Content uploaded by Valeria R Fantin
Author content
All content in this area was uploaded by Valeria R Fantin on Feb 09, 2016
Content may be subject to copyright.
Available via license: CC BY 4.0
Content may be subject to copyright.
ARTICLE
Received 19 Jun 2014 |Accepted 2 Aug 2014 |Published 25 Sep 2014
Glutamine deprivation stimulates
mTOR-JNK-dependent chemokine secretion
Naval P. Shanware1,*, Kevin Bray1,*, Christina H. Eng1, Fang Wang1, Maximillian Follettie1, Jeremy Myers1,
Valeria R. Fantin2& Robert T. Abraham2
The non-essential amino acid, glutamine, exerts pleiotropic effects on cell metabolism,
signalling and stress resistance. Here we demonstrate that short-term glutamine restriction
triggers an endoplasmic reticulum (ER) stress response that leads to production of the
pro-inflammatory chemokine, interleukin-8 (IL-8). Glutamine deprivation-induced ER stress
triggers colocalization of autophagosomes, lysosomes and the Golgi into a subcellular
structure whose integrity is essential for IL-8 secretion. The stimulatory effect of glutamine
restriction on IL-8 production is attributable to depletion of tricarboxylic acid cycle inter-
mediates. The protein kinase, mTOR, is also colocalized with the lysosomal membrane
clusters induced by glutamine deprivation, and inhibition of mTORC1 activity abolishes both
endomembrane reorganization and IL-8 secretion. Activated mTORC1 elicits IL8 gene
expression via the activation of an IRE1-JNK signalling cascade. Treatment of cells with a
glutaminase inhibitor phenocopies glutamine restriction, suggesting that these results will be
relevant to the clinical development of glutamine metabolism inhibitors as anticancer agents.
DOI: 10.1038/ncomms5900 OPEN
1Oncology Research Unit, Pfizer Worldwide Research and Development, 401 N. Middletown Road, Pearl River, New York 10965, USA. 2Oncology Research
Unit, Pfizer Worldwide Research and Development, 10777 Science Center Drive, La Jolla, California 92121, USA. * These authors contributed equally to this
work. Correspondence and requests for materials should be addressed to R.T.A. (email: Robert.Abraham@pfizer.com).
NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications 1
&2014 Macmillan Publishers Limited. All rights reserved.
Reprogramming of molecular and metabolic pathways
involved in intermediate metabolism is now recognized as
a hallmark of cancer1. Oncogenic signals drive constitutive
cell growth and proliferation, and place heavy demands on the
pathways responsible for providing the metabolic building blocks
needed for the synthesis of proteins, nucleic acids, lipids and
other macromolecules. To meet the increased demand for
biosynthetic precursors, cancer cells increase uptake of glucose
and other nutrients, and shift overall metabolism from bioenergy
(ATP) production and cell maintenance activities to anabolic
processes that support cell mass accumulation and mitotic cell
division2,3.
The shift toward anabolic metabolism is exemplified by the
altered catabolism of glucose in tumour tissues4. Normal, non-
proliferating cells primarily convert glucose to pyruvate via
glycolysis. Pyruvate is then imported into the mitochondria,
where it is converted into acetyl CoA for entry into the
tricarboxylic acid (TCA) cycle. The glucose-derived carbon is
then completely oxidized to produce carbon dioxide and ATP. In
contrast, tumour cells reduce pyruvate to lactate for export from
the cells. The glycolytic breakdown of glucose to lactate in
oxygenated tumour tissues is termed the Warburg effect5.In
addition to lactate, glycolysis generates intermediates that fuel
anabolic metabolism via the pentose-phosphate and serine
biosynthesis pathways4. Similarly, the TCA cycle is involved in
both energy production and in the generation of building blocks
for protein and lipid biosynthesis. The diversion of glucose-
derived carbon away from the mitochondria, together with the
withdrawal of TCA cycle intermediates for biosynthetic reactions,
creates a carbon deficit in the TCA cycle that must be corrected
by entry of carbon from other sources, a process termed
anaplerosis6. These and other alterations in nutrient uptake and
utilization in transformed cells have spawned considerable
interest in cancer metabolism as a promising area for the
discovery of novel antitumour agents7,8.
The non-essential amino acid, glutamine, is a major con-
tributor to anaplerotic replenishment of the TCA cycle, and
serves as a source of carbon and nitrogen for the synthesis
of proteins, lipids and amino acids9,10. Proliferating cells
avidly import extracellular glutamine, and catabolize it via
glutaminolysis, during which glutamine undergoes sequential
deamination in the mitochondria to glutamate and further into
the TCA cycle intermediate, a-ketoglutarate (a-KG)11.Asa
nitrogen donor, glutamine supports both nucleotide and non-
essential amino acid synthesis, in addition to protein
glycosylation through the hexosamine pathway12. Finally,
glutamine plays a key role in oxidative stress resistance by
serving as a source of glutamate for the production of
glutathione9. Many cancer cells exhibit strikingly increased rates
of glutamine uptake and metabolism. Notably, cells transformed
by the MYC proto-oncogene or oncogenic KRAS display
glutamine auxotrophy13–15. The increased sensitivity of certain
transformed cells to glutamine restriction suggests that drugs
interfering with glutamine catabolism might have clinically
exploitable antitumour activities16,17. An actionable target for
such inhibitors is the mitochondrial enzyme, glutaminase, which
catalyses the conversion of glutamine to glutamate. Clearly, our
understanding of the potential benefits and challenges of
therapeutic targeting of glutamine metabolism in cancer
patients will benefit from a more complete understanding of
the cellular responses to manipulations that deprive cancer cells
of glutamine or interfere with glutaminolysis.
We and others have recently described an unanticipated
contribution of glutaminolysis to autophagy, a cytoplasmic
pathway that delivers autophagosome-encapsulated macromole-
cules and organelles to lysosomes for degradation and recycling
into metabolic processes18–21. Cells normally exhibit a basal level
of autophagic flux that is strongly enhanced by certain
environmental stresses, such as nutrient starvation. Under such
stressful conditions, autophagy allows cells to degrade non-
essential macromolecules into products that support cellular
bioenergetics and viability22. A recent manuscript by Narita
et al.23 also identified an essential role for autophagy in the
context of a multi-component endomembrane structure termed
the TOR-autophagy spatial coupling complex (TASCC). In cells
undergoing oncogene-induced senescence (OIS), the TASCC is
formed by the spatial colocalization of the autophagy machinery
with lysosomes, and appears to facilitate the mass synthesis of
secretory proteins that comprise the senescence-associated
secretory phenotype (SASP).
In this report, we demonstrate that short-term glutamine
restriction results in a chemokine-secretory response that is
dependent on the induction of the ER stress-response pathway.
Glutamine restriction-induced ER stress triggers the reorganiza-
tion of subcellular organelles into a spatially localized, cytoplas-
mic compartment that supports a robust autophagic response,
together with activation of the mechanistic target of rapamycin
complex 1 (mTORC1). These events, in turn, drive the expression
and secretion of the pro-inflammatory cytokine, IL-8. Treatment
of the cells with a glutaminase inhibitor phenocopies these
responses, suggesting that the acquisition of a secretory
phenotype may have implications for therapeutic strategies aimed
at interfering with tumour-associated glutamine metabolism in
cancer patients.
Results
Glutamine deprivation induces the ER stress response. In initial
studies, we profiled the transcriptional changes induced by short-
term (24 h) glutamine deprivation in U2OS osteosarcoma cells.
Cell growth was partially reduced, with no appreciable differences
in cell viability induced by this glutamine restriction protocol
(Supplementary Fig. 1). Microarray analysis of gene expression in
glutamine-restricted U2OS cells revealed transcriptional changes
associated with increased endoplasmic reticulum (ER) stress,
lysosomal activity, autophagy and increased expression of several
cytokines that have been linked to the SASP (Table 1).
The ER stress-response pathway is triggered by the accumula-
tion of unfolded or misfolded proteins in the ER lumen, leading
to attenuation of global protein translation, accompanied by
selective increases in the transcription and translation of several
chaperone and stress-related proteins24,25. We observed robust
induction of canonical ER stress-responsive genes DDIT3 (also
known as CHOP or GADD153) and PP1R15A (GADD34) during
glutamine deprivation (Table 1). To confirm these results, we
performed quantitative reverse transcription PCR (RT–PCR)
analyses of mRNA derived from U2OS cells that were either
glutamine deprived or treated with thapsigargin (TPG), an
established ER stress inducer (Fig. 1a). Increased CHOP and
GADD34 levels were observed in both glutamine-deprived and
TPG-treated cells. Thus, glutamine restriction stimulated a gene
expression programme that overlapped with that triggered by the
canonical ER stress inducer, TPG. Immunoblot analyses revealed
that glutamine-deprived cells exhibited increased eukaryotic
translation initiation factor 2a(eIF2a) phosphorylation and
CHOP protein expression, two hallmarks of the ER stress
response25 (Fig. 1b).
Microarray analysis also revealed elevated expression of several
autophagy-related genes, including LC3 (MAP1LC3B), WIPI1
and UVRAG, and increased expression of several subunits of the
vacuolar ATPase (v-ATPase), a protein complex required for
lysosomal acidification, lysosome-dependent protein degradation
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900
2NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications
&2014 Macmillan Publishers Limited. All rights reserved.
and mTORC1 activation (Table 1)26. Notably, we found increased
expression of subunit V
1
C
1
(ATP6V1C1), a rate-limiting
component of the v-ATPase enzyme complex27. RT–PCR
confirmed the transcriptional induction of the autophagy-
related gene LC3 suggesting that glutamine restriction
upregulated autophagy in these cells (Fig. 1a).
Unexpectedly, glutamine deprivation stimulated the expression
of genes encoding cytokines previously associated with the SASP
response (Table 1)28. In repeat experiments in multiple cell lines,
we observed that the increased interleukin-8 (IL-8; also known as
CXCL8) mRNA was the most consistent SASP-related response
to glutamine deprivation. Therefore, we focused our subsequent
characterization efforts on IL8 (Table 1). Glutamine restriction-
induced IL8 transcription was confirmed by RT–PCR (Fig. 1a).
To profile the effects of glutamine restriction on cytokine
secretion, we analysed conditioned medium from glutamine-
deprived U2OS cells, as well as A549 lung cancer cells. Increased
levels of IL-8 secretion were detected in the conditioned medium
from both cell lines (Fig. 1c,d).
Glutamine deprivation induces autophagy. Microarray analysis
revealed increased expression of lysosomal and autophagy-related
genes following 24 h glutamine withdrawal (Table 1). To test
whether these transcriptional changes were consistent with the
induction of an autophagic response, autophagosomes were
visualized by the appearance of green fluorescent protein
(GFP)-labelled autophagosomes in U2OS cells stably expressing
GFP-tagged LC3. Cells treated with complete, serum-containing
medium lacking only glutamine exhibited a clear reduction in the
number of GFP-LC3-positive puncta at 8 and 24 h (Fig. 2a,b;
Supplementary Fig. 2A,B). Similarly, lower numbers of fluor-
escent autophagosomes were observed in glutamine-deprived
cells treated with the mTORC1 inhibitor, CCI-779 (ref. 29)
(Fig. 2a,b). A reduction in the number of steady-state
autophagosomes during glutamine deprivation could reflect
either a reduced rate of autophagosome formation, or an
accelerated rate of autophagosome turnover due to an increase
in the number of fusion events with lysosomes. To distinguish
between these possibilities, cells were subjected to glutamine
deprivation in the presence of bafilomycin A1 (BafA1), a vacuolar
HþATPase (v-ATPase) inhibitor that indirectly interferes with
autophagosome–lysosome fusion and blocks autophagosome
turnover30. BafA1 provoked a dramatic increase in GFP-LC3-
positive puncta in glutamine-deprived cells, consistent with
increased autophagic flux (Fig. 2a,b).
To examine whether increased autophagosome turnover was
accompanied by a concomitant increase in the degradation of
autophagic cargo, we monitored the levels of a well-established
autophagy substrate protein, p62 (refs 30,31). Glutamine
starvation resulted in a substantial, time-dependent reduction in
p62 levels (Fig. 2c), in spite of the increase in p62-encoding
mRNA transcripts under these conditions (Table 1). The
reduction in p62 levels in the glutamine-deprived cells was
reversed in the presence of BafA1, indicating that p62 reduction
was attributable to increased autophagic flux (Fig. 2c). In
addition, BafA1 treatment was required to visualize the mature,
lipidated LC3-II band in glutamine-deprived cells, indicating that
glutamine restriction increased the turnover of LC3-II in
autolysosomes (Fig. 2c). This conclusion was further substan-
tiated by studies comparing the effects of glutamine restriction on
p62 in autophagy-competent Atg5 þ/þimmortalized baby
mouse kidney (iBMK) cells with that seen in autophagy-defective
Atg5 /iBMK cells32,33. Glutamine withdrawal resulted in a
reduction in p62 in Atg5 þ/þcells, but not in their Atg5 /
counterparts (Supplementary Fig. 2C). Importantly, BafA1
reversed the reduction in p62 only in the autophagy-competent
Atg5 þ/þcells. Finally, we used the mCherry-GFP-LC3 dual-
reporter assay30 that allows discrimination of autophagosomes
from autolysosomes. In this assay, autophagosomes are visualized
as GFP- and mCherry-positive (yellow) puncta, whereas auto-
lysosomes exhibit only the mCherry (red) signal due to
quenching of GFP fluorescence in the acidic lysosomal
compartment. Glutamine deprivation led to a decrease in the
number of yellow vesicles (autophagosomes) with no appreciable
reduction in the number of red vesicles (autophagosomes þ
autolysosomes), consistent with enhanced autophagosome to
autolysosome turnover (Fig. 2d,e). We also tested the effect of
glutamine deprivation on phagophore formation. ULK1 and
ATG5 are involved in the early steps of autophagy, and are
physically associated with the phagophore but not retained on the
mature autophagosome34. The expression of fluorochrome-
tagged versions of these two proteins in cells therefore allows
monitoring of microscopically visible phagophores and immature
autophagosomes. Indeed, glutamine withdrawal in U2OS cells
expressing mCherry-ULK1 or mCherry-ATG5 fusion proteins
stimulated the appearance of mCherry-positive puncta (Fig. 2f,g).
In contrast, exposure to BafA1 did not stimulate the appearance
of these puncta, even though it led to autophagosome
accumulation (Supplementary Fig. 2D,E). The increases in
Table 1 | Gene expression changes following glutamine
deprivation.
Gene name Fold change Pvalue
Autophagy related
LC3 (MAP1LC3B) 4.8 0.0000
P62 (SQSTM1) 2.2 0.0001
WIPI1 3.7 0.0003
UVRAG 2.2 0.0043
Lysosome related
ATP6V1C1 1.4 0.0026
ATP6V0A2 4.0 0.0004
ATP6V1D 1.6 0.0043
ER stress related
DDIT3 (CHOP) 17.2 0.0022
XBP1 1.3 0.0941
PPP1R15A (GADD34) 21.5 0.0000
ATF6 1.5 0.0405
CCND1 (cyclin D1) 5.7 0.0024
SA-secretory proteome
IL-6 1.9 0.0076
IL-8 8.7 0.0325
CCL20 (MIP-3A) 5.9 0.0087
CCL3 (MIP-1A) 2.8 0.0383
Amphiregulin (Areg) 1.6 0.0475
Epiregulin (Ereg) 6.0 0.0082
bFGF (FGF2) 2.8 0.0022
KGF (FGF7) 4.7 0.0087
VEGF 2.5 0.0029
NGFB 7.0 0.0108
MMP1 3.3 0.0148
MMP3 2.3 0.0013
PAl-1 (SERPINE1) 2.6 0.0179
EGFR 5.1 0.0122
IL, interleukin.
U2OS cells were cultured for 24h in the presence or absence of glutamine, and gene microarray
analysis was performed. Significantly altered genes were grouped into the pathways specified in
the table. Numbers are derived from three replicate samples for each condition. The column
labelled fold change indicates the averaged ratio of expression in glutamine-restricted versus the
glutamine-replete cultures. The statistical significance (Pvalue) was determined by a two-tailed,
paired Student’s t-test.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900 ARTICLE
NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications 3
&2014 Macmillan Publishers Limited. All rights reserved.
ULK1- and ATG5-positive puncta indicated that glutamine
deprivation induced autophagy early in the autophagic process.
Collectively, these data provide strong evidence that glutamine
deprivation stimulated autophagosome formation, turnover and
cargo degradation.
Colocalization of the autophagy and secretory machinery.
In light of the endomembrane reorganization, and the
accompanying increases in autophagy and IL-8 secretion induced
by glutamine restriction, we hypothesized that these endomem-
brane clusters might be similar to the TASCC structure described
by Narita et al.23 We first determined whether glutamine
deprivation resulted in a change in lysosomal localization.
Indeed, U2OS cells showed a dramatic shift in the lysosomal
staining pattern after glutamine restriction, from diffuse to a
focal, perinuclear clustering (Supplementary Fig. 3A). We next
examined whether these lysosomal clusters overlapped with the
autophagosomal clusters noted in Fig. 2. mCherry-ULK1- and
ATG5-positive puncta were extensively colocalized with
lysosomes in both cell lines following glutamine restriction
(Fig. 3a). Colocalization of ATG5- and LAMP1-positive
endomembranes was also seen in H4 (neuroblastoma) and
HCT116 cells (Supplementary Fig. 3B).
To determine whether the endomembrane reorganization
response was unique to glutamine restriction, we examined
lysosome distribution after withdrawal of the essential amino
acid, leucine. Lysosomal clustering was evident in the leucine-
starved cells, but considerably less universal on a per cell basis
than removal of glutamine (Supplementary Fig. 3C,D). Leucine
deprivation was correspondingly less effective than glutamine
restriction with regard to autophagy induction, as measured by its
effects on p62 expression (Supplementary Fig. 3E). In complete
contrast to glutamine withdrawal, leucine deprivation failed to
induce IL-8 gene expression, despite robust induction of the ER
stress marker, CHOP (Supplementary Fig. 3F). These results
suggest that clustering of the autophagic machinery with
lysosomes was causally related to the increase in IL-8 production
observed in glutamine-restricted cells.
We had previously observed fewer GFP-LC3-positive auto-
phagosomes in glutamine-deprived U2OS cells (see Fig. 2). We
speculated that the spatial approximation of autophagic vesicles
with lysosomes accelerated autophagic trafficking to lysosomes, a
scenario that could explain the reduction in mature autophago-
somes observed after glutamine restriction. To explore this
possibility, GFP-LC3-expressing U2OS cells were subjected to
glutamine deprivation in the presence of BafA1 to visualize
autophagosomes that accumulated due to impaired lysosomal
fusion and degradative activity. BafA1 treatment caused the
appearance of GFP-LC3-positive puncta that were colocalized
with LAMP1-positive lysosomal clusters in glutamine-deprived
cells (Fig. 3b). BafA1 treatment also increased the colocalization
of p62 with lysosomes in these cells (Fig. 3c). These findings
support the hypothesis that glutamine deprivation stimulates
pEIF2α
Total EIF2α
CHOP
LC3
pS6K
Actin
+Q –Q
IL8
U2OS A549
+Q –Q
0.000
0.050
0.100
0.150
0.200
0.250
0.300
(+Q) (–Q) (–Q)
Arbitrary OD units
U2OS
0.000
0.200
0.400
0.600
0.800
1.000
(+Q)
Arbitrary OD units
A549
40 kDa
60 kDa
15 kDa
30 kDa
40 kDa
40 kDa
+Q –Q TPG
35
30
25
20
15
10
5
0
350
300
250
200
150
100
50
0
25
20
15
10
5
0
Gln (+) Gln (–) TPG Gln (+) Gln (–) TPG
Gln (+) Gln (–) TPG
Gln (+) Gln (–) TPG
35
30
25
20
15
10
5
0
Fold changeFold change
Fold change
Fold change
LC3 IL8
GADD34CHOP
Figure 1 | Glutamine deprivation induces IL-8 secretion. (a) RT–PCR analysis of LC3, IL-8, CHOP and GADD34 in U2OS cells subjected to glutamine
deprivation ( Q) or exposed to TPG (1 mM). (b) Immunoblot analysis of ER stress- and autophagy-related proteins in U2OS cells subjected to glutamine
deprivation or TPG treatment. Samples are run in triplicate. (c) Cytokine array analysis of conditioned media from U2OS and A549 cells grown in the
presence ( þQ) or absence ( Q) of glutamine. (d) IL-8 enzyme-linked immunosorbent assay of conditioned media from U2OS and A549 cells grown in
the presence ( þQ) or absence ( Q) of glutamine. Error bars in all figures represent s.d. of three biological replicates.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900
4NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications
&2014 Macmillan Publishers Limited. All rights reserved.
autophagic flux, in part through colocalization of the autophagy
induction and maturation apparatus with lysosomes.
The induction of SASP-related genes together with the
formation of a TASCC-like subcellular structure23 prompted us
to examine the potential involvement of the Golgi apparatus in
response to glutamine restriction. The Golgi apparatus plays key
roles in the sorting of newly translated proteins into various
intracellular compartments and the secretory pathway, and thus
may play an active role in glutamine deprivation-induced IL-8
secretion. Using RCAS1 and syntaxin 6 staining to visualize the
Golgi apparatus, we observed that glutamine deprivation
triggered Golgi–lysosome co-clustering (Fig. 3d; Supplementary
Fig. 4A). Similar results were obtained in H4 and HCT116 cells
(Supplementary Fig. 4B).
mTORC1 regulates endomembrane reorganization. The
rapamycin-sensitive mTORC1 kinase complex is generally con-
sidered to be a negative regulator of autophagy. In nutrient-
replete cells, active mTORC1 suppresses early events in the
autophagy pathway35,36. Conversely, nutrient starvation or
treatment with mTOR inhibitors suppresses mTORC1 activity,
leading to the de-repression of autophagy. To determine whether
the autophagic response to glutamine restriction was attributable
to attenuated mTORC1 activity, we examined phosphorylation of
S6 kinase 1 (S6K1), a known mTORC1 substrate (Fig. 3e). In
response to glutamine deprivation, U2OS cells reduced p62 levels,
indicating increased autophagy (Fig. 2c). In contrast, S6K1
phosphorylation was unchanged in response to glutamine
deprivation (Fig. 3e). Cells cultured in medium containing
either no amino acids or only essential amino acids displayed
nearly complete dephosphorylation of S6K1. These results
indicate that, unlike more global amino acid starvation, removal
of glutamine alone from the culture medium did not lead to a
reduction
in mTORC1 activity. Thus, glutamine deprivation-induced
autophagy, unlike global amino-acid deprivation, triggers an
autophagic response in the absence of reduced mTORC1 activity.
We next tested whether mTOR colocalized with the lysosomal
cluster provoked by glutamine deprivation. Under nutrient-
+Q –Q
U2OS mCherry-GFP-LC3 cells
60 kDa
40 kDa
15 kDa
–
–Q
–Q
–Q
0248240 2 4824
–Q + BAFA1
LC3-I
Hours
LC3-II
p62
Actin
(–) Q
*
*
*
Number of
yellow vesicles per cell
(autophagosomes)
Percentage of cells with
>10 puncta per cell
U20S mCherry -ULK1
U20S mCherry -ATG5
Percentage of cells with
>10 puncta per cell
Percentage of cells with
>10 puncta per cell
Number of red vesicles
per cell (autophagosomes
+autolysosomes)
+Q
–Q +Q
+Q
(+) Q (–) Q(+) Q
(–) Q (–) Q
(+) Q (+) Q
00
100
90
80
70
60
50
40
30
20
10
0
BafA1
CC1
Glutamine + +
+
+
++
+
–
–
–
–
––
–
–
–
–
–
70
60
50
40
30
20
10
0
70
80
90
60
50
40
30
20
10
0
55
10
10
15
15
20
20
25
25
35
30
CCI BafA1
Figure 2 | Glutamine deprivation induces autophagic flux. (a) Autophagosomes were visualized in U2OS cells stably expressing a GFP-LC3 reporter
construct. GFP-labelled puncta were examined after 24 h in the presence ( þQ) or absence ( Q) of glutamine, with or without co-addition of 100nM
CCI-779 (CCI) and 400nM BafA1. Scale bar, 10mm. (b) Graphical summary of experiments performed as described in a. Percentage of cells with 410
puncta per cell from three independent experiments is depicted. Bars represent mean±s.d. from three independent experiments (450 cells per
experiment). The statistical significance (Pvalue) was determined by a two-tailed, paired Student’s t-test. *Po0.05. (c) Immunoblot analysis of U2OS cells
subjected to glutamine deprivation with or without 400 nM BafA1. Cells were pretreated with BafA1 for 1 h before and during exposure to glutamine-
deficient medium. Autophagic activity was monitored by detection of p62 and LC3-II proteins. (d) U2OS mCherry-GFP-LC3 cells were cultured in the
presence ( þQ) or absence ( Q) of glutamine for 18 h. Red vesicles denote autolysosomes, whereas yellow vesicles represent autophagosomes.
Bars indicate numbers of yellow vesicles (autophagosomes) or red vesicles (autolysosomes) per cell±s.d. (e) Images of U2OS mCherry-GFP-LC3 cells
cultured for 18 h in the presence ( þQ) or absence ( Q) of glutamine. Scale bar, 10 mm. (f) Phagophore formation in mCherry-ULK1. Scale bar, 10 mm.
(g) mCherry-ATG5-expressing U2OS cells after 24 h in the presence ( þQ) or absence ( Q) of glutamine. Scale bar, 10 mm.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900 ARTICLE
NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications 5
&2014 Macmillan Publishers Limited. All rights reserved.
replete conditions, mTORC1 localizes to the lysosomal surface,
where its activity is coupled to the extrusion of amino acids
from the lysosomal lumen37,38. Immunofluorescence staining
revealed mTOR colocalization with the lysosome cluster in
glutamine-restricted cells, consistent with the retention of
mTORC1 activity in these cells (Fig. 3f). In cells deprived of all
amino acids, mTORC1 activity was completely suppressed and
mTORC1-lysosome colocalization was abolished, in agreement
with published reports (Fig. 3e,f)37. Complete amino-acid
deprivation also led to lysosomal dispersion throughout the
cytoplasm (Fig. 3f). Treatment of glutamine-starved cells with
WYE-125132, a highly selective inhibitor of mTOR kinase
activity39, blocked the formation of lysosomal clusters and
elicited a dispersed pattern of LAMP1 staining similar to that
observed on amino-acid starvation (Fig. 3f). However, a
substantial level of mTOR-lysosome colocalization remained in
the mTOR inhibitor-treated cells (Fig. 3f). These results suggested
that lysosomal localization of mTORC1 was necessary but not
sufficient to drive lysosomal clustering: mTORC1-dependent
signalling was also required for this response. WYE-125132
exposure or amino-acid starvation also blocked phagophore
clustering (Supplementary Fig. 5), indicating that the
formation of these glutamine deprivation-induced endo-
membrane structures was dependent on mTOR kinase activity.
These results suggested that glutamine deprivation leads to a
cytoplasmic endomembrane reorganization response with
parallels to the TASCC complex seen in cells undergoing OIS.
IL-8 secretion is autophagy independent. The TASCC was
shown to couple autophagic activity to IL-8 mRNA translation
and IL-8 secretion23. We therefore sought to test whether
a similar relationship between autophagy and chemokine
production existed in glutamine-restricted cancer cells. U2OS
cells were transfected with transcription activator-like effector
nucleases (TALENs) targeting the essential autophagy gene
ATG7. Clonal U2OS cell lines transfected with the ATG7-
TALEN (U2OS-TAL2-A7 or U2OS-TAL2-A8) expressed no
detectable ATG7 protein (Fig. 4a). ATG7 is an E1-like
enzyme that is required for autophagy due to its roles in the
conjugation of ATG12 to ATG5, and of ATG8 (LC3) to
phosphatidylethanolamine40. Relative to U2OS cells transfected
with the control TALEN (U2OS-TAL2-C1), the ATG7-TALEN
clones expressed no detectable ATG5/12 conjugates, and
displayed a marked accumulation of p62 in complete or amino-
acid-deficient medium, consistent with defective basal and
inducible autophagy (Fig. 4a). As predicted, the absence of
autophagy in the U2OS-TAL2 clones led to reduced survival in
response to long-term (6 days) culture in glutamine-free medium
(Fig. 4b,c).
The U2OS-TAL2 clones were then subjected to the standard
24-h glutamine deprivation protocol, followed by determinations
of IL-8 secretion. Contrary to our expectation, these autophagy-
defective cells were fully competent to express and secrete IL-8 in
response to glutamine restriction (Fig. 4d). Similar results were
observed in A549 cells transfected with the ATG7-targeted
–– BafA1BafA1
–Q
BafA1BafA1 ––
+Q
–Q
+Q
p62LAMP1Merge
Merge LAMP1 GFP -LC3
H
LAMP1 RCAS1 Merge
+Q
–Q
+Q
LAMP1mTORMerge
–Q +WYE-132–A.A–Q
p62
pS6K
Actin
S6K
60 kDa
–Ess. amino acids
–Q
+All amino acids
–All amino acids
60 kDa
60 kDa
40 kDa
mCherry
U2OS
mCherry-ULK1
U2OS
mCherry-ATG5
Lysotracker green Merge
Figure 3 | Glutamine deprivation induces endomembrane clustering and colocalization of mTORC1. (a) Colocalization of lysosomes (Lysotracker Green)
with phagophores (mCherry-ULK1 and mCherry-ATG5 puncta) in glutamine-deprived U2OS cells. Scale bar, 25 mm. (b) Colocalization of LAMP1-positive
lysosomes with GFP-LC3-labelled autophagosomes after glutamine deprivation in the absence or presence of 400 nM BafA1 in U2OS cells. Scale bar,
15 mm. (c) Immunofluorescence images of U2OS cells stained for p62 expression and LAMP1 following glutamine deprivation and BafA1 treatment. Nuclei
were stained with DAPI. Scale bar, 15 mm. (d) Immunofluorescence images of cells stained with LAMP1 (lysosomes) and the Golgi marker, RCAS1, following
glutamine deprivation. Nuclei were stained with DAPI. Scale bar, 15 mm. (e) U2OS cells were treated for 24 h with cell culture medium containing the
indicated amino acids. Cell lysates were immunoblotted to detect changes in S6K phosphorylation and p62. (f) Immunofluorescence staining for LAMP1
(green) and mTOR (red) in U2OS cells subjected to glutamine deprivation (24 h), in the presence of 100 nM WYE-125132 (WYE-132). AA denotes cells
deprived of all amino acids. Scale bar, 15 mm. Error bars in all figures represent s.d. of three biological replicates. Ess., essential.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900
6NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications
&2014 Macmillan Publishers Limited. All rights reserved.
TALEN (data not shown). In addition, IL-8 production was not
affected by treatment of glutamine-deprived U2OS cells with
BafA1, a pharmacological inhibitor of autophagy (Fig. 4e). Taken
together, these data indicate that, although autophagy is required
for optimal cell survival during extended glutamine deprivation,
the expression and secretion of IL-8 induced by short-term
glutamine withdrawal is independent of autophagic activity.
mTORC1-JNK signalling stimulates IL-8 secretion. To better
understand the mechanisms regulating glutamine deprivation-
induced IL-8 secretion, we examined the possible role of
mTORC1 in this response by treating glutamine-deprived U2OS
cells with WYE-125132. WYE-125132 blocked the formation of
the Golgi body–autophagy–lysosome cluster described above
(Fig. 5a) and strongly reduced IL-8 secretion in the glutamine-
deprived cells (Fig. 5b,c). Disruption of the Golgi apparatus by
brefeldin-A inhibited both the colocalization of Golgi endo-
membranes with lysosomes (Fig. 5a) and IL-8 secretion during
glutamine restriction (Fig. 5b). Collectively, these data indicate
that mTOR kinase activity is required for the juxtapositioning of
the Golgi apparatus, autophagy components and mTORC1-
associated lysosomes in glutamine-restricted cells. Moreover, both
an intact Golgi apparatus and CCI-779-sensitive mTORC1
activity (Fig. 5b,c) were required for the production of IL-8 by
glutamine-restricted cells. An unanticipated observation was that
mTORC1 inhibition by WYE-125132 or CCI-779 blocked IL-8
induction at the mRNA level, suggesting that an mTORC1-
dependent mechanism was driving IL8 gene transcription in these
cells (Fig. 5d).
The c-Jun NH2-terminal Kinase (JNK) is a stress-activated
protein kinase that is activated in response to ER stress41 and is a
known upstream activator of IL-8 (ref. 42). JNK was strongly
phosphorylated following 24 h glutamine deprivation. Inhibition
of mTORC1 activity with CCI-779 partially blocked JNK
phosphorylation. Suppression of mTORC1 did not simply lead
to a global shutdown of cellular translation since CHOP protein
induction was unaffected by CCI-779 (Fig. 6a). Suppression of
JNK with small interfering RNA (siRNA) blunted IL-8
transcription and secretion (Fig. 6b–d). In addition, treatment
with a JNK kinase inhibitor strongly inhibited glutamine
deprivation-induced IL-8 transcription and secretion, providing
strong evidence for the role of JNK in glutamine deprivation-
induced IL-8 secretion (Fig. 6e,f).
mTORC1 was previously reported to play an important role in
the ER stress-induced IRE1-JNK pathway41, prompting us to
examine the role of IRE1 in glutamine deprivation-induced IL-8
secretion. Knockdown of IRE1 with siRNAs (Fig. 6j) blocked both
IL-8 transcription and secretion following glutamine deprivation
(Fig. 6g,h). In contrast, IRE1 siRNA did not affect induction of
CHOP (Fig. 6i). IRE1 siRNA also strongly inhibited glutamine
deprivation-induced JNK phosphorylation (Fig. 6k), implying
that the IRE1-JNK arm of ER stress was specifically required for
glutamine deprivation-induced IL-8 secretion.
We next tested whether the ER stress inducer TPG also
induced IL-8 secretion via activation of the same pathway. U2OS
cells were treated with TPG with or without CCI-779 or
WYE-125132. TPG induced IL-8 transcription and secretion in
an mTOR-dependent manner (Supplementary Fig. 6A–C). TPG
also induced CHOP transcription in a mTORC1-independent
0.000
–0.500
0.500
1.000
1.500
2.000
2.500
U2OS_
TAL2-ATG7
U2OS_
TAL2-CTL
Absorbance 540 nM
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
TAL2-
ATG7-A8
TAL2-
CTL-C1
Arbitrary OD units
(+) Q
(–) Q
0
0.5
1
1.5
2
2.5
(–) Q (–) Q +BafA1
Arbitrary OD units
0 h
2 h
4 h
8 h
24 h
80 kDa
U2OS TAL2 C1 U2OS TAL2A7 U2OS TAL2A8
+Q +Q–Q –Q +Q –QHBSS HBSS HBSS
ATG7
p62 short
p62 long
ATG5/12
TAL2_CTL TAL2_ATG7
ATG5
LC3
Actin
60 kDa
60 kDa
50 kDa
30 kDa
15 kDa
40 kDa
Figure 4 | Autophagy is not required for glutamine deprivation-induced IL-8 secretion. (a) Immunoblot of autophagy-related proteins in U2OS cells
expressing a control TALEN (U2OS-TAL2-C1) or two independent subcloned TALEN cell lines targeting ATG7 (U2OS-TAL2-A7, U2OS-TAL2-A8).
p62 long and short refer to either long or short exposure times during film development. (b) U2OS cells expressing an ATG7-TALEN (TAL2-ATG7) or
control TALEN (TAL2-CTL) were subjected to glutamine starvation for 6 days followed by 6 days recovery in glutamine-replete medium. Cells were
then stained with SRB and absorbance was read at 540nM. Graph shows average results from five clonal sub-lines for each condition. The statistical
significance (Pvalue) was determined by a two-tailed, paired Student’s t-test. Po0.05. (c) Representative images of SRB-stained clones in b.
(d) IL-8 enzyme-linked immunosorbent assay (ELISA) of U2OS-TAL2-CTL-C1 (control) and U2OS-TAL2-ATG7-A8 (ATG7 knockout) following 24h
glutamine deprivation. (e) IL-8 ELISA of U2OS parental cells starved of glutamine for the indicated times with or without BafA1. Error bars in all
figures represent s.d. of three biological replicates. OD, optical density.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900 ARTICLE
NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications 7
&2014 Macmillan Publishers Limited. All rights reserved.
fashion, similar to the response induced by glutamine deprivation
(Supplementary Fig. 6D). Notably, TPG-induced JNK
phosphorylation was also suppressed by mTOR inhibition
(Supplementary Fig. 6E).
IL-8 secretion is a response to deficient anaplerosis. Finally,
we tested the effects of a glutamine metabolism inhibitor on IL-8
secretion. Bis-2-(5-phenylacetamido-1,3,4-thiadiazol-2-yl)ethyl
sulfide (BPTES) is a selective inhibitor of the kidney-type
glutaminase isoform (GLS1), which catalyses the deamination of
glutamine to glutamate. BPTES strongly induced both IL-8 and
CHOP, suggesting that BPTES treatment phenocopies the effects
of glutamine deprivation on these stress responses (Fig. 7a,b).
Metabolomic analyses revealed that both glutamine deprivation
and BPTES treatment depleted the TCA cycle intermediates
a-KG, citrate, succinate and fumarate. BPTES treatment also
led to an increase in intracellular glutamine pools, suggesting
that defective glutamine metabolism, rather than depletion
of intracellular glutamine levels, was responsible for the sequence
of events leading to IL-8 induction (Fig. 7c). To test this idea, a
cell-permeable form of a-KG, the TCA cycle intermediate pro-
duced by glutamine deamination, was examined for its ability to
reverse the stress-response mechanism that drives IL-8 secretion.
Treatment with dimethyl alpha-KG (DM-aKG) successfully
restored cellular TCA cycle pools to levels comparable to those
seen in the glutamine-replete condition (Fig. 7c). Importantly,
DM-aKG treatment reduced IL-8 transcription and secretion in
both glutamine-restricted (Fig. 7b,d) and BPTES-treated cells
(Fig. 7d). Finally, glutamine deprivation-induced JNK phos-
phorylation was also blocked by the addition of DM-aKG
(Fig. 7e). Together, these results indicate that the impairment of
TCA cycle function during glutamine restriction is partially
responsible for increases in IL-8 transcription and secretion
observed in these studies.
Discussion
Glutamine has emerged as both an important fuel for anabolic
metabolism and a regulator of key cellular responses, such as
growth factor receptor expression12, mTORC1 activation43,44 and
autophagy20,21. The present findings shed further light on the
complex interplay between glutamine and cell physiology by
demonstrating that a relatively brief period (24 h) of glutamine
restriction leads to a striking reorganization of several
cytoplasmic organelles, accompanied by stimulation of
autophagy and an autophagy-independent stress response
resulting in the induction of IL-8 expression and secretion. The
IL-8 release provoked by glutamine restriction was dependent on
the activity of a mTORC1-IRE1-JNK pathway that triggered both
transcription and secretion of this chemokine. Given the
pleiotropic actions of IL-8 in the tumour microenvironment,
the present findings have significant implications for both tumour
pathophysiology and therapeutic strategies aimed at interfering
with glutamine uptake and metabolism in cancer patients45.
The persistence of mTORC1 activity in glutamine-deprived
cells was unexpected, given that amino acid availability is a well-
established regulator of mTORC1 activity, and glutamine in
particular had been causally linked to amino-acid-dependent
mTORC1 activation43,44. The specific assay conditions used in
our study provide a plausible explanation for the maintenance of
mTORC1 activity in glutamine-deprived cells. Our experimental
protocol involved a relatively brief (24 h) exposure of cells to
standard, serum-containing culture medium lacking only
glutamine. In contrast, the earlier reports implicating glutamine
in amino-acid-dependent mTORC1 activation employed a
distinct protocol involving an initial period of complete amino-
acid starvation followed by re-addition of specific amino acids
and subsequent assessment of mTORC1 activity. Cells treated
under these conditions were dependent on glutamine and leucine
for mTORC1 reactivation after amino-acid starvation43,44.
Glutamine-induced mTORC1 activation in these studies
+Q
LAMP1RCAS1Merge
–Q
WYE-132 BrefA– –Q+Q
–Q + BrefA –Q + WYE-132
0.6
140 IL8
120
100
80
60
40
20
0+
Arbitrary OD units
Fold change
0.5
0.4
0.3
0.2
0.1
0+++Q– – – –+–+–
WYE132 WYE132 CCI-779
CCI779
Figure 5 | Glutamine deprivation-induced IL-8 secretion requires mTOR activity. (a) Immunofluorescence images of U2OS cells stained for
LAMP1 and the Golgi marker RCAS1 after 24 h glutamine deprivation with or without treatment with 1 mM BrefA or 100 nM WYE-125132. Scale bar, 15 mm.
(b) Cytokine array analysis of conditioned medium from U2OS cells subjected to glutamine deprivation in the presence of 1mM BrefA or 100 nM
WYE-125132. (c) IL-8 enzyme-linked immunosorbent assay in U2OS cells after 24 h glutamine deprivation with or without CCI-779 or WYE-125132.
(d) RT–PCR analysis of IL-8 mRNA expression in U2OS cells after 24 h glutamine deprivation with or without CCI-779 or WYE-125132. Error bars
in all figures represent s.d. of three biological replicates. OD, optical density.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900
8NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications
&2014 Macmillan Publishers Limited. All rights reserved.
involved at least two mechanisms, one related to glutamine-
dependent import of the mTORC1-activating amino acid,
leucine, and the other via the production of a-KG. The distinct
glutamine restriction protocol used in our studies permitted
examination of glutamine-related phenotypic responses in the
absence of confounding inhibitory effects on mTORC1 function.
The importance of persistent mTORC1 activity in glutamine-
deprived cells was highlighted by our findings that both the
endomembrane reorganization and autophagic responses induced
by glutamine restriction were strongly suppressed by pharmaco-
logic inhibitors of mTORC1 signalling functions. The endomem-
brane reorganization induced by glutamine deprivation involved
juxtapositioning of the Golgi with phagophores, autophagosomes
and lysosomes in a single cytoplasmic cluster. Relocalization of
cytoplasmic organelles was not a global response, in that the
distribution of mitochondria in the cytoplasm was not altered by
glutamine starvation (results not shown). Immunofluoresence
microscopy revealed the presence of mTOR in the lysosomal
clusters, consistent with recent reports indicating that active
mTORC1 is associated with the lysosomal compartment37,38.
Inhibition of mTORC1 activity with the rapalogue, CCI-779 or
the mTOR kinase inhibitor, WYE-125132, stimulated autophagy,
but did not lead to endomembrane clustering. Indeed, treatment
of glutamine-deprived cells with WYE-125132 blocked the
appearance of the endomembrane cluster, indicating that the
relocalization and/or stability of this structure were dependent on
mTORC1 signalling, presumably on the lysosomes.
The spatial colocalization of the autophagy machinery
with lysosomes appeared to drive rapid, efficient fusion of
mature autophagosomes with lysosomes, as indicated by our
inability to visualize GFP-LC3-labelled autophagosomes in these
clusters in the absence of the lysosomal inhibitor, BafA1. We
propose a bidirectional interplay between mTORC1 and the
endomembrane clusters in which autophagy fuels the lysosomal
lumen with amino acids that stimulate mTORC1 activity, and
mTORC1 in turn drives events that support the assembly and/or
maintenance of these clusters. The concentration of mTORC1
complexes near the Golgi apparatus might also facilitate the
60 kDa
30 kDa
+Q +Q
60 kDa
40 kDa
46 kDa
54 kDa
46 kDa
+–+–+–
siJNK #2siJNK #1
54 kDa
40 kDa
+++–––+++– ––+++–––
46 kDa
40 kDa
46 kDa
p62
CHOP
pS6K
Actin
Phospho-JNK
Total JNK
2,000 IL8
Fold change
0.6
0.5
0.4
0.3
0.2
0.1
0
0.6
0.8
1.2
1.4
1.6
1.8
1
0.4
0.2
0
+–+–+–Q
siCTL siJNK #1 siJNK #2
+–+–Q
JNKi
1,800
1,600
1,400
1,200
1,000
800
600
400
200
0
0
20
40
60
80
100
120
140
Fold change
Arbitrary OD units
JNK
Actin
–Q–Q
CCI
+Q
IL-8 +Q +Q –Q–Q–Q
0.0
Normalized fold
expression
0.2
0.4
0.6
0.8
1.0
Normalized fold
expression
Normalized fold
expression
0.0
0.0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
0.2
0.4
0.6
0.8
1.0
siIRE-BsiIRE-A
siCTL
IL-8
+Q +Q +Q –Q–Q–Q
siIRE-A
siIRE-AsiCTL
CHOP
+Q +Q +Q –Q–Q–Q
siIRE-B
siIRE-AsiCTL
+Q +Q +Q –Q–Q–Q
siIRE#2siIRE#1siCTL
IRE1
pSAPK/
JNK
Actin
siCTL siJNK #1 siJNK #2
3.000
2.500
2.000
1.500
1.000
0.500
0.000
siCONT (+Q)
siIRE-A (+Q)
siIRE-A (–Q)
siIRE-B (+Q)
siIRE-B (–Q)
siCONT (–Q)
Arbitrary OD units
Arbitrary OD units
Gene expression
Gene expression
Gene expression
JNKi
++––Q
Figure 6 | mTORC1 regulates ER stress-induced JNK signalling. (a) Immunoblot analysis of autophagy, ER stress- and mTOR-related proteins, and JNK in
U2OS cells after 24 h glutamine deprivation with or without CCI-779. (b) RT–PCR analysis of IL-8 mRNA expression in U2OS cells transfected with two
different siRNAs targeting JNK (JNK #1 and JNK #2) and subjected to 24 h glutamine deprivation. (c) IL-8 enzyme-linked immunosorbent assay (ELISA) of
conditioned media from the experiment described in b.(d) Immunoblot analysis of JNK protein expression in JNK siRNA-treated cells. (e) RT–PCR analysis
of IL-8 expression in U2OS cells after 24 h glutamine deprivation with or without the JNK inhibitor SP600125 (JNKi). (f) IL-8 ELISA of conditioned medium
from the experiment described in e.(g) IL-8 ELISA of U2OS cells transfected with two different siRNAs targeting IRE1 (IRE-A and IRE-B) and subjected to
glutamine deprivation for 24 h. (h) RT–PCR analysis of IL-8 mRNA expression from experiment described in g.(i) RT–PCR analysis of CHOP mRNA
expression from experiment described in g.(j) RT–PCR analysis of IRE1 mRNA expression from experiment described in g.(k) Immunoblot analysis of
experiment described in gshowing phosphorylation of JNK. Error bars in all figures represent s.d. of three biological replicates. OD, optical density.
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900 ARTICLE
NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications 9
&2014 Macmillan Publishers Limited. All rights reserved.
translation of mRNAs in nearby rough ER sites. This model is
reminiscent of recent findings demonstrating that autophagy
supports mTORC1 activation during nutrient starvation and that
active mTORC1 is required for the reformation of lysosomes
from autolysosomes46. Perhaps, the lysosomal clustering observed
in glutamine-deprived cells reflects, at least in part, the localized
regeneration of lysosomes in a subcellular compartment with
intense autophagic activity.
The endomembrane redistribution elicited by glutamine with-
drawal also shares features with the previously reported TASCC
in cells undergoing OIS23. In this study, fibroblasts induced to
undergo senescence by acute expression of the H-Ras oncoprotein
displayed autophagosome–lysosome clusters and concomitant
activation of autophagy and mTORC1. How might seemingly
disparate, stress-inducing events, such as glutamine restriction
and oncoprotein expression, trigger the formation of similar
autophagy-lysosome clusters? A reasonable hypothesis is that
glutamine deprivation and HRAS expression converge on a
common stress-response network provoked by the accumulation
of unfolded proteins in the ER (Table 1; Fig. 1; ref. 23). The ER
stress response is activated in response to a broad range
of stimuli, and results in global translational suppression,
accompanied by the induced expression of a gene set
whose encoded proteins act to restore ER homeostasis and
preserve cell viability, over the short term, and to trigger
apoptotic cell death in the setting of chronic ER stress24. Our
studies indicate that the ER stress response invoked by glutamine
restriction leads to the expression of genes associated with
autophagy and the SASP.
The relationship between autophagic activity and IL-8
secretion was the most critical difference between OIS-induced
IL-8 (ref. 23) and the chemokine secretion response evoked by
glutamine deprivation in the present study. Autophagic activity
was causally related to OIS-mediated chemokine production,
whereas the IL-8 expression induced by glutamine restriction was
maintained in cells in which the autophagy pathway was
genetically disrupted. Taken together, the earlier and present
findings suggest that OIS and glutamine deprivation induce a
similar clustering of a specific subset of cytoplasmic organelles
and concomitant increases in autophagic flux. Although both
stress-induced pathways converge on the production of IL-8 and
other chemokines, the contribution of autophagy to this shared
downstream response is fundamentally different in cell subjected
to OIS versus glutamine restriction.
Previous reports suggested that mTORC1 activity was required
for induction of the ER stress-stimulated IRE1-JNK pathway, but
was not involved in the activation of the parallel PERK and ATF6
pathways triggered by ER stress41. For example, rapamycin
selectively suppressed IRE1-JNK signalling and attenuated ER
stress-induced apoptosis in rat renal tubular epithelial cell lines41.
Our results indicate that glutamine deprivation strongly activates
JNK, and that this event is dependent on the upstream activities
of mTORC1 and IRE1. Taken together, these data support a
working model, which posits that glutamine deprivation provokes
an IRE-1- and mTORC1-dependent increase in JNK activity that,
in turn, leads to IL-8 expression and secretion. The precise
mechanism by which glutamine restriction leads to chemokine
secretion is incompletely understood. However, the observations
+Q –Q
60 kDa
40 kDa
46 kDa
54 kDa
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.0
0.2
0.1
0.2
0.0
0.0
0.0
0.1
2.0
0.5
0.3
0.8
0.1
0.1
0.0
0.1
0.0
1.5
220
0.8
1.2
1.2
2.4
1.1
Glutamine/lysine 20
1.2
1.0
0.8
0.6
0.4
0.2
0.0
CHOP Target IL-8
Gene expression
+Q
–Q+ 10 mM DMαKG
–Q –Q + 1 mM DM-αKG
Normalized fold
expression
15
5
10
Glutamate
α’-Ketoglutarate
Succinate
Fumarate
Malate
Oxaloacetate
Citrate/isocitrate 0
0.3
0.25
0.2
0.15
0.1
0.05
Arbitrary OD units
BPTES
+DM-αKG)
BPTES
(–Q + DM-
αKG)
(–Q)(+Q)
+Q –Q
–Q
+DMαKG
BPT
(–Q)(+Q) BPTES
Arbitrary OD units
1.000
0.800
0.600
0.400
0.200
0.000
1.200
–Q DMαKG
pS6K
pSAPK/JNK
Actin
Figure 7 | Glutamine starvation and glutamine metabolism inhibitors induce IL-8 secretion that is reversed by a-KG supplementation. (a) IL-8 enzyme-
linked immunosorbent assay (ELISA) on U2OS cells starved of glutamine or treated with BPTES (10 mM) for 24 h. (b) RT–PCR analysis of IL-8 and CHOP
mRNA expression in U2OS cells starved of glutamine for 24 h and co-treated with DMaKG. ( c) Metabolite analysis of U2OS cells starved of glutamine,
treated with BPTES or glutamine-starved co-treated with DMaKG. ( d) IL-8 ELISA of U2OS cells starved of glutamine or treated with BPTES for 24 h and co-
treated with DMaKG. (e) Immunoblot analysis of pS6K and pJNK in U2OS cells starved of glutamine for 24 h and co-treated with DMaKG. Error bars in all
figures represent s.d. of three biological replicates. OD, optical density.
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900
10 NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications
&2014 Macmillan Publishers Limited. All rights reserved.
that glutaminase inhibition phenocopied the effects of glutamine
withdrawal, and the reversal of these responses by the glutamine
deamination product, a-KG, argue that impaired TCA cycle
activity underlies this multi-factorial stress response. One possible
signal emanating from the loss of TCA cycle intermediates may
be enhanced production of reactive oxygen species (ROS). Initial
studies indicate that glutamine deprivation, but not leucine
deprivation, results in robust ROS production, and the use of the
ROS scavenger, N-acetylcysteine, partially blunts IL-8 secretion,
suggesting a role for oxidative stress in the observed response
(data not shown). The differential induction of IL-8 during
glutamine versus leucine starvation argues that the secretory
response is not simply a general, nonspecific outcome of amino-
acid deprivation.
The present findings have potential implications for
therapeutic strategies aimed at disrupting glutamine uptake
and/or metabolism in cancer cells16. Certain cancer cells, such
as those transformed by the Myc oncoprotein, are glutamine
auxotrophs, and undergo cell death in the setting of glutamine
withdrawal or in response to glutaminolysis inhibitors13,14.
A recent report indicated that MYCN-amplified neuroblastoma
cells undergo apoptotic death due to the activation of a nutrient
stress-response pathway leading to accumulation of the pro-
apoptotic PUMA and NOXA proteins13. The GLS inhibitor
BPTES and its analogues have shown interesting preclinical
activities and are currently in early clinical development47. Hence,
our observation that BPTES triggers IL-8 secretion in cultured
cells may be clinically relevant. If a similar response was invoked
in tumour tissue during therapeutic inhibition of glutamine
metabolism, the secretion of autocrine/paracrine-acting IL-8
might have significant effects on tumour pathophysiology and
the tumour–host interface, which could lead to changes in
therapeutic responsiveness. Previous reports demonstrated
that autocrine IL-8 signalling stimulates the outgrowth of
highly tumorigenic subpopulations that exhibit increased drug
resistance and metastatic behaviour, and that paracrine IL-8
signalling promotes angiogenesis and an inflammatory
microenvironment48. Additional studies are clearly needed to
more fully understand the impact of the chemokine-secretory
response on clinical outcomes in patients receiving drugs that
interfere with glutamine metabolism, as well as the potential
benefits of combining these agents with antagonists of IL-8 and
other tumour-derived chemokines.
Methods
Antibodies and chemicals.The following antibodies were used for western
blotting and immunofluorescence studies: LC3 (NB100-2220, Novus Biologicals,
dilution 1:1,000), p62 (BD Transduction, catalogue #610832, dilution 1:1,000),
actin (Sigma, A1978, dilution 1:100,000) and LAMP1 (ab25630, Abcam, dilution
1:100). Antibodies directed against GFP (#2555, dilution 1:100), phospho-S6K1
(p70 S6 Kinase Thr-389, #9205, dilution 1:1,000), S6K1 (p70 S6 Kinase, #2708,
dilution 1:1,000), mTOR (#2983, dilution 1:100), RCAS1 (#6960, dilution 1:100),
syntaxin 6 (#28690, dilution 1:100), CHOP (#2895, dilution 1:1,000),
phospho-EIF2aSer-51 (#9721, dilution 1:1,000), EIF2a(#2103, dilution 1:1,000),
phospho-SAPK/JNK, Thr183/Tyr185 (#9255, dilution 1:1,000) and JNK1 (#3708,
dilution 1:1,000) were obtained from Cell Signaling Technologies.
Bafilomycin A1 (catalogue #B1793), tunicamycin (catalogue #T7765),
thapsigargin (catalogue #T9063), brefeldin-A (catalogue #B7651),
dimethyl-a-ketoglutarate (catalogue #349631), BPTES (catalogue #SML0601)
and DON (6-diazo-5-oxo-L-norleucine; catalogue #D2141) were obtained from
Sigma-Aldrich. The JNK inhibitor SP600125 was obtained from R&D Systems
(catalogue #1496). CCI-779 and WYE-125132 were synthesized at Pfizer, and have
been described previously29,39. Lysotracker Green DND-26 (#L-7526) and Red
DND-99 (#L7528) were obtained from Invitrogen Technologies. All chemicals
and kits were used according to the manufacturers’ recommendations, unless
otherwise indicated.
Cell culture.All cell lines used in the study were grown in a humidified 5% CO
2
atmosphere. U2OS, H4, HCT116 and A549 cell lines were purchased from the
ATCC and iBMK (generously provided from the laboratory of Dr Eileen White)
cells were cultured in DMEM (Gibco, catalogue #11995), containing 2 mM
glutamine, supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–
streptomycin (complete DMEM). The iBMK cell lines 6.1B11 (atg5 þ/þ) and 7.1B4
(atg5 /) and U2OS GFP-LC3 cells have been described before20,32. U2OS
mCherry-ATG5 cells were generated by transfecting early passage U2OS cells with
mCherry-ATG5 (EX-M0036-M55 from Genecopoeia) and selecting for G418
resistance (500 mgml1). U2OS mCherry-ULK1 cells were generated by
transfecting U2OS cells with mCherry-ULK1 (EX-M0809-Lv111 from
Genecopoeia) and selecting for puromycin resistance (1 mgml1). U2OS mCherry-
ATG5 cells were subsequently transfected with GFP-LC3 to generate U2OS
mCherry-ATG5/GFP-LC3 cells. U2OS mCherry-GFP-LC3 cells were generated by
transducing U2OS cells with a mCherry-GFP-LC3 plasmid.
For glutamine deprivation, cells were plated overnight in complete DMEM,
briefly washed with phosphate-buffered saline (PBS) and then transferred into
glutamine-free medium (glutamine- and pyruvate-free DMEM (Cellgrow,
catalogue #15-017-CV)) supplemented with 10% dialyzed FBS (Gibco, catalogue
#26400) and 1 mM sodium pyruvate (Cellgro, catalogue #25-000-CI). The
corresponding glutamine-replete medium was prepared by addition of 2 mM
glutamine (Cellgro, catalogue #25-005-CV) to glutamine-free medium. Cells were
cultured for 24 h in glutamine-free medium, unless otherwise indicated. For leucine
deprivation, cells were plated in glutamine/leucine-free media (MP Biomedicals,
#1342149) supplemented with 10% dialyzed FBS, sodium pyruvate and either
2 mM glutamine or 0.8 mM leucine.
Immunoblotting.For immunoblotting experiments, 500,000 cells were plated
overnight in six-well tissue culture plates. After treatment for the indicated times,
cells were washed once with PBS, and cell lysates were prepared by scraping in
NuPAGE lithium dodecyl sulfate (LDS) buffer (Invitrogen, catalogue #NP0008),
followed by sonication in a water bath. Protein concentrations were determined
with the Bio-Rad RC/DC protein assay, and equal amounts of protein were elec-
trophoresed through NuPAGE 4–12% bis–tris gradient gels with 2-(N-morpholi-
no)ethanesulphonic acid (MES) running buffer. Proteins were transferred to
nitrocellulose and incuba ted with primary antibody overnight at 4 °C. Proteins
were detected using chemiluminescence with appropriate horseradish peroxidase-
conjugated secondary antibodies. Uncropped scans of the immunoblot membranes
are provided as Supplementary Figs 7–14.
Live cell microscopy.Live U2OS cells expressing epitope-tagged proteins were
imaged with an inverted Olympus IX51 fluorescent microscope at 20 magnifi-
cation. For quantification of GFP-LC3 puncta, cells displaying 410 brightly
fluorescent GFP-LC3 puncta were counted as positive. For phagophore–lysosome
colocalization studies, treated cells were co-stained with Lysotracker Green for
lysosomes and Hoechst-33342 for nuclei. For mCherry-GFP-LC3 cells, simulta-
neous images in the Green and Red channel were acquired for the same field
of cells followed by evaluation of the digitally merged image to estimate
autophagosome and autolysosome number.
Immunofluorescence analysis.U2OS cells were grown overnight on borosilicate
glass chambered slides (Nunc, catalogue #155409). At the end of the experiment,
cells were fixed and permeabilized with the Fix and Perm kit (GAS004, Invitrogen).
Immunostaining was then performed by overnight incubation with the appropriate
antibodies (1:100 dilutions) at 4 °C. The cells were then stained with fluorescein
isothiocyanate- or Cy3-conjugated secondary antibodies, together with
40,6-diamidino-2-phenylindole (DAPI) as a nuclear counterstain. Confocal images
were generated using an LSM 510 Meta microscope at 40 magnification.
Microarray analysis.Microarray analysis was performed as described49. Briefly,
2 million U2OS cells were cultured for 24 h in glutamine-free medium or the same
medium with added glutamine. Cells were collected in buffer RLT (Qiagen)
followed by RNA isolation using the RNeasy kit (#74104, Qiagen). Complementary
DNA synthesis and subsequent in vitro antisense RNA (cRNA) amplification and
biotin labelling were performed as described49. For each sample, 10 mg of biotin-
labelled cRNA was fragmented and hybridized to Human Genome U133 þ2
GeneChip oligonucleotide arrays (Affymetrix) using buffers and conditions
recommended by the manufacturer. GeneChips were then washed and stained with
Streptavidin R-phycoerythrin (Molecular Probes) using the GeneChip Fluidics
Station 450, and scanned with a Affymetrix GeneChip Scanner 3000. Microarray
data was converted into MIAME format and was uploaded into the Gene
Expression Omnibus (GEO) database with accession number GSE59931.
Real-time PCR analysis.U2OS cells (500,000) were plated overnight followed by
treatment for 24 h with glutamine-free medium, glutamine-free medium with
added glutamine or with glutamine-containing medium supplemented with 1 mM
TPG. Cells were collected in buffer RLT (Qiagen) followed by RNA isolation using
the RNeasy kit (#74104, Qiagen). cDNA was prepared with the iScript cDNA
synthesis kit (170–8890, Bio-Rad) according to the manufacturer’s instructions.
Real-time PCR analysis was then performed with standard cycling conditions on a
Bio-Rad CFX96 real-time system using TaqMan primer sets offering the best
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900 ARTICLE
NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications 11
&2014 Macmillan Publishers Limited. All rights reserved.
coverage. Relative transcript levels were calculated using the comparative C
t
method and normalized to the housekeeping gene, beta-microglobulin.
Cytokine array.Cytokine profiling was performed with the Human Cytokine
Array Kit, Panel A (#ARY005, R&D Systems). Briefly, conditioned medium from
cells treated for 24 h with glutamine-free or -supplemented medium was collected
and clarified by centrifugation. The cleared medium was diluted, mixed with a
biotinylated detection antibody cocktail and then incubated with the Human
Cytokine Array Panel A membranes. Following a wash to remove unbound
material, streptavidin-horseradish peroxidase and chemiluminescent detection
reagents were added sequentially and the cytokine array was visualized by
autoradiography.
IL-8 enzyme-linked immunosorbent assay.IL-8 levels in the cell culture
supernatants were measured with a quantitative sandwich enzyme immunoassay
(Quantikine ELISA Kit—D8000C, R&D Systems). Briefly, U2OS or A549 cells were
plated (5 105cells per well) in a six-well plate and were allowed to adhere
overnight. Cells were then treated for 24 h with glutamine-free or -supplemented
medium. Conditioned media were collected for determination of IL-8 levels with
the Quantikine ELISA kit.
siRNA transfections.U2OS cells were plated overnight in tissue culture-treated,
six-well plates (3 105cells per well), and were subsequently transfected with
50 nM siRNA per well mixed with Lipofectamine RNAiMax (13778-150, Invitro-
gen) according to the manufacturer’s protocol. Cells were subjected to glutamine
deprivation at 72 h post transfection. The following siRNA duplexes were used and
purchased from Ambion: control siRNA (#4390843), siJNK#1 (S28271), siJNK#2
(S28272), siIRE1#1 (S200430, S200432) and siIRE1#2 (S20800, S20801).
ATG7 TALENS.ATG7 TALENS were purchased from PNA Bio. U2OS cells were
plated at a density of 2 106in 10 cm plates. After 24 h, cells were tran sfected with
cytomegalovirus (CMV) TALEN_L2 (50-TATTGGAACACTGTATAACA-30),
CMV TALEN_R2 (50-TGTCCTTGGGAGCTTCATCC-30) and RG2S_2_CMV
(GFP/red fluorescent protein surrogate reporter) using Lipofectamine 2000 at a
ratio of 1:1:2. At 48 h post transfection, single cells were sorted into individual wells
of a 96-well plate. Control cell lines were generated from cells expressing one
TALEN module (which results in red fluorescence) by fluorescence-activated
cell sorting-based sorting for red fluorescent protein-only expressing cells
(U2OS_TAL2_CTL). ATG7-knockout cells were generated from cells expressing
both the left and right TALEN modules, which result in both red and green
fluorescence (U2OS_TAL2_A7, U2OS_TAL2_A8). For experiments,
U2OS_TAL2_CTL and U2OS TAL2_A8 were used unless otherwise indicated.
Sulphorhodamine B viability assay.Cells were plated at a density of 100,000 cells
per well in a 24-well plate. After 24 h, the medium was changed as described in the
Results section, and cells were incubated for 6 days. The culture medium was
replaced with complete medium and cells were allowed to recover for 6 days.
Following recovery, cells were fixed in 10% trichloroacetic acid for 1 h at 4°C. Cells
were washed five times in water and stained with a 0.057% (w/v) solution of SRB
(Sulphorhodamine B; Sigma, catalogue #S-9012) in 1% acetic acid. The plate was
then washed five times with 1% acetic acid. SRB was solubilized in 10 mM
unbuffered Tris base and absorbance was read at 540 nm.
Metabolic analyses.Cells were seeded in triplicate in six-well plates. After the
indicated treatments, the cell culture medium was removed from plates and cells
were washed three times with cold PBS. The cell culture plates were stored at
80 °C until analysis. Intracellular metabolites were extracted by adding 0.25 ml of
cold organic extraction buffer consisting of 40/40/20 mixture of acetonitrile/
methanol/water for 15 min on ice followed by addition of 0.25 ml water for 5 min
on ice. Ten mlof2mM iso-ATP and 5-fluoro-2-methylpyridine (injection stan-
dards) was added to the pooled metabolites, and the resulting solutions were
centrifuged at 16,000 gat 4 °C for 15 min. Two Liquid Chromatography-Multiple
Reaction Monitoring (LC-MRM) methods were used for analysis of the indicated
metabolites: (1) an Imtakt Unison UK-Phenyl 75 2.0 mm column coupled with
AB SCIEX API4000 for monitoring TCA metabolites in the negative ionization
mode, with mobile phase A of 100% water containing 0.1% formic acid and mobile
phase B of 100% acetonitrile containing 0.1% formic acid, was used at flow rate of
0.3 ml min 1and column oven temperature of 30 °C. The following gradient was
used: 0–4 min 0% solvent B, 4–5 min 0–100% solvent B, 5–8 min 100% solvent B,
8–9 min 100–0% B and 9–15 min 0% solvent B. (2) An Imtakt Scherzo SM-C18
150 2.0 mm column coupled with AB SCIEX API4000 for monitoring amino
acids in the positive ionization mode, with mobile phase A of 100% water con-
taining 0.1% formic acid and mobile phase B of 100% acetonitrile containing 0.1%
formic acid, was used with the flow rate of 0.3 ml min 1at 30 °C. The following
gradient was used: 0–6 min 0% solvent B, 6–16 min 50% solvent B, 16–18 min
50–100% solvent B, 18–20 min 100% solvent B, 20–20.5 min 100–0% solvent B and
20.5–30 min 0% solvent B.
Statistics.All experiments presented were performed in triplicate. Error bars on
graphs represent s.d. unless otherwise stated. Pvalues were calculated using a
paired student’s t-test, two tailed, type 1.
References
1. Hanahan, D. & Weinberg, R. A. Hallmarks of cancer: the next generation. Cell
144, 646–674 (2011).
2. DeBerardinis, R. J., Lum, J. J., Hatzivassiliou, G. & Thompson, C. B. The
biology of cancer: metabolic reprogramming fuels cell growth and proliferation.
Cell Metab. 7, 11–20 (2008).
3. Cantor, J. R. & Sabatini, D. M. Cancer cell metabolism: one hallmark, many
faces. Cancer Discov. 2, 881–898 (2012).
4. Lunt, S. Y. & Vander Heiden, M. G. Aerobic glycolysis: meeting the metabolic
requirements of cell proliferation. Annu. Rev. Cell Dev. Biol. 27, 441–464
(2011).
5. Hsu, P. P. & Sabatini, D. M. Cancer cell metabolism: Warburg and beyond. Cell
134, 703–707 (2008).
6. Deberardinis, R. J., Sayed, N., Ditsworth, D. & Thompson, C. B. Brick by
brick: metabolism and tumor cell growth. Curr. Opin. Genet. Dev. 18, 54–61
(2008).
7. Cheong, H., Lu, C., Lindsten, T. & Thompson, C. B. Therapeutic targets in
cancer cell metabolism and autophagy. Nat. Biotechnol. 30, 671–678 (2012).
8. Dang, C. V., Hamaker, M., Sun, P., Le, A. & Gao, P. Therapeutic targeting of
cancer cell metabolism. J. Mol. Med. 89, 205–212 (2011).
9. Shanware, N. P., Mullen, A. R., DeBerardinis, R. J. & Abraham, R. T.
Glutamine: pleiotropic roles in tumor growth and stress resistance. J. Mol. Med.
89, 229–236 (2011).
10. DeBerardinis, R. J. & Cheng, T. Q’s next: the diverse functions of glutamine in
metabolism, cell biology and cancer. Oncogene 29, 313–324 (2010).
11. Mullen, A. R. et al. Reductive carboxylation supports growth in tumour cells
with defective mitochondria. Nature 481, 385–388 (2012).
12. Wellen, K. E. et al. The hexosamine biosynthetic pathway couples growth
factor-induced glutamine uptake to glucose metabolism. Genes Dev. 24,
2784–2799 (2010).
13. Qing, G. et al. ATF4 regulates MYC-mediated neuroblastoma cell death upon
glutamine deprivation. Cancer Cell 22, 631–644 (2012).
14. Wise, D. R. et al. Myc regulates a transcriptional program that stimulates
mitochondrial glutaminolysis and leads to glutamine addiction. Proc. Natl
Acad. Sci. USA 105, 18782–18787 (2008).
15. Son, J. et al. Glutamine supports pancreatic cancer growth through a
KRAS-regulated metabolic pathway. Nature 496, 101–105 (2013).
16. Dang, C. V. Therapeutic targeting of Myc-reprogrammed cancer cell
metabolism. Cold Spring Harb. Symp. Quant. Biol. 76, 369–374 (2011).
17. Wise, D. R. & Thompson, C. B. Glutamine addiction: a new therapeutic target
in cancer. Trends Biochem. Sci. 35, 427–433 (2010).
18. Yang, Z. & Klionsky, D. J. Mammalian autophagy: core molecular machinery
and signaling regulation. Curr. Opin. Cell Biol. 22, 124–131 (2010).
19. Mizushima, N. Autophagy in protein and organelle turnover. Cold Spring Harb.
Symp. Quant. Biol. 76, 397–402 (2011).
20. Eng, C. H., Yu, K., Lucas, J., White, E. & Abraham, R. T. Ammonia derived
from glutaminolysis is a diffusible regulator of autophagy. Sci. Signal. 3, ra31
(2010).
21. Cheong, H., Lindsten, T., Wu, J., Lu, C. & Thompson, C. B. Ammonia-induced
autophagy is independent of ULK1/ULK2 kinases. Proc. Natl Acad. Sci. USA
108, 11121–11126 (2011).
22. Rabinowitz, J. D. & White, E. Autophagy and metabolism. Science 330,
1344–1348 (2010).
23. Narita, M. et al. Spatial coupling of mTOR and autophagy augments secretory
phenotypes. Science 332, 966–970 (2011).
24. Ron, D. & Walter, P. Signal integration in the endoplasmic reticulum unfolded
protein response. Nat. Rev. Mol. Cell Biol. 8, 519–529 (2007).
25. Wek, R. C., Jiang, H. Y. & Anthony, T. G. Coping with stress: eIF2 kinases and
translational control. Biochem. Soc. Trans. 34, 7–11 (2006).
26. Jefferies, K. C., Cipriano, D. J. & Forgac, M. Function, structure and
regulation of the vacuolar (H þ)-ATPases. Arch. Biochem. Biophys. 476, 33–42
(2008).
27. Perez-Sayans, M., Garcia-Garcia, A., Reboiras-Lopez, M. D. & Gandara-Vila, P.
Role of V-ATPases in solid tumors: importance of the subunit C (Review). Int.
J. Oncol. 34, 1513–1520 (2009).
28. Coppe, J. P., Desprez, P. Y., Krtolica, A. & Campisi, J. The senescence-
associated secretory phenotype: the dark side of tumor suppression. Annu. Rev.
Pathol. 5, 99–118 (2010).
29. Yu, K. et al. mTOR, a novel target in breast cancer: the effect of CCI-779, an
mTOR inhibitor, in preclinical models of breast cancer. Endocr. Relat. Cancer 8,
249–258 (2001).
30. Klionsky, D. J. et al. Guidelines for the use and interpretation of assays
for monitoring autophagy in higher eukaryotes. Autophagy 4, 151–175
(2008).
ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900
12 NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications
&2014 Macmillan Publishers Limited. All rights reserved.
31. Bjorkoy, G. et al. p62/SQSTM1 forms protein aggregates degraded by
autophagy and has a protective effect on huntingtin-induced cell death. J. Cell
Biol. 171, 603–614 (2005).
32. Mathew, R. et al. Autophagy suppresses tumorigenesis through elimination of
p62. Cell 137, 1062–1075 (2009).
33. Mathew, R. et al. Autophagy suppresses tumor progression by limiting
chromosomal instability. Genes Dev. 21, 1367–1381 (2007).
34. Itakura, E. & Mizushima, N. Characterization of autophagosome formation site
by a hierarchical analysis of mammalian Atg proteins. Autophagy 6, 764–776
(2010).
35. Sengupta, S., Peterson, T. R. & Sabatini, D. M. Regulation of the mTOR
complex 1 pathway by nutrients, growth factors, and stress. Mol. Cell 40,
310–322 (2010).
36. Gibbons, J. J., Abraham, R. T. & Yu, K. Mammalian target of rapamycin:
discovery of rapamycin reveals a signaling pathway important for normal and
cancer cell growth. Semin. Oncol. 36 (Suppl 3), S3–S17 (2009).
37. Sancak, Y. et al. Ragulator-Rag complex targets mTORC1 to the lysosomal
surface and is necessary for its activation by amino acids. Cell 141, 290–303
(2010).
38. Zoncu, R. et al. mTORC1 senses lysosomal amino acids through an inside-out
mechanism that requires the vacuolar H( þ)-ATPase. Science 334, 678–683
(2011).
39. Yu, K. et al. Beyond rapalog therapy: preclinical pharmacology and antitumor
activity of WYE-125132, an ATP-competitive and specific inhibitor of
mTORC1 and mTORC2. Cancer Res. 70, 621–631 (2010).
40. Tanida, I., Ueno, T. & Kominami, E. LC3 conjugation system in mammalian
autophagy. Int. J. Biochem. Cell Biol. 36, 2503–2518 (2004).
41. Kato, H. et al. mTORC1 serves ER stress-triggered apoptosis via selective
activation of the IRE1-JNK pathway. Cell Death Differ. 19, 310–320 (2012).
42. Hoffmann, E., Dittrich-Breiholz, O., Holtmann, H. & Kracht, M.
Multiple control of interleukin-8 gene expression. J. Leukoc. Biol. 72,
847–855 (2002).
43. Nicklin, P. et al. Bidirectional transport of amino acids regulates mTOR and
autophagy. Cell 136, 521–534 (2009).
44. Duran, R. V. et al. Glutaminolysis activates Rag-mTORC1 signaling. Mol. Cell
47, 349–358 (2012).
45. Waugh, D. J. & Wilson, C. The interleukin-8 pathway in cancer. Clin. Cancer
Res. 14, 6735–6741 (2008).
46. Yu, L. et al. Termination of autophagy and reformation of lysosomes regulated
by mTOR. Nature 465, 942–946 (2010).
47. Gross, M. I. et al. Antitumor activity of the glutaminase inhibitor CB-839 in
triple-negative breast cancer. Mol. Cancer Ther. 13, 890–901 (2014).
48. Welte, G. et al. Interleukin-8 derived from local tissue-resident stromal cells
promotes tumor cell invasion. Mol. Carcinog. 51, 861–868 (2012).
49. Damelin, M. et al. Delineation of a cellular hierarchy in lung cancer reveals
an oncofetal antigen expressed on tumor-initiating cells. Cancer Res. 71,
4236–4246 (2011).
Acknowledgements
We would like to thank the members of the Oncology Research Unit’s Metabolism group
for helpful discussions. We would also like to thank Christine Loreth, Wenyan Zhong
and Sylvia Vong for expert technical assistance. N.P.S. and K.B. acknowledge critical
support from the Pfizer Postdoctoral Training Program.
Author contributions
N.P.S. and K.B. designed and performed all the experiments in the paper. C.H.E.
and V.R.F. provided valuable insight and guidance. F.W. and J.M. performed the
metabolomics analysis. M.F. performed the microarray analysis and R.T.A. supervised
all experiments and edited the final manuscript.
Additional information
Supplementary Information accompanies this paper at http://www.nature.com/
naturecommunications
Competing financial interests: The authors are employees and shareholders of Pfizer, Inc.
Reprints and permission information is available online at http://npg.nature.com/
reprintsandpermissions/
How to cite this article: Shanware, N. P. et al. Glutamine deprivation stimulates
mTOR-JNK-dependent chemokine secretion. Nat. Commun. 5:4900
doi: 10.1038/ncomms5900 (2014).
This work is licensed under a Creative Commons Attribution 4.0
International License. The images or other third party material in this
article are included in the article’s Creative Commons license, unless indicated otherwise
in the credit line; if the material is not included under the Creative Commons license,
users will need to obtain permission from the license holder to reproduce the material.
To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
NATURE COMMUNICATIONS | DOI: 10.1038/ncomms5900 ARTICLE
NATURE COMMUNICATIONS | 5:4900 | DOI: 10.1038/ncomms5900 | www.nature.com/naturecommunications 13
&2014 Macmillan Publishers Limited. All rights reserved.
