Optimisation of LRRK2 inhibitors and assessment of functional efficacy in cell-based models of neuroinflammation

Article abstract of DOI:10.1038/onc.2014.225

LRRK2IN1 is a highly potent inhibitor of leucine-rich repeat kinase 2 (LRRK2, IC₅₀ = 7.9 nM), an established target for treatment of Parkinson's disease. Two LRRK2IN1 analogues 1 and 2 were synthesised which retained LRRK2 inhibitory activity (1: IC₅₀ = 72 nM; 2: IC₅₀ = 51 nM), were predicted to have improved bioavailability and were efficacious in cell-based models of neuroinflammation. Analogue 1 inhibited IL-6 secretion from LPS-stimulated primary human microglia with EC₅₀ = 4.26 μM. In order to further optimize the molecular properties of LRRK2IN1, a library of truncated analogues was designed based on docking studies. Despite lacking LRRK2 inhibitory activity, these compounds show antineuroinflammatory efficacy at micromolar concentration. The compounds developed were valuable tools in establishing a cell-based assay for assessing anti-neuroinflammatory efficacy of LRRK2 inhibitors. Herein, we present data that IL-1β stimulated U87 glioma cell line is a reliable model for neuroinflammation, as data obtained in this model were consistent with results obtained using primary human microglia and astrocytes.

Full text of DOI:10.1038/onc.2014.225

Oncogene (2014), 1–9
© 2014 Macmillan Publishers Limited All rights reserved 0950-9232/14
www.nature.com/onc
SHORT COMMUNICATION
The p38-MK2-HuR pathway potentiates
EGFRvIII–IL-1β-driven IL-6 secretion in glioblastoma cells
FMS Gurgis1,7, YT Yeung1,2,7, MXM Tang1, B Heng3, M Buckland4, AJ Ammit2, J Haapasalo5,6, H Haapasalo5, GJ Guillemin3, T Grewal2 and
L Munoz1
The microenvironment of glioblastoma (GBM) contains high levels of inflammatory cytokine interleukin 6 (IL-6), which contributes
to promote tumour progression and invasion. The common epidermal growth factor receptor variant III (EGFRvIII) mutation in
GBM is associated with significantly higher levels of IL-6. Furthermore, elevated IL-1β levels in GBM tumours are also believed to
activate GBM cells and enhance IL-6 production. However, the crosstalk between these intrinsic and extrinsic factors within the
oncogene-microenvironment of GBM causing overproduction of IL-6 is poorly understood. Here, we show that EGFRvIII potentiates
IL-1β-induced IL-6 secretion from GBM cells. Importantly, exacerbation of IL-6 production is most effectively attenuated in
EGFRvIII-expressing GBM cells with inhibitors of p38 mitogen-activated protein kinase (p38 MAPK) and MAPK-activated protein
kinase 2 (MK2). Enhanced IL-6 production and increased sensitivity toward pharmacological p38 MAPK and MK2 inhibitors in
EGFRvIII-expressing GBM cells is associated with increased MK2-dependent nuclear–cytoplasmic shuttling and accumulation of
human antigen R (HuR), an IL-6 mRNA-stabilising protein, in the cytosol. IL-1β-stimulated activation of the p38 MAPK–MK2-HuR
pathway significantly enhances IL-6 mRNA stability in GBM cells carrying EGFRvIII. Further supporting a role for the p38 MAPK–MK2-
HuR pathway in the development of inflammatory environment in GBM, activated MK2 is found in more than 50% of investigated
GBM tissues and correlates with lower grade and secondary GBMs. Taken together, p38 MAPK–MK2-HuR signalling may enhance
the potential of intrinsic (EGFRvIII) and extrinsic (IL-1β) factors to develop an inflammatory GBM environment. Hence, further
improvement of brain-permeable and anti-inflammatory inhibitors targeting p38 MAPK, MK2 and HuR may combat progression
of lower grade gliomas into aggressive GBMs.
Oncogene advance online publication, 4 August 2014; doi:10.1038/onc.2014.225
INTRODUCTION
IL-6 expression is tightly regulated by extrinsic and intrinsic
signals targeting transcription factors like cAMP response
element-binding protein (CREB) and activating transcription
factor 1 (ATF-1) to control IL-6 promoter activity.7 In addition,
IL-6 mRNA is characterised by rapid turnover, which is controlled
by cellular factors, including human antigen R (HuR).7,9 Upon cell
stimulation, HuR monomer in the nucleus undergoes dimerisation
and translocates to the cytoplasm, where HuR dimer binds and
stabilizes its target mRNA.10 The nuclear–cytoplasmic shuttling of
HuR is critical for HuR-dependent mRNA stabilisation, and is
regulated via several signalling pathways.11 Most interestingly,
activation of p38 mitogen-activated protein kinase (p38 MAPK),
possibly via IL-1β and tumour necrosis factor-α (TNF-α), is
implicated in HuR-triggered stabilisation of mRNA coding for key
inflammatory mediators, including IL-6, IL-8, cyclooxygenase-2
(COX-2) and TNF-α.12–15 p38 MAPK-induced activation of MAPK-
activated protein kinase 2 (MAPK-APK2 or MK2) appears crucial for
HuR translocation and stabilisation of target mRNAs.16,17 In non-
stimulated cells, MK2 and p38 MAPK exist as a preformed inactive
complex in the nucleus. After activation by upstream kinases, for
example, MAPK kinase 6, p38 MAPK phosphorylates MK2 at
threonine 334 (T334), which enables the p38 MAPK–MK2 complex
Glioblastomas (GBM) are among the most lethal and least
successfully treated solid tumours. These brain tumours contain
high levels of inflammatory cytokines, in particular interleukins (IL)
IL-1β, IL-6 and IL-8, which promote GBM proliferation, stemness,
angiogenesis and invasion.1
Approximately 50% of all GBMs overexpress wild-type EGFR,
often together with the truncated and constitutively active
epidermal growth factor receptor variant III (EGFRvIII) mutant.2
Oncogenic activity of EGFRvIII correlates with overproduction of
IL-63 and is associated with upregulated EGFR effector Akt
activity.4 IL-6 is also produced by tumour-infiltrating microglia
and astrocytes5,6 and IL-1β, which is significantly elevated in GBM,
most likely via enhanced mitogen-activated protein kinase (MAPK)
signalling, promotes IL-6 production.7,8 Thus, IL-6 production is
triggered by multiple stimuli involving several tumour- and
microenvironment-induced signalling events. However, the cur-
rent understanding of IL-6 synthesis in GBM is almost exclusively
based on models utilising cancer or glial cells in isolation, not
incorporating the contribution of tumour-microenvironment
interactions.
1School of Medical Sciences/Pharmacology, University of Sydney, New South Wales, Australia; 2Faculty of Pharmacy, University of Sydney, New South Wales, Australia; 3Australian
School of Advanced Medicine, Macquarie University, New South Wales, Australia; 4School of Medical Sciences/Pathology, University of Sydney, New South Wales, Australia;
5Tampere University Hospital, Department of Pathology and Fimlab laboratories, Tampere, Finland and 6Tampere University Hospital, Unit of Neurosurgery, Tampere, Finland.
Correspondence: Dr L Munoz, School of Medical Sciences/Pharmacology, University of Sydney, New South Wales 2006, Australia.
E-mail: lenka.munoz@sydney.edu.au
7These authors contributed equally to this work.
Received 24 July 2013; revised 3 June 2014; accepted 20 June 2014

The MK2-HuR pathway in glioblastoma inflammation
FMS Gurgis et al
2
translocation. MK2 T334 phosphorylation exposes the MK2
substrate-binding site, which might potentiate HuR translocation
kinetics.18 Therefore, inhibition of the p38 MAPK–MK2 pathway
could interfere with cytoplasmic HuR accumulation, which is
rapidly developing into a new marker and potential hallmark of
cancer cells.19–21
Here we investigated IL-6 synthesis and secretion in the
oncogene-microenvironment context. Inflammatory cytokines,
such as IL-1β and TNF-α, show enhanced potential to upregulate
IL-6 secretion in EGFRvIII-expressing GBM cells. In these cells,
augmented potency of p38 MAPK and MK2 inhibitors to block IL-6
secretion is associated with an increased involvement of p38
MAPK and MK2 in nuclear–cytoplasmic HuR translocation and IL-6
mRNA stabilisation. In addition, the oncogenic EGFRvIII mutant
alters expression and activity of key regulators and binding
partners of HuR. Identification of activated MK2 in GBM tumours,
in particular lower grade and secondary GBMs, implicate the p38-
MK2-HuR pathway as a target to inhibit the development of an
inflammatory environment in GBM.
p38 MAPK signalling has been linked to the trafficking
of secretory vesicles.24,25 p38 MAPK inhibition reduced both
intra- and extracellular IL-6 levels much more efficiently in U87-
EGFRvIII (83.4 8.0%) compared with U87 cells (51.3 11.4%,
Supplementary Figure 1D). Taken together, p38 MAPK and MK2
inhibitors more efficiently inhibit inflammatory IL-6 cytokine
production in IL-1β-treated GBM cells expressing EGFRvIII.
We next compared IL-6 mRNA expression of IL-1β-incubated
U87 and U87-EGFRvIII cells SB203580 using RT–PCR (Figure 1d).
In line with increased IL-6 secretion (Figure 1a), basal (1.9 0.6-
fold) and IL-1β-induced IL-6 mRNA levels in U87-EGFRvIII cells
were increased compared with controls (139.0 8.5-fold and
40.8 19.6-fold, respectively; Figure 1d). SB203580 significantly
reduced IL-6 mRNA levels upon IL-1β stimulation by 34.0 11.2%
and 82.9 4.2% in U87 and U87-EGFRvIII cells, respectively. By
contrast, p38 MAPK inhibition did not reveal increased potency to
reduce IL-1β-induced IL-8 mRNA levels in U87-EGFRvIII cells
compared with controls (Supplementary Figure 1E). Hence,
increased potency of SB203580 to block cytokine production in
EGFRvIII cells specifically affect IL-6 mRNA levels.
Then we compared the phosphorylation kinetics (t = 0–120 min)
of p38 MAPK, MK2 and heat-shock protein 27 (Hsp27),
a downstream target of MK2, in IL-1β-stimulated U87 and
U87-EGFRvIII cells (Supplementary Figure 2). However, p38 MAPK,
MK2 and Hsp27 phosphorylation kinetics were similar in U87 and
U87-EGFRvIII cells, indicating that EGFRvIII or IL-1β-induced
EGFR wild-type transactivation26 did not impact on p38
MAPK-dependent MK2 and Hsp27 phosphorylation.
RESULTS AND DISCUSSION
Oncogenic EGFRvIII upregulates basal secretion of IL-6 and IL-8 in
unstimulated GBM cells.3,22 To address if EGFRvIII could enhance
IL-6 secretion in GBM cells upon extrinsic stimulation, we treated
U87 and U87-EGFRvIII cells with increasing amounts of IL-1β
(0–100 ng/ml) and compared IL-6 secretion (Figure 1a). IL-1β
stimulated U87-EGFRvIII consistently secreted approximately four-
fold higher quantities of IL-6 when compared with U87 cells.
Increasing concentrations of IL-1β also elevated IL-8 secretion in
U87-EGFRvIII cells (Supplementary Figure 1A).
p38 MAPK inhibitors efficiently reduce IL-6 secretion in
IL-1β-treated U251 GBM cells.8 To investigate the role of p38
MAPK–MK2 signalling for IL-6 secretion in U87 and U87-EGFRvIII
cells, cells were pre-incubated with p38 MAPK inhibitor SB203580
(SB) or MK2 inhibitor, compound sc-221948 (thereafter sc-48,
Figure 1b). IL-1β-induced IL-6 secretion in U87 cells was reduced
by 27.6 4.3% with SB203580; however, this reduction was not
statistically significant. SB203580 was much more potent to inhibit
IL-1β-induced secretion of IL-6 from U87-EGFRvIII cells
(68.2 15.7%). Similarly, IL-6 secretion in IL-1β-incubated U87-
EGFRvIII cells was more effectively inhibited with the MK2 inhibitor
sc-48 as compared with controls (68.2 18.8% vs 6.8 1.1%).
Augmented susceptibility of U87-EGFRvIII cells toward SB203580
was not due to increased SB203580 internalisation
(Supplementary Figures 1B and C).
To address efficacy of p38 MAPK and MK2 inhibitors, we
analysed p38 MAPK and MK2 phosphorylation SB203580 and
sc-48. As above, p38 MAPK activation by IL-1β was comparable in
U87 and U87-EGFRvIII cells (Figure 2a, lanes 2 and 5). In line with
published data,27 SB203580 attenuated IL-1β-induced p38 MAPK
phosphorylation by 49.9 11.8% in U87 cells (lanes 2–3) and by
75.6 17.9% in U87-EGFRvIII cells (lanes 5–6). Despite increased
potency of SB203580 to inhibit p38 MAPK phosphorylation in
U87-EGFRvIII cells, this did not translate into increased inhibition
of p38 MAPK activity, as SB203580 completely blocked MK2 phos-
phorylation in both U87 and U87-EGFRvIII cells (lanes 3 and 6).
We then addressed MK2 inhibitor sc-48 efficacy. U87 and
U87-EGFRvIII cells were pre-treated with sc-48 and stimulated with
IL-1β for 15 min. As sc-48 only moderately inhibits MK2 activation,
but attenuates MK2 kinase activity,28 we analysed phosphorylation
of the MK2 substrate Hsp27. In line with data shown above,
IL-1β-induced MK2 phosphorylation was evident in both cell lines,
and was not downregulated with sc-48 (Figure 2b, lanes 3 and 6).
Hsp27 phosphorylation IL-1β and MK2 inhibitor sc-48 were
comparable in U87 and U87-EGFRvIII cells (lanes 2, 3 and 5, 6).
Taken together, inhibition of p38 MAPK and MK2 more potently
reduced IL-6 secretion in IL-1β-activated U87-EGFRvIII cells
(Figure 1), yet magnitude of p38 MAPK, MK2 and Hsp27 inhibition,
as judged by phosphorylation, appeared comparable in U87 and
U87-EGFRvIII cells (Figure 2).
To investigate this further, we examined CREB and ATF-1,
downstream targets of p38 MAPK that regulate IL-6 promoter
activity. However, high CREB and ATF-1 phosphorylation, indica-
tive of upregulated CREB/ATF-1 activity, was evident in control
and IL-1β-incubated cells (Supplementary Figure 3A). In addition,
SB203580 did not significantly attenuate IL-6 promoter-driven
luciferase activity in IL-1β-stimulated U87 and U87-EGFRvIII cells
(Supplementary Figure 3B). Taken together, the p38 MAPK–MK2
pathway does not seem to play a major role in the transcriptional
regulation of IL-6 expression in U87 cells.
To address the contribution of other EGFR effector pathways23
in IL-1β-induced IL-6 secretion, we next compared IL-6 secretion
upon incubation with p38 MAPK (069A), MK2 (sc-38), MAPK kinase
Mek1/2 (PD098059) and c-Jun N-terminal kinase (JNK) (SP600125)
inhibitors (Figure 1c). As above, p38 MAPK and MK2 inhibitors
were significantly (4threefold) more effective in attenuating
IL-1β-induced IL-6 secretion in U87-EGFRvIII cells. Most strikingly,
Mek1/2 or JNK inhibition did not downregulate IL-6 secretion in
U87 and U87-EGFRvIII cells.
We next compared p38 MAPK, Mek1/2 and JNK inhibitors to
block TNF-α and IL-1β-induced IL-6 and IL-8 expression
(Supplementary Table 1). In all settings, Mek1/2 and JNK inhibitors
reduced IL-6 and IL-8 secretion with similar efficacy in U87 and
U87-EGFRvIII cells or were even less effective in EGFRvIII-
expressing cells. Hence, oncogenic EGFR does not seem to alter
the contribution of Mek1/2 and JNK in IL-6 and IL-8 secretion from
IL-1β- or TNF-α-stimulated GBM cells. Interestingly, only IL-1β- and
TNF-α-induced IL-6, but not IL-8, secretion was highly susceptible
to p38 MAPK and MK2 inhibition in U87-EGFRvIII cells
(Supplementary Table 1 and Figures 1b and c), strongly indicating
that the p38 MAPK–MK2 pathway specifically targets molecular
events controlling some, but not all members of the IL family.
We next studied the involvement of the p38 MAPK–MK2
pathway in HuR-dependent post-transcriptional regulation of IL-6
mRNA expression (Figure 3a). Indeed, siRNA-targeting HuR
effectively downregulated HuR levels in U87 (87.4 0.1%) and
U87-EGFRvIII cells (82.6 2.6%). HuR knockdown was associated
Oncogene (2014), 1 – 9
© 2014 Macmillan Publishers Limited

The MK2-HuR pathway in glioblastoma inflammation
FMS Gurgis et al
3
Figure 1. Inhibition of p38 MAPK–MK2 signalling attenuates IL-6 secretion in glioblastoma cells. (a) U87 and U87-EGFRvIII (5 × 104) cells were
treated with IL-1β (0–100 ng/ml) for 24 h and secretion of IL-6 was determined by enzyme-linked immunosorbent assay (ELISA). Values were
normalised to cell viability (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT)) and expressed as fold increase (mean s.e.m.)
compared with untreated U87 cells. (b–c) U87 and U87-EGFRvIII (5 × 104) cells were starved for 2 h, pre-treated with compounds (10 μM)
inhibiting (b) p38 MAPK (SB203580, SB), MK2 (sc-221948, sc-48) or (c) p38 MAPK (069A,36), MK2 (sc-203138, sc-38 1 μM), MEK1/2 (PD98059, PD)
or JNK (SP600125, SP) for 60 min and treated with IL-1β (10 ng/ml) for 24 h. Secretion of IL-6 was determined using IL-6 ELISA kit and
normalised to cell viability as above. See Supplementary Material for further details. (d) U87 and U87-EGFRvIII (3 × 105) cells were starved for
2 h, pre-treated with the 10 μM SB203580 (SB) and treated with IL-1β (10 ng/ml) for 24 h. IL-6 mRNA was determined by RT–PCR and normalised
to 18 s mRNA (for details see Supplementary Material and Munoz L et al.37) data are expressed as fold increase compared with untreated cells.
All data are mean s.e.m. from at least three independent experiments performed in duplicates. (**Po0.01, ***Po0.001, one-way analysis of
variance followed by Newman–Keuls post-test using Prism 5 GraphPad Software).
with 56.5 3.4% reduced IL-6 production in IL-1β-stimulated U87
cells (Figure 3a). Moreover, HuR knockdown in IL-1β-treated
U87-EGFRvIII cells decreased IL-6 secretion by 76.3 3.0%
(Figure 3a), indicating increased involvement of HuR in IL-6 mRNA
stabilisation in EGFRvIII.
HuR accumulated in the cytoplasm (72.1 5.4%, lane 5). These
findings suggested that EGFRvIII may enhance HuR translocation
in GBM cells.
In order to explain increased cytosolic HuR accumulation in
U87-EGFRvIII cells, we compared expression and activation of
several HuR-interacting proteins in U87 and U87-EGFRvIII cells
(Figure 3c). Total amounts of p38 MAPK and MK2 in U87 and
U87-EGFRvIII were comparable (Figure 3c and Supplementary
Figure 4A). Slightly elevated expression of MK2 was found in
EGFRvIII-expressing cells, but the levels did not reach statistical
significance when compared with controls. Intriguingly, U87-
EGFRvIII cells showed a twofold difference in the expression and
inactivation of cyclin-dependent kinase 1 (Cdk1). Cdk1 is a cell
cycle kinase that is inactivated through phosphorylation in the
event of DNA damage, causing G2 arrest.29 Importantly, HuR
shuttling is controlled by the Cdk1-mediated phosphorylation of
Ser202 that acts to retain HuR in the nucleus.30 Increased Cdk1
inactivation, most likely due to the EGFRvIII-mediated replicative
stress and DNA damage causing G2 arrest,31 correlated with
Accumulation of cytoplasmic HuR may be common in tumours,
including GBM.20,21 To possibly identify a link between oncogenic
EGFR and nuclear–cytoplasmic HuR shuttling, U87 and U87-
EGFRvIII cells were harvested and whole-cell lysates were
subjected to subcellular fractionation. Both cell lines express
comparable amounts of HuR (Figure 3c). Non-malignant primary
human astrocytes served as control. Nuclear fractions, enriched
with nuclear marker lamin A/C, and cytosolic fractions were
prepared (Figure 3b). Activated HuR forms a stable dimer,10 hence
reducing conditions were used to ensure HuR monomer (37 kDa)
analysis. Consistent with published data, HuR was predominantly
found in nuclear fractions of primary human astrocytes (lane 2).
In U87 GBM cells, increased quantities of HuR were evident in the
cytoplasmic fraction (lane 3). Most strikingly, in U87-EGFRvIII cells,
© 2014 Macmillan Publishers Limited
Oncogene (2014), 1 – 9

The MK2-HuR pathway in glioblastoma inflammation
FMS Gurgis et al
4
decreased levels of pHuR(S202) in U87-EGFRvIII (Figure 3c) and
increased cytosolic HuR accumulation (Figure 3b). It is important
to note that only unphosphorylated HuR shuttles into the cytosol
and pHuR(S202) could not be detected in the cytosolic fractions.
This mechanism could contribute to increase HuR accumulation in
the cytoplasm and thereby increase efficacy of p38 MAPK and
MK2 inhibitors, as p38 MAPK–MK2 signalling is activated by DNA
damage and causes inactivation of Cdk1 and G2 arrest.29 In
addition, we found that U87-EGFRvIII cells express significantly
more chromosome region maintenance 1 (CRM1/exportin 1)
protein (Figure 3c). CRM1 is a ubiquitous transport receptor that
recognises the nuclear export sequence of activated MK2 and
Figure 2. IL-1β activates the p38 MAPK–MK2 pathway in U87 and U87-EGFRvIII cells. (a–b) U87 and U87-EGFRvIII (4 × 105) cells were starved for
2 h, pre-treated with (a) p38 MAPK inhibitor SB203580 (SB, 10 μM) or (b) MK2 inhibitor sc-48 (10 μM) for 60 min and treated with IL-1β (10 ng/ml)
for 15 min. Protein concentration of whole-cell lysates was determined and 25 μg cellular protein was subjected to western blot analysis as
described (Yeung Y et al.8, see Supplementary Material) and analysed for p38 MAPK phosphorylation (p-p38), total p38 MAPK (p38),
phosphorylated MK2 (p-MK2), total MK2 (MK2), phosphorylated Hsp27 (p-Hsp27), total Hsp27 (Hsp27), β-actin and β-tubulin (all Cell Signalling)
as indicated. Relative levels of p-p38 and p-Hsp27 were normalised to total p38 and total Hsp27, respectively, expressed as fold increase
compared with untreated (Ctr) cells. Representative blots and quantification of three (a) and five (b) independent experiments are shown.
Data represent the mean s.e.m. (***Po0.001, one-way analysis of variance followed by Newman–Keuls post-test using Prism 5 GraphPad
Software).
Figure 3. HuR is critical for IL-6 secretion in U87 glioblastoma cells. (a) U87 and U87-EGFRvIII (2 × 105) cells were transfected with siRNA
targeting HuR (5′-ACUUAUUCGGGAUAAAGUATT-3′, 5′-UACUUUAUCCCGAAUAAGUTT-3′) (for details see Supplementary Material). Scrambled
siRNA served as control. After 72 h, cells were stimulated with IL-1β (10 ng/ml) for 24 h. IL-6 secretion was determined by enzyme-linked
immunosorbent assay (ELISA) as described above. Whole-cell lysates were subjected to western blot analysis to confirm HuR knockdown
(87.4 0.1% and 82.6 2.6% in U87 and U87-EGFRvIII cells, respectively). Representative blot and data analysis (mean s.e.m.; ***Po0.001,
one-way analysis of variance followed by Newman–Keuls post-test) of three independent experiments is shown. (b) U87 and U87-EGFRvIII
(1 × 107) cells were subjected to subcellular fractionation (see Supplementary Material for details). Primary human astrocytes served as control
in b and were isolated as described.38 Protein concentrations of nuclear and cytosolic fractions were determined and 15–25 μg of total protein
was subjected to western blot analysis. Cellular distribution of HuR, nuclear marker lamin A/C and β-tubulin was analysed. Nuclear and
cytosolic HuR distribution was quantified and normalised to lamin A/C or β-tubulin, respectively. Representative blots and data (mean s.e.m.,
*Po0.05, **Po0.01, ***Po0.001, one-way analysis of variance followed by Newman–Keuls post-test) from three independent experiments is
shown. (c) Whole-cell lysates from unstimulated U87 EGFRvIII (40 μg) were subjected to western blot analysis and analysed for EGFR and
EGFRvIII, p38 MAPK (p38), total MK2 (MK2), phosphorylated Cdk1 (p-Cdk1), total Cdk1 (Cdk1), total HuR (HuR), phosphorylated HuR (pHuR
(S202),30 CRM1 and β-tubulin (all Cell Signalling) as indicated. Representative blots and data (mean s.e.m., *Po0.05; **Po0.01; ***Po0.001,
paired Student's t-test) from three independent experiments are shown. (d) Nuclear (15 μg) and cytosolic (30 μg) fractions from U87 and
U87-EGFRvIII were prepared and analysed by western blotting for total p38 MAPK (p38), phosphorylated and total MK2 (p-MK2 and MK2,
respectively), phosphorylated and total Cdk1 (p-Cdk1 and Cdk1, respectively), β-tubulin and lamin A/C. Protein expression was quantified and
normalised against lamin A/C (nuclear fractions) and β-tubulin (cytosolic fraction), expressed as fold increase compared with U87.
Representative blots and data (mean s.e.m., *Po0.05; paired Student's t-test) from three independent experiments are shown.
Oncogene (2014), 1 – 9
© 2014 Macmillan Publishers Limited

The MK2-HuR pathway in glioblastoma inflammation
FMS Gurgis et al
5
exports MK2-complexes into the cytoplasm.32 Thus, increased
CRM1 expression may enhance MK2-HuR export from the nucleus
of U87-EGFRvIII cells, where MK2 is predominantly located.
Next, the subcellular distribution of p38 MAPK and MK2 was
analysed (Figure 3d). U87 cells showed less nuclear p38 MAPK
(lane 1) compared with U87-EGFRvIII cells (lane 2) and MK2 also
appeared enriched in nuclear fractions from U87-EGFRvIII cells
(lane 2). Similarly, expression and phosphorylation of Cdk1 were
enhanced in the nuclear fractions from EGFRvIII-expressing cells
(lane 2). Together, this data suggested that increased nuclear
localisation of several HuR-interacting proteins, in particular MK2,
in U87-EGFRvIII cells improves signal efficacy to activate and
promote nuclear HuR translocation. These changes in subcellular
p38 MAPK, MK2 and HuR distribution correlate with increased
efficacy of SB203580 and sc-48 to attenuate IL-6 secretion in
these cells.
As IL-1β induces nuclear HuR export33 and MK2 facilitates HuR
shuttling,34 we investigated if MK2 inhibition could attenuate
U87
U87-EGFRvIII
scramble
HuR siRNA
Ctr IL-1β
scramble
Ctr IL-1β
HuR siRNA
Ctr IL-1β
Ctr
IL-1β
HuR
β-tubulin
3
4
5
6
7
8
1
Primary
astrocytes
Cyt Nuc
U87 U87-EGFRvIII
U87 U87-
EGFRvIII
Cyt Nuc Cyt Nuc
EGFR
HuR
lamin A/C
β-tubulin
EGFRvIII
p38
MK2
p-Cdk1
1
2
3
4
5
6
Cdk1
HuR
p-HuR
CRM1
β-tubulin
1
2
nucleus
cytosol
U87
U87- U87
EGFRvIII
U87-
EGFRvIII
p38
p-MK2
total MK2
p-Cdk1
Cdk1
lamin A/C
β-tubulin
1
2
3
4
© 2014 Macmillan Publishers Limited
Oncogene (2014), 1 – 9

The MK2-HuR pathway in glioblastoma inflammation
FMS Gurgis et al
6
Figure 4. MK2 inhibition attenuates nuclear–cytoplasmic translocation of HuR and IL-6 mRNA stability in U87 cells. (a–b) U87 and U87-EGFRvIII
(1 × 107) cells were starved for 2 h, pre-treated with sc-48 (10 μM) for 60 min and activated with IL-1β (10 ng/ml) for 8 h. Nuclear and cytosolic
fractions were prepared and analysed by western blotting for the distribution of HuR, lamin A/C and β-tubulin (for details see Supplementary
Material). Relative HuR levels were quantified and normalised to lamin A/C (nuclear fractions) or β-tubulin (cytosolic fractions). Representative
blots and data (mean s.e.m., *Po0.05, paired Student's t-test) from three independent experiments are shown. (c–d) U87 and U87-EGFRvIII
(4 × 105) cells were starved for 2 h, pre-treated with sc-48 (10 μM) for 60 min (d) and stimulated with IL-1β (10 ng/ml) for 4 h (c–d). Cells were
then washed and incubated with actinomycin D (5 μg/ml). Total RNA was extracted at indicated timepoints and IL-6 mRNA quantified. Results
are expressed as % mRNA remaining over time (for details see Supplementary Material). Data are mean s.e.m. from three independent
experiments. (e) Representative images of GBM tissues stained with antibody recognising activated MK2 (pT334-MK2) (for details see
Supplementary Material).
Oncogene (2014), 1 – 9
© 2014 Macmillan Publishers Limited

The MK2-HuR pathway in glioblastoma inflammation
FMS Gurgis et al
7
Table 1. Formalin-fixed, paraffin-embedded tissues from 205 brain tumour patients with grade II–IV astrocytomas who underwent surgery in the
Unit of Neurosurgery, Tampere University Hospital, Tampere, Finland, during 1983–2001 were used for the construction of TMA
Relative p-MK2 expression
Score
0
1
2
3
Histological type
N
n (%)
n (%)
n (%)
n (%)
Astrocytoma (WHO grade 2)
Anaplastic astrocytoma (WHO grade 3)
Glioblastoma (WHO grade 4)
Subtotal
25
31
149
8 (32)
10 (32)
78 (52)
96
3 (12)
2 (6)
11 (7)
16
12 (48)
15 (48)
52 (35)
79
2 (8)
4 (13)
8 (5)
14
Total
205
96 (47%)
109 (53%)
N
p-MK2 negative (o5% positive cells) n (%)
p-MK2 positive (45% positive cells) n (%)
Significance (χ2-test)
P = 0.02
Primary GBM
119
5
63 (53)
0 (0)
56 (47)
5 (100)
49 (43)
31 (70)
18 (38)
29 (61)
50 (54)
Secondary GBM
IDH-1 (negative)
IDH -1 (positive)
p53 (negative)
p53 (positive)
EGFR (no
115
44
48
48
92
66 (57)
13 (30)
30 (62)
19 (39)
42 (46)
P = 0.002
P = 0.025
amplification)
EGFR (amplification)
44
26 (59)
18 (41)
n.s.
Abbreviations: EGFR, epidermal growth factor receptor; GBM, glioblastoma; IDH, isocitrate dehydrogenase; n.s., non significant; p-MK-2, phosphorylated MK2;
TMA, tissue microarrays; WHO, World Health Organization. The study was approved by the Ethical Review Board of Tampere University Hospital, Tampere,
Finland (Dnro R07042), the National Authority for Medicolegal Affairs of Finland (Dnro 1502/04/046/07) and the Human Ethics Committee of the University of
Sydney (HREC 2013/131). Activated MK2 (p-MK2) was detected with rabbit polyclonal antibody recognising phosphorylated MK2 (T334, Cell Signaling
Technology, Beverly, MA, USA, for details see Supplementary Material). Tissue sections were counterstained with haematoxylin, and p-MK2 immunoreactivity
was evaluated by two pathologists using a consultation microscope (for representative images, see Figure 4e).
IL-1β-induced nuclear export of HuR in GBM cells. Therefore U87
and U87-EGFRvIII were pre-treated with sc-48, followed by
incubation with IL-1β for 8 h (Figure 4). As HuR activation involves
cysteine-dependent dimerisation, non-reducing conditions to
detect HuR monomers (~37 kDa) and dimers (~70 kDa) were
chosen (for details see Supplementary Material). The identity of
the band corresponding to HuR dimer was confirmed by its
sensitivity toward HuR depletion in immunoblots from HuR
knockdown lysates (Supplementary Figure 4B). Monomeric HuR
was found in the nucleus of unstimulated and IL-1β sc-48-
incubated U87 cells (Figure 4a, lane 1–3). Since HuR is abundant in
the nucleus, but under normal or non-stimulated conditions
almost undetectable in the cytosol, HuR shuttling is generally
based on analysis of cytoplasmic fractions.30 Nevertheless, we
observed a 1.5-fold increase in monomeric HuR nuclear levels in
IL-1β-treated U87 cells (Figure 4a, lane 1–2). Comparable amounts
of HuR monomer and dimer were detected in the cytosol of U87
cells in the presence or absence of IL-1β (Figure 4a, lanes 4–5). This
suggests that IL-1β increases HuR expression (see also Figure 3a),
without increasing HuR cytoplasmic accumulation. Importantly,
MK2 inhibitor sc-48 significantly reduced cytosolic HuR monomers
and dimers in U87 cells (lane 6), which was accompanied by a
twofold increase in nuclear HuR (lane 3). Hence, blocking MK2-
dependent HuR cytoplasmic accumulation could reduce IL-6
expression in U87 cells.
Most strikingly, U87-EGFRvIII cells treated with IL-1β displayed a
0.5-fold decrease in nuclear HuR and 2.5-fold HuR increase in the
cytosolic fraction (Figure 4b, lanes 2 and 5). As observed in U87
cells, MK2 inhibition by sc-48 effectively reduced cytosolic HuR
(see decreased monomer and dimer in lane 6), which was
accompanied by a twofold increased nuclear amounts of HuR
monomers (lane 3).
To further explore the molecular mechanism underlying
enhanced HuR shuttling in EGFRvIII-expressing cells after IL-1β
treatment, we compared the localisation of MK2 and phosphor-
ylation state of HuR in U87 cells EGFRvIII (Supplementary
Figure 5A–B). In U87 cells, MK2 is primary localised in the cytosol
(Supplementary Figure 5B, lane 1) and upon IL-1β stimulation,
activated MK2 was distributed in both nuclear (Supplementary
Figure 5A, lanes 2–3) and cytosolic fractions (Supplementary
Figure 5B, lanes 2–3; for quantification see bottom panels). In
addition, IL-1β stimulation of U87 cells did not significantly
increase phosphorylation-dependent inactivation of nuclear Cdk1
and consistent with these observations, pHuR(S202) levels
remained unchanged in these cells (Supplementary Figure 5A,
lanes 1–3). Consistent with data shown above (Figure 3d), in
unstimulated U87-EGFRvIII cells increased amounts of MK2 were
found in the nucleus (Supplementary Figure 5A, lane 4) and
activated p-MK2 remained localised in the nucleus upon IL-1β
stimulation (lanes 5–6), whereas low p-MK2 levels were detected
in the cytoplasmic fractions (Supplementary Figure 5B, lanes 5–6).
Enhanced nuclear accumulation of activated MK2 was accom-
panied by significant phosphorylation-dependent inactivation of
Cdk1 (Supplementary Figure 5A, lanes 5–6) and 0.63-fold decrease
in pHuR(S202) levels (Supplementary Figure 5A, lanes 6). Taken
together, this could explain the lack of increased cytoplasmic HuR
levels in IL-1β-stimulated U87 cells compared with U87-EGFRvIII
cells (Figure 4a). Similarly, p38 MAPK inhibitor SB203580 was more
effective in preventing nuclear export of HuR in U87-EGFRvIII cells
(Supplementary Figure 5C). In summary, MK2 accumulation in the
nucleus and increased nuclear–cytosolic shuttling of HuR could
be responsible for enhanced IL-6 production in IL-1β-stimulated
U87-EGFRvIII cells.
To address if inhibition of HuR shuttling by MK2 inhibitor sc-48
altered IL-6 mRNA stability, U87 EGFRvIII cells were pre-treated
sc-48 (10 μM) and stimulated with IL-1β for 4 h. Transcription was
then halted using actinomycin D, and real-time RT–PCR was used
to measure IL-6 mRNA degradation over time to determine the
kinetics of IL-6 mRNA decay. In the absence of sc-48, IL-6 mRNA
decay was slower in IL-1β-treated U87-EGFRvIII cells (decay
constant k = 0.66 0.2; t1/2 = 1.47 0.4 h) compared with U87 cells
(decay constant k = 1.88 0.4; t1/2 = 0.34 0.1 h, Figure 4c).
© 2014 Macmillan Publishers Limited
Oncogene (2014), 1 – 9

The MK2-HuR pathway in glioblastoma inflammation
FMS Gurgis et al
8
U87
U87-EGFRvIII
IL-1
IL-6
IL-6
IL-1
EGFRvIII
P
P
p-p38
p-MK2
p-p38
p-p38
Nucleus
Nucleus
IL-6 mRNA
IL-6 mRNA
stabilisation
p-MK2
stabilisation
p-MK2
p-Cdk1
p-Cdk1
p-HuR
HuR
HuR
p-HuR
HuR
HuR
Cytoplasm
Cytoplasm
Figure 5. Proposed model of the molecular mechanism underlying IL-6 production in an inflammatory microenvironment with elevated IL-1β
levels stimulating glioblastoma cells with and without EGFRvIII. In U87 cells, MK2 is primarily located in the cytosol. Upon IL-1β treatment, the
activated p38 MAPK–MK2 complex translocates to the nucleus, which leads to phosphorylation-dependent inactivation of Cdk1. This process
is amplified in U87-EGFRvIII cells where MK2 is predominantly located in the nucleus. This potentiates Cdk1 phosphorylation and
consequently, increased amounts of inactive nuclear Cdk1 shift the phosphorylation equilibrium of HuR toward reduction of phosphorylated
HuR (pHuR S202) levels. Importantly, only unphosphorylated HuR is able to translocate into the cytosol of U87-EGFRvIII after IL-1β treatment.
In the cytoplasm, HuR stabilises IL-6 mRNA levels, leading to increased IL-6 protein production and secretion.
As shown in Figure 4d, sc-48 reduced IL-1β-induced IL-6
mRNA stability in U87-EGFRvIII cells (decay constant k = 2.1 0.9;
t1/2 = 0.48 0.2 h), but increased IL-6 mRNA stability in U87
cells (decay constant k = 0.55 0.1; t1/2 = 1.43 0.7 h). Increased
involvement of HuR in IL-6 mRNA stabilisation in U87-EGFRvIII cells
may increase sensitivity toward the MK2 inhibitor sc-48, which as
shown here interferes with HuR shuttling, IL-6 mRNA stability and
IL-6 secretion (Figures 4b–d, Figure 1b).
are characterised by increased amounts of cytosolic HuR, an
mRNA-stabilising protein.9,20,21 This is associated with upregulated
IL-6 secretion, which is a key factor contributing to GBM tumour
progression.3,4 Most importantly, IL-1β-induced and EGFRvIII-
exacerbated HuR translocation and IL-6 secretion is efficiently
attenuated by p38 MAPK–MK2 pathway inhibition. Mechanisti-
cally, EGFRvIII increases nuclear accumulation of p38 MAPK and
MK2 in U87 glioma cells. When activated by IL-1β, the proximity
and nuclear enrichment of these two inflammatory kinases is
associated with enhanced Cdk1 inactivation, HuR dephosphoryla-
tion and shuttling, consequently increasing IL-6 mRNA stability
and IL-6 secretion (see model in Figure 5). The accumulation of
HuR in the cytoplasm of unstimulated EGFRvIII-expressing cells
is further connected to upregulation of Cdk1 and CRM1, which is
currently under investigation. Most importantly, activated MK2 is
expressed in 450% of investigated GBM tissues, predominantly in
lower grade and secondary gliomas, suggesting potential for MK2
inhibitors as anti-inflammatory agents to combat progression of
lower grade gliomas into aggressive GBMs.
Our data indicated that MK2 activation may be associated with
brain tumour progression. We therefore compared p-MK2 staining
in glioma tumours (n = 205) of different grade (Table 1). Non-
malignant brain samples (n = 6) served as control and all showed
no p-MK2 immunoreactivity (Figure 4e). In diffuse astrocytomas
(WHO (World Health Organization), grade 2, n = 25), 68% tumours
showed weak, moderate or strong p-MK2 staining in more than
5% cells. In anaplastic astrocytomas (WHO grade 3, n = 31), 68%
tumours were classified as positive for p-MK2. In the most
aggressive GBM cohort (WHO grade 4, n = 149), 48% tumours
showed weak, moderate or strong staining for p-MK2 (Table 1 and
Figure 4e). Statistical association of MK2 activation with lower
WHO grade gliomas was observed (P = 0.036, χ2 test). Lower grade
gliomas often develop into secondary GBMs. Analysis of all
secondary GBMs (n = 5) showed strong staining for p-MK2.
Furthermore, we found correlation between p-MK2-positive
tumours with isocitrate dehydrogenase 1 and p53 mutation
(Table 1), both markers of lower grade gliomas and secondary
GBMs.35 Thus, MK2 may contribute to the progression of lower
grade glioma tumours into GBMs, possibly involving the devel-
opment of an inflammatory microenvironment enriched with
inflammatory cytokines, including IL-6. p-MK2 immunohistochem-
istry did not correlate with EGFR amplification, a marker of
primary, but not lower grade or secondary GBMs.35 Nevertheless,
from a cohort of 44 tumours with EGFR amplification, 18 tumours
(41%) expressed activated MK2, suggesting that a subgroup of
GBM tumours with EGFR amplification could hinge on the
molecular mechanism described in this study.
ABBREVIATIONS
ATF-1, activating transcription factor 1; Cdk1, cyclin-dependent
kinase 1; COX-2, cyclooxygenase-2; CREB, cAMP response
element binding; CRM1, chromosome region maintenance 1;
DTT, dithiothreitol; EGFR, epidermal growth factor receptor; FBS,
fetal bovine serum; GBM, glioblastoma; HuR, human antigen R;
Hsp27, heat-shock protein 27; IDH, isocitrate dehydrogenase;
IL, interleukin; JNK, c-Jun N-terminal kinase; LPS, lipopolysacchar-
ide; MAPK, mitogen-activated protein kinase; Mek1/2, MAPK
kinase 1/2; MK2, MAPK-activated protein kinase 2; MKK3, MAPK
kinase 3; PI3K, phosphatidylinositol-3-kinase; TNF-α, tumour
necrosis factor-α.
In summary, this study provides insight into the differential
regulation of IL-6 production in the oncogene-microenvironment
context of GBM biology. GBM cells expressing oncogenic EGFRvIII
CONFLICT OF INTEREST
The authors declare no conflict of interest.
Oncogene (2014), 1 – 9
© 2014 Macmillan Publishers Limited

The MK2-HuR pathway in glioblastoma inflammation
FMS Gurgis et al
9
ACKNOWLEDGEMENTS
interleukin-6 independently at different post-transcriptional levels. J Biol Chem
2002; 277: 3065–3068.
18 Gaestel M. MAPKAP kinases - MKs- two's company, three's a crowd. Nat Rev Mol
Cell Biol 2006; 7: 120–130.
This study was supported by grants from National Foundation for Medical Research
and Innovation and the University of Sydney Brown Fellowship to LM. TG
acknowledges support from the National Health and Medical Research Council of
Australia (510294) and the University of Sydney (2010-02681). GJG is supported by
the Australian Research Council (FT120100397) and Tour de Cure. AJA is supported
by the National Health and Medical Research Council of Australia (1025637).
We would like to thank Maggie Lee for technical assistance with immunohistochem-
istry and Myriam Gorospe for pHuR(S202) antibody.
19 Miyata Y, Watanabe S, Sagara Y, Mitsunari K, Matsuo T, Ohba
K et al.
High expression of HuR in cytoplasm, but not nuclei, is associated with
malignant aggressiveness and prognosis in bladder cancer. PLoS ONE 2013; 8:
e59095.
20 Bolognani F, Gallani A-I, Sokol L, Baskin D, Meisner-Kober N. mRNA stability
alterations mediated by HuR are necessary to sustain the fast growth of
glioma cells. J Neurooncol 2012; 106: 531–542.
21 Filippova N, Yang X, Wang Y, Gillespie GY, Langford C, King PH et al. The
RNA-binding protein HuR promotes glioma growth and treatment resistance. Mol
Cancer Res 2011; 9: 648–659.
22 Bonavia R, Inda MM, Vandenberg S, Cheng S-Y, Nagane M, Hadwiger P et al.
EGFRvIII promotes glioma angiogenesis and growth through the NFkB,
interleukin-8 pathway. Oncogene 2011; 31: 4054–4066.
23 Avraham R, Yarden Y. Feedback regulation of EGFR signalling: decision making by
early and delayed loops. Nat Rev Mol Cell Biol 2011; 12: 104–117.
24 Brzezinska AA, Johnson JL, Munafo DB, Ellis BA, Catz SD. Signalling mechanisms
for Toll-like receptor-activated neutrophil exocytosis: key roles for interleukin-1-
receptor-associated kinase-4 and phosphatidylinositol 3-kinase but not Toll/IL-1
receptor (TIR) domain-containing adaptor inducing IFN-β (TRIF). Immunology
2009; 127: 386–397.
25 Sumara G, Formentini I, Collins S, Sumara I, Windak R, Bodenmiller B et al.
Regulation of PKD by the MAPK p38δ in insulin secretion and glucose home-
ostasis. Cell 2009; 136: 235–248.
REFERENCES
1 Yeung YT, McDonald KL, Grewal T, Munoz L. Interleukins in glioblastoma patho-
physiology: implications for therapy. Br J Pharmacol 2013; 168: 591–606.
2 The Cancer Genome Atlas Research, Comprehensive genomic characterization
defines human glioblastoma genes and core pathways. Nature 2008; 455:
1061–1068.
3 Inda M, Bonavia R, Mukasa A, Narita Y, Sah D, Vandenberg S et al. Tumor
heterogeneity is an active process maintained by
a mutant EGFR-induced
cytokine circuit in glioblastoma. Genes Dev 2010; 24: 1731–1745.
4 Jin X, Yin J, Kim S-H, Sohn Y-W, Beck S, Lim YC et al. EGFR-AKT-Smad signaling
promotes formation of glioma stem-like cells and tumor angiogenesis by
ID3-driven cytokine induction. Cancer Res 2011; 71: 7125–7134.
5 Zhang J, Sarkar S, Cua R, Zhou Y, Hader W, Yong VW. A dialog between glioma
and microglia that promotes tumor invasiveness through the CCL2/CCR2/
interleukin-6 axis. Carcinogenesis 2012; 33: 312–319.
6 Schulz I, Engel C, Niestroj A, Zeitschel U, Menge K, Kehlen A et al. Heteroarylk-
etones inhibit astroglial interleukin-6 expression via a STAT3/NF-kappaB signaling
pathway. J Neuroinflammation 2011; 8: 86.
7 Spooren A, Mestdagh P, Rondou P, Kolmus K, Haegeman G, Gerlo S. IL-1β potently
stabilizes IL-6 mRNA in human astrocytes. Biochem Pharmacol 2011; 81:
1004–1015.
26 Cheng C, Kuo C, Lin C, Hsieh H, Yang C. IL-1β induces expression of matrix
metalloproteinase-9 and cell migration via a c-Src-dependent, growth factor
receptor transactivation in A549 cells. Br J Pharmacol 2010; 160: 1595–1610.
27 Munoz L, Selig R, Yeung Y, Peifer C, Laufer SA. Fluorescence polarisation binding
assay to develop inhibitors of inactive p38a mitogen-activated protein kinase.
Anal Biochem 2010; 401: 125–133.
8 Yeung Y, Bryce N, Adams S, Braidy N, Konayagi M, McDonald K et al. p38 MAPK
inhibitors attenuate pro-inflammatory cytokine production and invasiveness of
human U251 glioblastoma cells. J Neurooncol 2012; 109: 35–44.
9 Nabors LB, Gillespie GY, Harkins L, King PH. HuR a RNA stability factor, is expressed
in malignant brain tumors and binds to adenine- and uridine-rich elements within
the 3-untranslated regions of cytokine and angiogenic factor mRNAs. Cancer Res
2001; 61: 2154–2161.
10 Meisner N-C, Hintersteiner M, Mueller K, Bauer R, Seifert J-M, Naegeli H-U et al.
Identification and mechanistic characterization of low-molecular-weight
inhibitors for HuR. Nat Chem Biol 2007; 3: 508–516.
11 Doller A, Pfeilschifter J, Eberhardt W. Signalling pathways regulating nucleo-
cytoplasmic shuttling of the mRNA-binding protein HuR. Cell Signal 2008; 20:
2165–2173.
12 Lin W-N, Lin C-C, Cheng H-Y, Yang C-M. Regulation of cyclooxygenase-2 and
cytosolic phospholipase A2 gene expression by lipopolysaccharide through the
RNA-binding protein HuR: involvement of NADPH oxidase, reactive oxygen
28 Mourey RJ, Burnette BL, Brustkern SJ, Daniels JS, Hirsch JL, Hood WF et al.
A
Benzothiophene inhibitor of mitogen-activated protein kinase-activated
protein kinase 2 inhibits tumor necrosis factor α production and has oral anti-
inflammatory efficacy in acute and chronic models of inflammation. J Pharmacol
Exp Ther 2010; 333: 797–807.
29 Gurgis FMS, Ziaziaris W, Munoz L. Mitogen-activated protein kinase–activated
protein kinase 2 in neuroinflammation, heat shock protein 27 phosphorylation,
and cell cycle: role and targeting. Mol Pharmacol 2014; 85: 345–356.
30 Kim HH, Abdelmohsen K, Lal A, Pullmann R, Yang X, Galban S et al. Nuclear
HuR accumulation through phosphorylation by Cdk1. Genes Dev 2008; 22:
1804–1815.
31 Li L, Dutra A, Pak E, Labrie JE, Gerstein RM, Pandolfi PP et al. EGFRvIII expression
and PTEN loss synergistically induce chromosomal instability and glial tumors.
Neuro Oncol 2009; 11: 9–21.
32 Turner JG, Dawson J, Sullivan DM. Nuclear export of proteins and drug resistance
in cancer. Biochem Pharmacol 2012; 83: 1021–1032.
species and mitogen-activated protein kinases. Br
1691–1706.
13 Fernau NS, Fugmann D, Leyendecker M, Reimann K, Grether-Beck S, Galban S
et al. Role of HuR and p38 MAPK in ultraviolet B-induced post-transcriptional
regulation of COX-2 expression in the human keratinocyte cell line HaCaT. J Biol
Chem 2010; 285: 3896–3904.
14 Wang S, Zhang J, Zhang Y, Kern S, Danner RL. Nitric oxide-p38 MAPK signaling
stabilizes mRNA through AU-rich element-dependent and -independent
mechanisms. J Leukocyte Biol 2008; 83: 982–990.
J Pharmacol 2011; 163:
33 Liao W, Wang W, Chang W, Tseng JT. The RNA-binding protein HuR stabilizes
cytosolic phospholipase A2α mRNA under interleukin-1β treatment in non-small
cell lung cancer A549 cells. J Biol Chem 2011; 286: 35499–35508.
34 Tran H, Maurer F, Nagamine Y. Stabilization of urokinase and urokinase receptor
mRNAs by HuR is linked to its cytoplasmic accumulation induced by activated
mitogen-activated protein kinase-activated protein kinase 2. Mol Cell Biol 2003;
23: 7177–7188.
35 Ohgaki H, Kleihues P. The definition of primary and secondary glioblastoma. Clin
Cancer Res 2013; 19: 764–772.
15 Ramgolam VS, DeGregorio SD, Rao GK, Collinge M, Subaran SS, Markovic-Plese S
et al. T Cell LFA-1 engagement induces HuR-dependent cytokine mRNA stabili-
zation through a Vav-1, Rac1/2, p38MAPK and MKK3 signaling cascade. PLoS ONE
2010; 5: e14450.
16 Winzen R, Kracht M, Ritter B, Wilhelm A, Chen CY, Shyu AB et al. The p38 MAP
kinase pathway signals for cytokine-induced mRNA stabilization via MAP kinase-
activated protein kinase 2 and an AU-rich region-targeted mechanism. EMBO J
1999; 18: 4969–4980.
36 Munoz L, Ralay Ranaivo H, Roy SM, Hu W, Craft JM, McNamara LK et al.
A novel p38alpha MAPK inhibitor suppresses brain proinflammatory cytokine
up-regulation and attenuates synaptic dysfunction and behavioral deficits in an
Alzheimer's disease mouse model. J Neuroinflammation 2007; 4: 21.
37 Munoz L, Ramsay E, Manetsch M, Ge Q, Peifer C, Laufer SA et al. Novel p38 MAPK
inhibitor ML3403 has potent anti-inflammatory activity in airway smooth muscle.
Eur J Pharmacol 2010; 635: 212–218.
38 Guillemin GJ, Smythe G, Takikawa O, Brew BJ. Expression of indoleamine
2,3-dioxygenase and production of quinolinic acid by human microglia, astro-
cytes, and neurons. Glia 2005; 49: 15–23.
17 Neininger A, Kontoyiannis D, Kotlyarov A, Winzen R, Eckert R, Volk HD et al. MK2
targets AU-rich elements and regulates biosynthesis of tumor necrosis factor and
Supplementary Information accompanies this paper on the Oncogene website (http://www.nature.com/onc)
© 2014 Macmillan Publishers Limited
Oncogene (2014), 1 – 9