modified Eagle’s medium; Gibco) supplemented with 10 %
(v/v) newborn calf serum, 100 units/ml penicillin and 100 μg/ml
streptomycin (Gibco). L (parental) and L-Wnt3a cells were
maintained in high-glucose DMEM (Gibco) supplemented
with 10 % (v/v) newborn calf serum, 100 units/ml penicillin
and 100 μg/ml streptomycin (Gibco). Conditioned media was
prepared according to A.T.C.C. guidelines.
Experiments were performed on confluent cultures after
overnight serum-starvation in RPMI 1640 with no glucose, and
supplemented with glucose as indicated. Transient transfections
of RAW264.7 cells with TOPFLASH/FOPFLASH constructs
were performed with FuGENETM HD transfection reagent
(Roche). At 24 h after transfection, cells were serum-starved
overnight in 0.5 mM glucose. A luciferase assay was performed
using the Dual-Glo luciferase assay system (Promega). pRLTK (Renilla luciferase-thymidine kinase) was co-transfected for
normalization.
Protein isolation, subcellular fractionation and immunoblotting
Cell lysates were prepared as previously described [24].
Subcellular fractionation was performed using the NE-PER
kit (Pierce). Protein concentrations were determined by BCA
(bicinchoninic acid) protein assay (Pierce). Immunoblotting and
visualization of immunoreactive proteins have been described
previously [24].
RNA isolation, cDNA synthesis and quantitative real-time PCR
RNA was isolated with TRIzol® (Invitrogen). cDNA was
synthesized from 2.5 μg of total RNA using Thermoscript
(Invitrogen). Real-time PCR was performed using the Applied
Biosystems 7900 instrument and software (Applied Biosystems).
RT (reverse transcriptase) and no-template controls were included
in all experiments. Primers were designed using the Primer
Express Software (Applied Biosystems) and verified by melt
profile analysis. SYBR green mastermix with ROX (Invitrogen)
was used for PCR reactions. Gene expression was analysed
by normalization to housekeeping genes ACTB (β-actin),
c The Authors Journal compilation c 2008 Biochemical Society
Species gene primer sequence
The primers used in the present study were: mouse Actb forward, 5 -TGTCCACCTTCCAGCAGATGT-3 and reverse, 5 AGCTCAGTAACAGTCCGCCTAGA-3 ; mouse Tbp forward,
5 -ATGAGCCAGAATTATTTCCTGGATTA-3 and reverse, 5 GATGTTTTCAAATGCTTCATAAATCTCT-3 ; mouse Hprt1
forward, 5 -CCCAAAATGGTTAAGGTTGCAA-3 and reverse, 5 -CTCATTATAGTCAAGGGCATATCCAA-3 ; mouse βcatenin (ctnnb1) forward, 5 -TCTTCAGGACAGAGCCAATGG
-3 and reverse, 5 -ACCAGAGTGAAAAGAACGGTAGCT-3 ;
mouse CyclinD1 forward, 5 -GATGTGAGGGAAGAGGTGAAGGT-3 and reverse, 5 -CAATGAGAATCTGGTTCTGAACGT-3 ; and mouse Axin2 forward, 5 -TCACAGCCCTTGTGGTTCAAG-3 and reverse, 5 -GGTAGATTCCTGATGGCCGTAGT-3 .
Statistical analysis
Results are presented as means +
− S.E.M. with the number of
experiments indicated in the legend. Statistical significance was
assessed using one-way ANOVA and Bonferroni’s post-hoc test.
RESULTS AND DISCUSSION
Glucose regulates β-catenin levels in cell lines
Most cells have high levels of β-catenin due to the stable pool of
β-catenin bound to cadherins which can mask any changes to the
cytosolic and nuclear pools of β-catenin. Therefore, in the present
study, we used the J774.2 and RAW264.7 macrophage cell lines as
we have previously shown that these cells are glucose responsive
[9,25] and have very low levels of classical cadherins (Figure 1A).
Thus the β-catenin seen in these cell lines represents the cytosolic
and nuclear pools of β-catenin that are involved in transducing
Wnt signals and TCF-dependent transcriptional activation. After
starvation in low glucose, restoring glucose levels led to a rapid,
dose-dependent increase in β-catenin levels in both cell lines
(Figures 1B and 1C). Importantly, changes in β-catenin levels
were observed between 5 mM and 20 mM glucose, indicating
that the effect can occur within the physiological range of glucose
concentrations.
Glucose effect on β-catenin requires the hexosamine pathway
Several mechanisms have been described that allow cells to sense
glucose concentrations as described above. We have previously
shown that the β-catenin target gene Id2 was increased by glucose
and that this effect was dependent on the hexosamine pathway
[9]. In the present study we provide evidence that the hexosamine
pathway is involved in regulating β-catenin levels. The evidence
supporting this is: (i) the effect was blocked by azaserine and
DON (6-diazo-5-oxonorleucine) (Figure 2A) which are inhibitors
of GFAT (glutamine:fructose-6-phosphate amido transferase), the
rate-limiting enzyme in the hexosamine biosynthetic pathway
[26]; (ii) GFAT requires glutamine as an amino donor [26] and the
glucose effect was lost when glutamine is not present (Figure 2B);
and (iii) the effects of glucose can be replicated by low levels of
GlcN (glucosamine), which can directly enter the hexosamine
Glucose regulation of Wnt signalling
Figure 1
213
Effect of glucose on β-catenin protein levels
(A) Western blotting was used to assess relative levels of cadherins in HepG2, RAW264.7
and J774.2 cells under growing culture conditions. β-actin served as a loading control.
(B and C) Representative Western blots and densitometry quantifications of β-catenin from
three independent experiments performed in triplicate expressed relative to the untreated group
(means +
− S.E.M.; *P < 0.05). β-actin served as a loading control. RAW264.7 cells (B) and
J774.2 cells (C) were serum-starved overnight in 0.5 mM glucose and then treated for 2 h with
increasing concentrations of glucose.
pathway downstream of GFAT (Figure 2C). Similar results were
observed in J774.2 cells (results not shown).
N-linked glycosylation mediates the glucose effect on β-catenin
The hexosamine pathway produces UDP-GlcNAc, which can
be used as a donor for N-linked or O-linked glycosylation of
proteins [5]. To determine which of these types of glycosylation
was responsible for the effect of glucose on β-catenin, we
first used PUGNAc, an inhibitor of the enzyme that mediates
removal of O-GlcNAc from protein. PUGNAc did not mimic
the glucose effect (Figure 3A); however, in the same samples,
PUGNAc did increase the overall level of O-GlcNAc (Figure 3B),
Figure 2 Involvement of the hexosamine pathway in the effects of glucose
on β-catenin
(A–C) RAW264.7 cells were serum-starved in 0.5 mM glucose overnight, and treated for
2 h. Results are presented as a representative β-catenin Western blot and densitometry
quantifications of three independent experiments performed in triplicate expressed relative
to untreated groups (means +
− S.E.M.; *P < 0.05). β-actin served as a loading control.
(A) GFAT inhibitors azaserine (Aza) (10 μM) or DON (10 μM) blocked the effect of 20 mM
glucose on β-catenin. (B) Glucose (20 mM) does not increase β-catenin levels in the absence
of glutamine. (C) GlcN treatment mimics the effect of glucose.
demonstrating that PUGNAc is in fact inhibiting O-GlcNAc
removal. We then investigated the role of N-linked glycosylation
with the inhibitor tunicamycin. We found that inhibiting N-linked
glycosylation with tunicamycin completely blocked the effect of
c The Authors Journal compilation c 2008 Biochemical Society
214
Figure 3
S. H. Anagnostou and P. R. Shepherd
Involvement of N-linked glycosylation in the glucose effect on β-catenin
(A–E) RAW264.7 cells were serum-starved overnight in 0.5 mM glucose and treated for 2 h as indicated. Representative Western blots and densitometry quantifications of three independent
experiments performed in triplicate are shown. Values are means +
− S.E.M. (*P < 0.05 compared with untreated cells). β-actin served as a loading control. (A) PUGNAc (40 μM) did not mimic the
effect of low glucose, and had no effect with high glucose (20 mM). (B) PUGNAc (40 μM) does increase non-specific O -GlcNAc levels in the samples from (A). (C) Tunicamycin (10 μg/ml; Tun)
blocks the glucose induction of β-catenin. (D) Increasing concentrations of 2DOG decreased β-catenin levels in low glucose and competed out the effect of 5 mM glucose (E).
glucose on β-catenin (Figure 3C). 2DOG (2-deoxy-D-glucose)
has also been shown to inhibit N-linked glycosylation via a
different mechanism than that of tunicamycin [27,28]. In the
present study we show that 2DOG treatment decreases β-catenin
levels both in the presence and absence of glucose (Figures 3D and
3E). This provides further evidence that N-linked glycosylation
is required for the glucose effect on β-catenin. Similar results
were observed in J774.2 cells (results not shown).
Increased β-catenin protein levels with glucose treatment
is the result of decreased degradation
To investigate the mechanism by which β-catenin protein levels
were increased, we first examined the β-catenin transcript
level. Neither glucose nor glucosamine treatments, or the absence
c The Authors Journal compilation c 2008 Biochemical Society
of glutamine, affected β-catenin mRNA levels (Figure 4A).
Phosphorylation of β-catenin by GSK3 is an essential step in
targeting it for proteasomal degradation, so lower levels of GSK3
activity might explain increases in β-catenin [10,11]. However,
we found that the total levels of GSK3β did not change with
glucose or glucosamine treatment and were not affected by the
absence of glutamine (Figure 4B). GSK3 is also inactivated by
phosphorylation so the ratio of phosphorylated to total β-catenin
is a reflection of the degree of GSK3 activity in the destruction
complex. This was also unaffected by glucose or glucosamine
treatment, or by the absence of glutamine (Figure 4B).
We do find, however, that the ratio of phosphorylated to
total β-catenin is decreased following glucose or glucosamine
treatment (Figure 4C), indicating that less GSK3β-mediated
Glucose regulation of Wnt signalling
Figure 4
Regulation of GSK3 signalling by glucose
(A–C) RAW264.7 cells were serum-starved overnight in 0.5 mM glucose and treated for 2 h
as indicated. (A) β-catenin transcript does not change with glucose treatment. After treatment
with 20 mM glucose, 20 mM glucose and no glutamine, or 0.2 mM GlcN, quantitative real-time
PCR was performed to assess β-catenin transcript levels. Results are presented relative to the
untreated group and values are means +
− S.E.M. from two experiments performed in triplicate.
*P < 0.05 compared with the untreated samples. (B) GSK3β levels are unchanged with glucose
treatment (in the presence or absence of glutamine) or GlcN treatment. Phospho-GSK3β and total
GSK3β Western blots were analysed under these conditions and densitometry quantifications
from three independent experiments performed in triplicate are expressed relative to untreated
groups (means +
− S.E.M). (C) Glucose and GlcN decrease the proportion of β-catenin that is
phosphorylated. Phospho-β-catenin and total β-catenin Western blots from three independent
experiments performed in triplicate were analysed, and the ratio of phospho-to-total β-catenin
from three independent experiments is presented relative to the untreated group. Values are
means +
− S.E.M. (*P < 0.05).
phosphorylation of β-catenin is occurring, thus stabilizing βcatenin protein. As we found no evidence for GSK3 activity
changing, the most likely explanation was that the glucose
was affecting Wnt signalling as this leads to a conformational
change in the destruction complex such that less GSK3-mediated
phosphorylation of β-catenin occurs without any direct effect on
total levels of GSK3 activity.
Glucose induces autocrine activation of the canonical Wnt
signalling pathway
As there is less phosphorylation of β-catenin occurring without
any effect on GSK3, we investigated whether glucose was
increasing canonical Wnt signalling. Previous studies suggest
mechanisms by which this could occur. Glucose could be
stimulating the secretion of Wnts, as it is known that the
secretion of some Wnts requires N-linked glycosylation [29,30]
215
and autocrine regulation of Wnt signalling has been reported
[31,32]. Another hypothesis is that glucose could be regulating
N-linked glycosylation of the Frizzled proteins or their coreceptors (LRP5/6). This could be altering Wnt signalling as it is
known that some receptors are functionally regulated by N-linked
glycosylation [8]. For example, there is evidence that LRP6 is
N-glycosylated and this affects the localization and signalling
capacity of this receptor [33]. In each of these hypotheses,
extracellular Wnt signalling would be required for the glucose
effect on β-catenin. To test this we used recombinant Dkk1, which
attenuates Wnt signalling by blocking the interaction between
Wnts and the Frizzled–LRP complex [34]. We find that Dkk1
completely abrogates the effects of glucose on β-catenin levels
(Figure 5A). The extracellular Wnt-binding protein sFRP2 also
blocks Wnt signalling [35], and we find that it also blocks the
effect of glucose on β-catenin (Figure 5B). Taken together these
results provide evidence that the Wnt signalling pathway could
be involved in the glucose effect on β-catenin.
To more directly test whether the canonical Wnt pathway was
being activated by changes in glucose levels, we investigated
the phosphorylation status of LRP6, as it has previously been
shown that Wnt signalling at the receptor level leads to the
phosphorylation of LRP6 [36,37]. As a control experiment we
found that treatment of RAW264.7 cells with Wnt3a-conditioned
medium induced the phosphorylation of LRP6 (Figure 5C). We
went on to show that glucose induced phosphorylation of LRP6
and that this was mediated by the hexosamine pathway as the
glucose effect required the presence of glutamine and could
be mimicked by glucosamine (Figure 5C). Finally, glucoseinduced phosphorylation of LRP6 was blocked by tunicamycin
(Figure 5D), indicating that N-linked glycosylation is required
for glucose-induced Wnt signalling. Taken together these results
provide strong evidence that glucose-induced increases in βcatenin arise due to an autocrine activation of the Wnt signalling
system, and that this involves the hexosamine pathway and Nglycosylation of proteins.
Glucose-induced changes in β-catenin leads to TCF-dependent
transcriptional events
An important question was whether the increased β-catenin
with glucose treatment has functional consequences for gene
expression. To assess whether glucose-induced β-catenin was
mediating the transcriptional events that would be expected upon
Wnt stimulation, we first performed subcellular fractionation
experiments and showed that glucose or glucosamine treatment
led to an increase in β-catenin in the nucleus of RAW264.7
cells (Figure 6A) and J774.2 cells (results not shown). Next we
demonstrated that glucose increased the transcriptional activity of
a β-catenin/TCF reporter system in RAW264.7 cells (Figure 6B).
Axin2 and CyclinD1 are well-recognized β-catenin target genes.
We were unable to detect Axin2 mRNA in RAW264.7 cells
(results not shown), but we showed that levels of CyclinD1
mRNA increased following exposure of RAW264.7 cells to
glucose (Figure 6C). This glucose effect on CyclinD1 required
glutamine and was mimicked by glucosamine (Figure 6C), and
was blocked by tunicamycin (Figure 6D). In J774.2 cells, where
both Axin2 and CyclinD1 transcripts were readily detectable,
we found that glucose induced the expression of both CyclinD1
and Axin2 and this was blocked by tunicamycin (Figure 6E).
Taken together these results clearly demonstrate that glucose,
via the hexosamine pathway and N-linked glycosylation,
increases the level of functional β-catenin, which accumulates
in the nucleus and binds to members of the TCF/LEF family to
activate transcription of target genes.
c The Authors Journal compilation c 2008 Biochemical Society
216
Figure 5
S. H. Anagnostou and P. R. Shepherd
Glucose regulation of canonical Wnt signalling
RAW264.7 cells were serum-starved in 0.5 mM glucose overnight, and treated for 2 h with 20 mM glucose plus the indicated treatment. Representative Western blots are shown with densitometry
quantifications of three independent experiments performed in triplicate expressed relative to untreated groups (means +
− S.E.M.; *P < 0.05 compared with untreated). β-actin or non-specific protein
stain served as a loading control. (A and B) Wnt inhibitory proteins Dkk1 (100 ng/ml) and sFRP2 (250 ng/ml) block the glucose effect on β-catenin. (C) Phospho-LRP6 levels in cells treated with
20 mM glucose, 0.2 mM GlcN, 20 mM glucose with no glutamine (Glut), or with control or Wnt3a conditioned medium (CM). (D) Phospho-LRP6 levels in RAW264.7 cells treated with 20 mM
glucose and 10 μg/ml tunicamycin (Tun) as indicated.
Implications for glucose regulation of the canonical Wnt
signalling pathway
The present study provides the first evidence that physiologically
relevant changes in glucose concentrations regulate β-catenin
protein levels. Macrophage cell lines were used as they have
low levels of cadherins and we had previously shown that Id2
levels were increased with glucose treatment in these cells. Use
of cell culture models in the present study allowed dissection of
the signalling mechanisms involved in the glucose effect. We find
that the glucose effect is mediated via the hexosamine pathway
and involves regulation of the canonical Wnt signalling system;
however, we also find that β-catenin levels rapidly increase 2–3fold in liver, muscle and fat following refeeding of fasted rats and
levels are also elevated in liver of insulinopaenic, hyperglygaemic
streptozotocin-induced diabetic rats (S. H. Anagnostou and P. R.
Shepherd, unpublished work). This indicates both that changes in
β-catenin occur in tissues involved in regulating glucose metabolism under physiologically relevant conditions and that these
effects are also most likely due to changes in glucose levels in vivo.
The finding that glucose induces changes in β-catenin levels
has potential implications for understanding how nutrients can
c The Authors Journal compilation c 2008 Biochemical Society
regulate a range of biological processes. These mechanisms
allow fine-tuning of metabolic responses to react rapidly to
changes in glucose concentrations. Although a number of such
mechanisms have already been described (see Introduction),
several lines of evidence suggest that a mechanism involving the
hexosamine pathway, regulation of Wnt signalling and changes
in β-catenin levels could make an important contribution to
processes regulating glucose homoeostasis. First, the hexosamine
pathway is known to play a role in the processes regulating glucose
metabolism in muscle, liver, fat and in β-cells [5,38,39], and in
the glucose-dependent expression of leptin [40]. Secondly, there
is evidence that Wnt signalling plays a crucial role in regulating
the differentiation of β-cells and adipocytes, both of which are
key tissues involved in regulating glucose homoeostasis, so any
glucose-induced changes in β-catenin levels could potentially
affect these processes [15,18–20,41,42]. A third line of evidence
comes from the finding that, in mice overexpressing β-catenin
in liver, a number of the genes implicated in switching from
gluconeogeneis to glycolysis are up-regulated, in particular
pyruvate kinase [21]. Changes in glucose levels are known to play
a role in regulating the gluconeogenic/glycolytic switch, but the
full details of the mechanisms involved are not fully understood.
Glucose regulation of Wnt signalling
Figure 6
217
Effect of glucose on TCF-dependent transcriptional events
(A–E) Cells were serum-starved overnight in 0.5 mM glucose and treated for 2 h (A and C–E) or 6 h (B). (A) Cells were treated with no glucose (lanes 1, 4 and 7), 20 mM glucose (lanes 2, 5
and 8) or 0.2 mM GlcN (lanes 3, 6 and 9). Nuclear and cytoplasmic fractions were analysed by Western blot. CREB (nuclear) and α-tubulin (cytoplasmic) served as fractionation controls. Western
blots are representative of three independent experiments performed in duplicate. (B) RAW264.7 cells were transfected with the TCF reporter construct TOPFLASH or FOPFLASH control and treated
with or without 20 mM glucose. Data represent firefly/Renilla signal from three independent experiments performed in triplicate, expressed as the fold-induction relative to no glucose samples.
(C–E) Quantitative real-time PCR was performed and transcript levels are expressed as the fold-change relative to untreated samples (means +
− S.E.M. from at least three independent experiments in
triplicate; *P < 0.05). (C) CyclinD1 transcript levels in RAW264.7 cells treated with glucose (20 mM), glucose and no glutamine (Glut) or 0.2 mM GlcN. (D) CyclinD1 transcript levels in RAW264.7
cells treated with glucose (20 mM) and 10 μg/ml tunicamycin (Tun) as indicated. (E) CyclinD1 and Axin2 transcript levels in J774.2 cells treated with 20 mM glucose and 10 μg/ml tunicamycin
(Tun).
We propose that changes in β-catenin levels might play a role
in this process. Finally there is now strong genetic evidence
that TCF7L2 plays an important role in regulating the processes
involved in glucose homoeostasis, and as TCF7L2 is an effector
of β-catenin this also indicates that changes in β-catenin levels
are likely to have an impact on glucose homoeostasis [22,23].
The findings of the present study also have potential
implications for understanding cancer biology. The Wnt/βcatenin pathway plays a key role in the development of many
tumour types [10–12] and it is known that tumours have
greatly increased levels of glucose uptake and flux through the
hexosamine pathway [26,43,44]. In addition, N-glycosylation
is involved in both tumorigenesis and metastasis [44–46]. This
suggests that the glucose-dependent mechanism we describe
could be promoting tumour growth by contributing to Wnt
signalling, conceivably via the previously described autocrine
Wnt signalling loop found in some cancer cells [31,32].
Together the findings of the present study provide evidence for
a new regulatory mechanism by which certain cells can sense
glucose levels and respond appropriately at the cellular level.
Further work is required to understand whether this mechanism
for regulating Wnt signalling exists in all cells, or whether it
is restricted to cells involved in metabolic processes. Further
analysis of this mechanism in cancer cells may lead to new
approaches for targeting Wnt signalling in cancer.
ACKNOWLEDGEMENTS
We thank Dr Anassuya Ramachandran for technical assistance.
FUNDING
This work was supported by the Health Research Council of New Zealand [grant numbers
05/257, 07/080A].
REFERENCES
1 MacDonald, P. E., Joseph, J. W. and Rorsman, P. (2005) Glucose-sensing mechanisms in
pancreatic β-cells. Philos. Trans. R. Soc. Lond. B Biol. Sci. 360, 2211–2225
2 Carling, D. (2004) The AMP-activated protein kinase cascade: a unifying system for
energy control. Trends Biochem. Sci. 29, 18–24
c The Authors Journal compilation c 2008 Biochemical Society
218
S. H. Anagnostou and P. R. Shepherd
3 Winder, W. W. and Thomson, D. M. (2007) Cellular energy sensing and signaling by
AMP-activated protein kinase. Cell Biochem. Biophys. 47, 332–347
4 Brownlee, M. (2001) Biochemistry and molecular cell biology of diabetic complications.
Nature 414, 813–820
5 Marshall, S. (2006) Role of insulin, adipocyte hormones, and nutrient-sensing pathways
in regulating fuel metabolism and energy homeostasis: a nutritional perspective of
diabetes, obesity, and cancer. Science STKE 2006, RE7
6 Zachara, N. E. and Hart, G. W. (2006) Cell signaling, the essential role of O-GlcNAc!
Biochim. Biophys. Acta 1761, 599–617
7 Vosseller, K., Sakabe, K., Wells, L. and Hart, G. W. (2002) Diverse regulation of protein
function by O-GlcNAc: a nuclear and cytoplasmic carbohydrate post-translational
modification. Curr. Opin. Chem. Biol. 6, 851–857
8 Lau, K. S., Partridge, E. A., Grigorian, A., Silvescu, C. I., Reinhold, V. N., Demetriou, M.
and Dennis, J. W. (2007) Complex N-glycan number and degree of branching cooperate
to regulate cell proliferation and differentiation. Cell 129, 123–134
9 Gronning, L. M., Tingsabadh, R., Hardy, K., Dalen, K. T., Jat, P. S., Gnudi, L. and
Shepherd, P. R. (2006) Glucose induces increases in levels of the transcriptional repressor
Id2 via the hexosamine pathway. Am. J. Physiol. Endocrinol. Metab. 290, E599–E606
10 Clevers, H. (2006) Wnt/β-catenin signaling in development and disease. Cell 127,
469–480
11 Gordon, M. D. and Nusse, R. (2006) Wnt signaling: multiple pathways, multiple
receptors, and multiple transcription factors. J. Biol. Chem. 281, 22429–22433
12 Nusse, R. (2005) Wnt signaling in disease and in development. Cell Res. 15, 28–32
13 Prestwich, T. C. and MacDougald, O. A. (2007) Wnt/β-catenin signaling in adipogenesis
and metabolism. Curr. Opin. Cell Biol. 19, 612–617
14 Ni, Z., Anini, Y., Fang, X., Mills, G., Brubaker, P. L. and Jin, T. (2003) Transcriptional
activation of the proglucagon gene by lithium and β-catenin in intestinal endocrine L
cells. J. Biol. Chem. 278, 1380–1387
15 McLin, V. A., Rankin, S. A. and Zorn, A. M. (2007) Repression of Wnt/β-catenin signaling
in the anterior endoderm is essential for liver and pancreas development. Development
134, 2207–2217
16 Rulifson, I. C., Karnik, S. K., Heiser, P. W., ten Berge, D., Chen, H., Gu, X., Taketo, M. M.,
Nusse, R., Hebrok, M. and Kim, S. K. (2007) Wnt signaling regulates pancreatic β cell
proliferation. Proc. Natl. Acad. Sci. U.S.A. 104, 6247–6252
17 Fujino, T., Asaba, H., Kang, M. J., Ikeda, Y., Sone, H., Takada, S., Kim, D. H., Ioka, R. X.,
Ono, M., Tomoyori, H. et al. (2003) Low-density lipoprotein receptor-related protein 5
(LRP5) is essential for normal cholesterol metabolism and glucose-induced insulin
secretion. Proc. Natl. Acad. Sci. U.S.A. 100, 229–234
18 Christodoulides, C., Laudes, M., Cawthorn, W. P., Schinner, S., Soos, M., O’Rahilly, S.,
Sethi, J. K. and Vidal-Puig, A. (2006) The Wnt antagonist Dickkopf-1 and its receptors
are coordinately regulated during early human adipogenesis. J. Cell Sci. 119,
2613–2620
19 Ross, S. E., Hemati, N., Longo, K. A., Bennett, C. N., Lucas, P. C., Erickson, R. L. and
MacDougald, O. A. (2000) Inhibition of adipogenesis by Wnt signaling. Science 289,
950–953
20 Wright, W. S., Longo, K. A., Dolinsky, V. W., Gerin, I., Kang, S., Bennett, C. N., Chiang,
S. H., Prestwich, T. C., Gress, C., Burant, C. F. et al. (2007) Wnt10b inhibits obesity in
ob /ob and agouti mice. Diabetes 56, 295–303
21 Tan, X., Apte, U., Micsenyi, A., Kotsagrelos, E., Luo, J. H., Ranganathan, S., Monga, D. K.,
Bell, A., Michalopoulos, G. K. and Monga, S. P. (2005) Epidermal growth factor receptor:
a novel target of the Wnt/β-catenin pathway in liver. Gastroenterology 129, 285–302
22 Grant, S. F., Thorleifsson, G., Reynisdottir, I., Benediktsson, R., Manolescu, A., Sainz, J.,
Helgason, A., Stefansson, H., Emilsson, V., Helgadottir, A. et al. (2006) Variant of
transcription factor 7-like 2 (TCF7L2) gene confers risk of type 2 diabetes. Nat. Genet. 38,
320–323
23 Cauchi, S., El Achhab, Y., Choquet, H., Dina, C., Krempler, F., Weitgasser, R., Nejjari, C.,
Patsch, W., Chikri, M., Meyre, D. and Froguel, P. (2007) TCF7L2 is reproducibly
associated with type 2 diabetes in various ethnic groups: a global meta-analysis.
J. Mol. Med. 85, 777–782
24 Chaussade, C., Rewcastle, G. W., Kendall, J. D., Denny, W. A., Cho, K., Gronning, L. M.,
Chong, M. L., Anagnostou, S. H., Jackson, S. P., Daniele, N. and Shepherd, P. R. (2007)
Evidence for functional redundancy of class IA PI3K isoforms in insulin signalling.
Biochem. J. 404, 449–458
Received 15 July 2008/24 September 2008; accepted 30 September 2008
Published as BJ Immediate Publication 30 September 2008, doi:10.1042/BJ20081426
c The Authors Journal compilation c 2008 Biochemical Society
25 O’Rourke, L., Gronning, L. M., Yeaman, S. J. and Shepherd, P. R. (2002)
Glucose-dependent regulation of cholesterol ester metabolism in macrophages by insulin
and leptin. J. Biol. Chem. 277, 42557–42562
26 Milewski, S. (2002) Glucosamine-6-phosphate synthase: the multi-facets enzyme.
Biochim. Biophys. Acta 1597, 173–192
27 Datema, R. and Schwarz, R. T. (1979) Interference with glycosylation of glycoproteins.
Inhibition of formation of lipid-linked oligosaccharides in vivo . Biochem. J. 184,
113–123
28 Kurtoglu, M., Gao, N., Shang, J., Maher, J. C., Lehrman, M. A., Wangpaichitr, M.,
Savaraj, N., Lane, A. N. and Lampidis, T. J. (2007) Under normoxia, 2-deoxy-D-glucose
elicits cell death in select tumor types not by inhibition of glycolysis but by interfering
with N-linked glycosylation. Mol. Cancer Ther. 6, 3049–3058
29 Komekado, H., Yamamoto, H., Chiba, T. and Kikuchi, A. (2007) Glycosylation and
palmitoylation of Wnt-3a are coupled to produce an active form of Wnt-3a. Genes Cells
12, 521–534
30 Kurayoshi, M., Yamamoto, H., Izumi, S. and Kikuchi, A. (2007) Post-translational
palmitoylation and glycosylation of Wnt-5a are necessary for its signalling. Biochem. J.
402, 515–523
31 Bafico, A., Liu, G., Goldin, L., Harris, V. and Aaronson, S. A. (2004) An autocrine
mechanism for constitutive Wnt pathway activation in human cancer cells. Cancer Cell 6,
497–506
32 Schlange, T., Matsuda, Y., Lienhard, S., Huber, A. and Hynes, N. E. (2007) Autocrine WNT
signaling contributes to breast cancer cell proliferation via the canonical WNT pathway
and EGFR transactivation. Breast Cancer Res. 9, R63
33 Khan, Z., Vijayakumar, S., de la Torre, T. V., Rotolo, S. and Bafico, A. (2007) Analysis of
endogenous LRP6 function reveals a novel feedback mechanism by which Wnt negatively
regulates its receptor. Mol. Cell. Biol. 27, 7291–7301
34 Niehrs, C. (2006) Function and biological roles of the Dickkopf family of Wnt modulators.
Oncogene 25, 7469–7481
35 Kawano, Y. and Kypta, R. (2003) Secreted antagonists of the Wnt signalling pathway.
J. Cell Sci. 116, 2627–2634
36 Zeng, X., Tamai, K., Doble, B., Li, S., Huang, H., Habas, R., Okamura, H., Woodgett, J. and
He, X. (2005) A dual-kinase mechanism for Wnt co-receptor phosphorylation and
activation. Nature 438, 873–877
37 Davidson, G., Wu, W., Shen, J., Bilic, J., Fenger, U., Stannek, P., Glinka, A. and Niehrs, C.
(2005) Casein kinase 1γ couples Wnt receptor activation to cytoplasmic signal
transduction. Nature 438, 867–872
38 McClain, D. A., Hazel, M., Parker, G. and Cooksey, R. C. (2005) Adipocytes with increased
hexosamine flux exhibit insulin resistance, increased glucose uptake, and increased
synthesis and storage of lipid. Am. J. Physiol. Endocrinol. Metab. 288, E973–E979
39 Veerababu, G., Tang, J., Hoffman, R. T., Daniels, M. C., Hebert, Jr, L. F., Crook, E. D.,
Cooksey, R. C. and McClain, D. A. (2000) Overexpression of
glutamine:fructose-6-phosphate amidotransferase in the liver of transgenic mice results
in enhanced glycogen storage, hyperlipidemia, obesity, and impaired glucose tolerance.
Diabetes 49, 2070–2078
40 Wang, J., Liu, R., Hawkins, M., Barzilai, N. and Rossetti, L. (1998) A nutrient-sensing
pathway regulates leptin gene expression in muscle and fat. Nature 393, 684–688
41 Heiser, P. W., Lau, J., Taketo, M. M., Herrera, P. L. and Hebrok, M. (2006) Stabilization of
β-catenin impacts pancreas growth. Development 133, 2023–2032
42 Longo, K. A., Wright, W. S., Kang, S., Gerin, I., Chiang, S. H., Lucas, P. C., Opp, M. R. and
MacDougald, O. A. (2004) Wnt10b inhibits development of white and brown adipose
tissues. J. Biol. Chem. 279, 35503–35509
43 Gatenby, R. A. and Gillies, R. J. (2004) Why do cancers have high aerobic glycolysis?
Nat. Rev. Cancer 4, 891–899
44 Bironaite, D., Nesland, J. M., Dalen, H., Risberg, B. and Bryne, M. (2000) N-glycans
influence the in vitro adhesive and invasive behaviour of three metastatic cell lines.
Tumour Biol. 21, 165–175
45 Granovsky, M., Fata, J., Pawling, J., Muller, W. J., Khokha, R. and Dennis, J. W. (2000)
Suppression of tumor growth and metastasis in Mgat5-deficient mice. Nat. Med. 6,
306–312
46 Zhao, Y. Y., Takahashi, M., Gu, J. G., Miyoshi, E., Matsumoto, A., Kitazume, S. and
Taniguchi, N. (2008) Functional roles of N-glycans in cell signaling and cell adhesion in
cancer. Cancer Sci. 99, 1304–1310