Acidic pH is a metabolic switch for 2-Hydroxyglutarate generation and signaling

Article abstract of DOI:10.1074/jbc.M116.738799

2-Hydroxyglutarate (2-HG) is an important epigenetic regulator, with potential roles in cancer and stem cell biology. The D-(R)-enantiomer (D-2-HG) is an oncometabolite generated from α-ketoglutarate (α-KG) by mutant isocitrate dehydrogenase, whereas L-(S)-2-HG is generated by lactate dehydrogenase and malate dehydrogenase in response to hypoxia. Because acidic pH is a common feature of hypoxia, as well as tumor and stem cell microenvironments, we hypothesized that pH may regulate cellular 2-HG levels. Herein we report that cytosolic acidification under normoxia moderately elevated 2-HG in cells, and boosting endogenous substrate α-KG levels further stimulated this elevation. Studies with isolated lactate dehydrogenase-1 and malate dehydrogenase-2 revealed that generation of 2-HG by both enzymes was stimulated severalfold at acidic pH, relative to normal physiologic pH. In addition, acidic pH was found to inhibit the activity of the mitochondrial L-2-HG removal enzyme L-2-HG dehydrogenase and to stimulate the reverse reaction of isocitrate dehydrogenase (carboxylation of α-KG to isocitrate). Furthermore, because acidic pH is known to stabilize hypoxia-inducible factor (HIF) and 2-HG is a known inhibitor of HIF prolyl hydroxylases, we hypothesized that 2-HG may be required for acid-induced HIF stabilization. Accordingly, cells stably overexpressing L-2-HG dehydrogenase exhibited a blunted HIF response to acid. Together, these results suggest that acidosis is an important and previously overlooked regulator of 2-HG accumulation and other oncometabolic events, with implications for HIF signaling.

Full text of DOI:10.1074/jbc.M116.738799

JBC Papers in Press. Published on August 10, 2016 as Manuscript M116.738799
The latest version is at http://www.jbc.org/cgi/doi/10.1074/jbc.M116.738799
Acid and 2‐Hydroxyglutarate
Acidic pH is a Metabolic Switch for 2-Hydroxyglutarate Generation and Signaling
1
2
3
Sergiy M. Nadtochiy , Xenia Schafer , Dragony Fu ,
4
2
1*
Keith Nehrke , Joshua Munger , Paul S. Brookes .
1
2
3
4
From: Departments of Anesthesiology, Biochemistry, Biology, Medicine,
University of Rochester, Rochester, NY 14642, USA.
Running Title: Acid & 2-hydroxyglutarate
*
To whom correspondence should be addressed: Paul S. Brookes, PhD., Department of Anesthesiology,
University of Rochester Medical Center, 601 Elmwood Avenue, Rochester, NY 14642, USA. Tel. (585) 273-
626. paul_brookes@urmc.rochester.edu
1
Keywords: Hypoxia, 2-Hydroxyglutarate, Cancer, Acidosis, Metabolism
ABSTRACT
2
-hydroxyglutarate (2-HG) is an important
The field of cancer biology has long been
rapt by the notion of a cancer-specific metabolic
phenotype, perhaps most famously embodied in
the “Warburg effect”, wherein glycolytic
metabolism predominates in cancer cells despite
epigenetic regulator, with potential roles in cancer
and stem cell biology. The D (R) enantiomer (D-2-
HG) is an oncometabolite generated from α-
ketoglutarate (α-KG) by mutant isocitrate
dehydrogenase (ICDH), while L (S) 2-HG is
generated by lactate dehydrogenase (LDH) and
malate dehydrogenase (MDH) in response to
hypoxia. Since acidic pH is a common feature of
hypoxia, as well as tumor and stem cell
microenvironments, we hypothesized that pH may
regulate cellular 2-HG levels. Herein we report
that cytosolic acidification under normoxia
moderately elevated 2-HG in cells, and boosting
endogenous substrate α-KG levels further
stimulated this elevation. Studies with isolated
LDH-1 and MDH-2 revealed that generation of 2-
HG by both enzymes was stimulated several-fold
at acidic pH, relative to normal physiologic pH. In
addition, acidic pH was found to inhibit the
activity of the mitochondrial L-2-HG removal
enzyme L-2-HG dehydrogenase, and to stimulate
the reverse reaction of ICDH (carboxylation of α-
KG to isocitrate). Furthermore, since acidic pH is
known to stabilize hypoxia-inducible factor (HIF),
and 2-HG is a known inhibitor of HIF prolyl
hydroxylases, we hypothesized that 2-HG may be
required for acid-induced HIF stabilization.
Accordingly, cells stably over-expressing L-
O availability and largely intact mitochondrial
2
respiratory function (1).
A prominent feature of the cancer
metabolic phenotype (reviewed in (2)) is an
elevated level of the small metabolic acid 2-
hydroxyglutarate (2-HG), derived from the TCA
cycle intermediate α-ketoglutarate (α-KG) (3). The
D (R) enantiomer of 2-HG (D-2-HG) was shown
to be generated by mutant forms of isocitrate
dehydrogenase (ICDH) that are associated with a
variety of cancers including aggressive gliomas
(4). In addition, more recently the L (S)
enantiomer (L-2-HG) was shown to be generated
under
hypoxic
conditions
by
lactate
dehydrogenase (LDH) and malate dehydrogenase
(MDH) (5, 6). We have also reported elevated
D/L-2-HG levels in the heart following ischemic
preconditioning (7).
In addition to synthesis, 2-HG levels are
regulated by a pair of dehydrogenases that convert
2-HG back to α-KG (i.e., L-2-HGDH and D-2-
HGDH). Mutations in these enzymes manifest as
the hydroxyglutaric acidurias, devastating
inherited metabolic diseases with symptoms
including epilepsy and cerebellar ataxia (8-11).
However, the importance of these dehydrogenases
in regulating 2-HG levels in other settings is not
clear.
2
HGDH exhibited a blunted HIF response to acid.
Together these results suggest that acidosis is an
important and previously overlooked regulator of
2
-HG accumulation and other oncometabolic
events, with implications for HIF signaling.
The downstream signaling roles of D-2-
HG in cancer biology and of L-2-HG in hypoxia
1
Copyright 2016 by The American Society for Biochemistry and Molecular Biology, Inc.

Acid and 2‐Hydroxyglutarate
or stem cell biology, are thought to be mediated by
epigenetic effects (12-14), owing to competitive
inhibition of the α-KG-dependent dioxygenase
superfamily of enzymes. This includes the JmjC
domain-containing histone demethylases, the TET
we hypothesized that α-KG may be limiting in
acidic pH alone. As such, we found that inhibiting
the TCA cycle enzyme α-KGDH with
ketomethyvalerate (KMV) boosted cellular α-KG
levels by ~5-fold regardless of pH (Figure 1E).
Furthermore, while KMV alone drove a small
increase in 2-HG levels at pH 7.4 (Figure 1D), the
combination of KMV plus acidic pH yielded an
overall 13-fold increase in 2-HG levels. Together
these data indicate that reproducing the acidic pH
and elevated α-KG levels seen in hypoxia is
sufficient to increase 2-HG in cells. Acidosis may
therefore be an important variable that drives 2-
HG elevation in hypoxia.
5
-methylcytosine hydroxylases, the EGLN prolyl-
hydroxylases that regulate hypoxia inducible
factor (HIF), and the AlkB homolog family of
DNA/RNA demethylases (15-18). As such, 2-HG
is a potentially important link between metabolism
and epigenetic signaling.
A common feature of hypoxia, as well as
the tumor and stem cell microenvironments, is
metabolic acidosis. However, the role of pH in
regulating 2-HG formation and disposal has not
been considered. We tested the hypothesis that
acidic pH is a regulator of 2-HG metabolism at the
cellular, mitochondrial, and isolated enzyme
levels. Our results indicate that acidic pH can
independently drive elevated 2-HG levels, and we
propose that pH regulation of 2-HG may have
important implications for 2-HG signaling in
hypoxia and other settings.
Operating on the premise that the 2-HG
1
observed in cells was the L-isomer (see footnote ),
we next investigated the mechanism of acid-
induced 2-HG elevation, by examining the effects
of pH on the known enzymes that generate and
remove this metabolite – namely LDH, MDH,
ICDH and 2-HGDH. The purity of commercial
+
isolated LDH-1, MDH-2 and NADP -dependent
ICDH were verified by gel electrophoresis, with
ICDH found to be somewhat impure (Figure 2A).
A spectrophotometric assay for the consumption
of NADH by LDH in the presence of α-KG
(Figure 3A inset) was performed across a range of
pH values spanning 6.6–7.8 (i.e. encompassing the
pH values studied in intact cells), and revealed that
NADH consumption was approximately 2.5-fold
faster at pH 6.8 than at pH 7.4 (Figure 3A). In
agreement with the cell data (Figure 1D), the rate
was also [α-KG] dependent, being faster at 10 mM
than at 1 mM α-KG. The spectrophotometric data
were confirmed by LC-MS/MS measurement of 2-
HG generation (Figure 3B), with approximate
molar equivalents of 2-HG produced across the pH
RESULTS & DISCUSSION
We investigated the role of pH as an
independent variable in the regulation of 2-HG
levels in HEK293 cells, using an ammonium plus
5
-(N-Ethyl-N-isopropyl) amiloride (EIPA) pH
clamp system (19). This system was capable of
depressing and maintaining intracellular pH to 6.8
within 2 hrs. (Figures 1A–C). This magnitude of
pH depression (approximately 0.5 units) is similar
to that previously reported upon exposure of cells
to hypoxia/anoxia (20-22). As such, the pH-
dependent signaling effects measured herein can
be considered to represent those that may be
expected to occur in hypoxic tissues.
2
range 6.6–7.8 (Figure 2B, r = 0.97).
Although this degree of acidification
drove a 4-fold elevation in 2-HG levels (Figure
D, left two bars), this was somewhat lower than
the >8-fold elevation reported in the same cell type
in response to hypoxia (5, 6). Since 2-HG
generation also depends on the availability of
substrate α-KG, which is elevated in hypoxia (6),
The native activity of LDH (i.e.,
conversion of pyruvate to lactate) was unaffected
by pH (Figure 3C); unsurprising for an enzyme
that typically drives metabolic acidosis.
Fractionally, the 2-HG generating capacity of
LDH was at least 4 orders of magnitude lower
than its native activity (Figure 3A vs. 3C),
indicating that acidic pH unmasks a latent 2-HG
generating activity in LDH without impacting its
native activity. This observation also suggests that
the generation of 2-HG is not a quantitatively
important metabolic sink for NADH.
1
1
1
Note – The methodology used herein was incapable of
distinguishing between L and D 2-HG. However, since the D
enantiomer is primarily found in cancers associated with
ICDH mutation, while the L enantiomer is primarily made by
LDH and MDH under hypoxia, we assume that the bulk of 2-
HG measured herein was the L enantiomer.
2

Acid and 2‐Hydroxyglutarate
As a control, the effect of hypoxia on the
ability of isolated LDH to generate 2-HG at pH 7.4
was also studied. Hypoxia was verified by
measuring the spectrum of oxy- vs. deoxy-
hemoglobin in the cuvet (Figure 4A). In contrast
to the situation seen in cells (5, 6), no effect of pO2
on 2-HG generation by isolated LDH was seen
KG indeed consumed NADPH in a manner that
was accelerated at acidic pH (Figure 5A), no 2-HG
generation was seen (Figure 5B). Instead, ICDH
was observed to carboxylate α-KG to isocitrate,
and this reaction was stimulated at acidic pH
(Figure 5C). This is somewhat expected because
the reaction uses protons as a substrate (Figure 5C
inset). Such reversal of ICDH activity was
previously shown to be triggered under hypoxic
conditions (23), and is an important pathway for
diversion of carbon from anaplerotic glutamine
toward lipid synthesis (via ATP citrate lyase) in
cancer cells. Thus, while native ICDH does not
appear to be a source for acid-triggered 2-HG
generation, these data suggest that acidic pH may
regulate numerous oncometabolic pathways,
including reductive α-KG carboxylation.
(
Figures 4B and 4C). These data suggest that
hypoxic stimulation of 2-HG generation by LDH
is not a property of the isolated enzyme, but rather
it requires the intact hypoxic cell environment,
likely including acidosis.
Similar to LDH, the generation of 2-HG
by MDH was also found to be higher at pH 6.8
than at pH 7.4 (Figures 3D and 3E). As seen with
LDH, spectrophotometric and LC-MS/MS data for
2
MDH correlated well (Figure 2C, r = 0.98), with
approximate molar equivalents of 2-HG produced
across the pH range 6.6–7.8. Notably, unlike LDH
Since cellular 2-HG levels are also
regulated by specific dehydrogenases, we next
tested the effects of pH on mitochondrial L-2-
HGDH activity, using both a formazan-linked
colorimetric assay (10) and LC-MS/MS to
quantify α-KG generation from the substrate L-2-
(
Figure 3C) the native activity of MDH was itself
pH sensitive, being inhibited ~30% at pH 6.8
versus pH 7.4 (Figure 3F). At first glance, this
result might suggest an acid-induced activity
switch, away from native and toward 2-HG-
generation. However, it should again be noted that
the 2-HG generating activity of MDH was at least
2
HG. Both assays correlated well (Figure 2D, r =
0.98), and showed that L-2-HGDH activity was
inhibited by acidic pH (Figure 6). Together with
the data in Figure 3 this result indicates that acidic
pH both promotes the formation of L-2-HG and
inhibits its removal, suggesting a coordinated
metabolic response to elevate L-2-HG under acidic
conditions. Based on the data in Figures 1, 3 and
6, we conclude that pH is an important
determinant of cellular 2-HG levels, and that
acidosis may play a key role in the L-2-HG
elevation observed in hypoxia (5, 6).
3
orders of magnitude lower than its native
activity. Although 2-HG generation has been
termed a “metabolic error” and the 2-HG
dehydrogenases referred to as metabolite repair
enzymes (11), our data suggest that the 2-HG
synthetic capacity of MDH unmasked by acidic
pH is fractionally small, and does not represent a
major diversion of carbon away from the TCA
cycle.
Comparing LDH and MDH, at the pH
range seen in cells (6.8 vs. 7.4), and at the same α-
KG level (10 mM), the relative acid-induced
stimulation of 2-HG generation was slightly larger
for LDH (2-fold) vs. MDH (1.3-fold). In addition,
the pH response of LDH appeared to initiate at
slightly more neutral pH, whereas the MDH
response was flatter until more acidic pH values
were attained. These data, plus the >3-fold greater
specific activity for 2-HG generation by LDH vs.
MDH, suggest that LDH may be a quantitatively
more important source of 2-HG in response to
acid.
Biochemically, the reaction catalyzed by
LDH and MDH during generation of 2-HG is the
same as the native reactions of these enzymes,
namely the NADH dependent conversion of an α-
keto acid to its corresponding α-hydroxy acid
(Figure 7). The only difference between substrates
is the corresponding R group, and our current
findings suggest that the mechanism of acid-
induced 2-HG generation by LDH and MDH may
involve changes in access to the active site,
permitting entry of a larger R group. Such a model
leads to a simple prediction: if LDH under acidic
conditions can accommodate the R group of α-KG
((CH ) –COOH)), it should also be able to
The ability of native (non-mutant) ICDH
to convert α-KG to 2-HG in a pH-sensitive manner
was also tested. While ICDH in the presence of α-
2
2
accommodate oxaloacetate (OAA). Figures 7B
and 7C show that this prediction is correct, with
3

Acid and 2‐Hydroxyglutarate
LDH able to convert OAA to malate (i.e. to
perform the MDH reaction) in an acid-sensitive
manner. A crystal structure for LDH in the
presence of the inhibitor oxamate (R group = NH2,
PDB 1I0Z), shows the region of the protein that
may accommodate the R group occupied by 2
molecules of water, with no acid-sensitive residues
such as histidine nearby (24). As such, the
mechanism by which acidic pH may allow larger
substrates into the active site of LDH remains to
be elucidated. In addition, the ability of LDH (or
MDH) to accept even larger substrates such as α-
ketobutyrate is unknown, but could be worthy of
investigation because elevated α-hydroxybutyrate
has been shown to be a component of the
metabolic signature of mitochondrial dysfunction
across multiple organisms (25, 26). In general, the
potential for acidic pH to remodel metabolite
profiles by promoting α-keto to α-hydroxy acid
conversion, remains largely unexplored.
with extracellular pH (pHex) tending more toward
acidity than intracellular pH (pH ) (28, 29).
in
Furthermore, although it has been suggested that
exposure to acidic pHex may play a role in
promoting “stem-ness” (30), there has been much
recent controversy regarding the effects of acid
exposure on stem cell phenotype (31). Thus, while
it could be speculated that a simple elevation in 2-
HG levels can transmit a pH signal to the
epigenetic machinery and bring about a cancer or
stem cell phenotype, the situation is likely to be far
more complex in-vivo. As the data in Figure 1
suggest, the availability of substrate α-KG is also a
major determinant of 2-HG generation, and the
fact that α-KG is also a substrate for the
dioxygenase signaling enzymes that 2-HG inhibits,
suggests that the α-KG/2-HG ratio is likely to be
an important determinant of 2-HG signaling
outcomes.
Finally, the current findings may also hold
Downstream signaling by 2-HG is thought
therapeutic implications for the hydroxyglutaric
acidurias. Lactic acidosis has been reported in
some 2-HG aciduric patients (32), and it could be
speculated that acidotic episodes may trigger 2-
HG accumulation. As such, the careful
management of pH to avoid acidosis, for example
by treatment with dichloroacetate, (33) may be an
overlooked therapeutic approach in these patients.
to be mediated by the inhibition of α-KG-
dependent dioxygenases, such as the EGLN
prolyl-hydroxylases that regulate HIF (15). It is
known that acidic pH alone can activate HIF-1α
(
27), and therefore we hypothesized that 2-HG
may play a role in acid-induced HIF activation. To
test this, a cell line with stable transgenic over-
expression of L-2-HGDH was generated (Figure
8
1
). The L-2-HGDH-Tg cells (clone #2) exhibited
2-fold greater L-2-HGDH activity compared to
EXPERIMENTAL PROCEDURES
All reagents including isolated enzymes
mock transfected controls (Figure 8B).
Furthermore, exposure of cells to acidosis plus
KMV as in Figure 1 resulted in elevated 2-HG in
control cells with a blunted response in L-2-
HGDH-Tg cells (Figure 8C). Measurement of HIF
levels by western blot (Figure 9) revealed that HIF
stabilization in response to acid plus KMV was
significantly blunted in L-2-HGDH-Tg cells (lane
were purchased from Sigma (St. Louis, MO)
unless otherwise stated. Male C57BL/6J mice
were bred in-house from stocks obtained from Jax
Labs (Bar Harbor, ME) and were maintained in a
thermoneutral environment on a 12 hr. light/dark
cycle with food and water available ad libitum. All
experiments were in compliance with the NIH
“Guide for the Care and Use of Laboratory
Animals” and were approved by a local ethics
committee (UCAR protocol #2007-087). Enzyme
activities were assayed in 25 mM potassium
phosphate buffer at 37°C, adjusted to specific pH
values as indicated in the figures. Conversion of α-
KG to L-2-HG (mediated by LDH, MDH, or
ICDH) and was monitored as the oxidation of
NADH (0.1 mM) spectrophotometrically at 340
nm. For hypoxia experiments on isolated enzymes,
an “open-flow” apparatus was used, in which the
spectrophotometer cuvet was fitted with a lid and
stirring apparatus, allowing purging of the
3
vs. 4). In contrast, HIF induction by hypoxia was
similar between both control and L-2-HGDH-Tg
cells, indicating that canonical HIF signaling was
intact. These data indicate that 2-HG generation is
required for acid-induced HIF stabilization.
Although the current data suggest that pH
is an important regulator of 2-HG signaling, the
relative importance of the pH  2-HG axis in
cancer or stem cell biology remains to be
determined. While acidic pH is often associated
with the high glycolytic metabolism of cancer
cells, the regulation of pH in cancer is complex,
4

Acid and 2‐Hydroxyglutarate
headspace with gas mixtures defined by mass-flow
controllers (34). Gas mixtures were humidified to
prevent sample evaporation, and successful
achievement of anoxia after 1 hr. of purging with
argon was verified by the spectrum of oxy/deoxy-
hemoglobin (Hb, 100 μM) in a matching open-
flow neighboring cuvet. Oxy-Hb was freshly
prepared by dithionite reduction followed by gel
volumes media and pelleting at 500 x g for 3 min.
Cells were resuspended in 3 volumes media and
enumerated on a hemocytometer after Trypan blue
6
2
staining. Cells were seeded at 10 cells per 75 cm
flask, 24 hr. prior to experiments. Cells were
incubated in bicarbonate-free DMEM with 4 mM
L-glutamine, 0.1 mM pyruvate, 10 mM glucose,
and 10 mM HEPES (adjusted to pH 7.4 or 6.5) at
37°C. Intracellular acidification (19) was achieved
by supplementing pH 6.5 media for 15 min. with
filtration and incubation under 100% O and white
2
light.
The disappearance and formation of L-2-
HG, α-KG, citrate/isocitrate and other metabolites
was determined by LC-MS/MS analysis of the
cuvet reactions. Metabolites were extracted in
30 mM NH Cl (ACROS organics) and 10 µM 5-
4
(N-Ethyl-N-isopropyl) amiloride (EIPA), followed
by washing and re-incubation in pH 6.5 media
with amiloride alone, for the duration of the
experiment (20 hr). This protocol brought
intracellular pH to 6.8 within 2 hrs. (Figure 1C).
Where appropriate, the α-KGDH inhibitor
ketomethylvalerate (KMV) was present at 20 mM.
Intracellular pH was monitored by fluorescence
microscopy using the ratiometric fluorescent pH
8
0% methanol and then analyzed using reverse
phase chromatography with an ion-paring reagent
in a Shimadzu HPLC coupled to a Thermo
Quantum triple-quadrupole mass spectrometer (7).
Data were analyzed using MzRock machine
learning
tool
kit
(
http://code.google.com/p/mzrock/),
which
indicator
2',7'-Bis-(2-Carboxyethyl)-5-6-
automates analysis of targeted metabolites data
based on chromatographic retention time, whole
molecule mass, collision energy, and resulting
fragment mass. Metabolite concentrations were
determined from standard curves constructed
using known concentrations.
carboxyfluorescein-acetoxymethyl ester (BCECF-
AM, Invitrogen, Eugene OR) (36). Cells were
loaded with 2 µg/ml BCECF-AM) and incubated
for 30 min. at 37°C. Cells were then washed once
and fluorescence microscopy performed using an
Eclipse TE2000-S (Nikon, Avon MA) and data
were analyzed using TILL Photonics Imaging
System Software. For cellular metabolite analysis,
a bolus of 80 % (vol/vol) methanol at -80°C was
added after incubations, and cells serially
extracted with the same, followed by analysis by
LC-MS/MS, as for enzyme assays above.
L-2-HG dehydrogenase activity was
assayed (10) in frozen mitochondria isolated from
male C57BL6/J mouse hearts as previously
described (35). Mitochondria were snap-frozen in
liquid N and stored at -80°C. Mitochondrial
2
protein (0.3 mg/ml) was incubated in 20 mM
HEPES buffer supplemented with 0.85 mM
To create a stable cell line expressing L-2-
MgCl , 0.05 % (vol/vol) Triton X-100, and 1.5
HGDH, the expression vector pTS4 was created
by PCR cloning of the rat L-2-HGDH open
2
mM iodonitrotetrazolium (ACROS organics).
After 5 min., 500 μM L-2-HG was added, and the
reaction monitored spectrophotometrically at 500
reading
frame
into
pcDNA3.1-V5HisC
(Invitrogen, Carlsbad, CA) as a BstEII fragment.
Rat cDNA was amplified using primers F-5'-ACA
CGG TCA CCA TGT GGC CGA CCC TGC GCT
ACG-3' and R-5'-ACA CGG TCA CCT AAC
TTA AAC CTT TGC TGT GCT TCC TC-3'.
HEK293 cells were transfected with pTFS4 using
Lipofectamine 3000 (Invitrogen). Stable clonal
isolates were expanded in DMEM with 10 % heat
inactivated FBS under 1 mg/ml G418 selective
pressure for 10 days and thereafter maintained in
media containing 200μg/ml G418. The presence of
transgenic L-2-HGDH was monitored by
enzymatic assay, and by western blotting for V5
and mCherry tags. pcDNA3.1mCherry was
-
1
-1
nm (ε=19300 cm M ) for 60 min. At the end of
each run, de-novo α-KG formation was confirmed
by LC-MS/MS.
HEK-293 cells (ATCC, Manassas VA)
were maintained in DMEM (Life Technologies,
Grand Island NY) with 25 mM D-glucose, 4 mM
L-glutamine, 0.1 mM sodium pyruvate, 10 % heat-
inactivated FBS (Life Technologies) and 100
µg/ml
penicillin-streptomycin
(Life
Technologies), in 5% CO at 37°C. For passaging,
2
cells at 75-90% confluence were released by 3 min
incubation with 0.25% trypsin-EDTA (Life
Technologies), followed by collection in 5
5

Acid and 2‐Hydroxyglutarate
included with pTFS4 at 1:10 to visualize
transfected cells, and was transfected on its own to
create a negative control (mock transfected) cell
line.
For 2-HG signaling experiments, cells
were incubated in a hypoxic chamber (<0.1%) at
NOTE ADDED IN PROOF
During revision of this manuscript, a
report emerged on LDH as a source of high levels
of 2-HG found in testis (37), including data
showing this phenomenon to be pH sensitive. The
relative role of acid as a 2-HG stimulus in testis
under physiologically relevant conditions is
unclear. We consider this study complimentary to
our current findings which span multiple enzymes
and cell systems, and overall these discoveries
highlight the importance of pH as a driver of
metabolic signaling.
3
7 °C for 20 hrs. For western blot monitoring of
HIF, cell lysates were harvested under hypoxia in
SDS-PAGE sample buffer containing MG132, and
following SDS-PAGE proteins were quantified by
western blot. Antibodies employed were: anti-V5
(
Novex, Carlsbad CA), anti-mCherry (AbCam,
Cambridge MA), anti-Hif-1α (Novus Biologics,
Littleton CO), anti-β-actin (Santa Cruz Biotech’,
Santa Cruz CA). Blots were developed using
HRP-linked secondary antibodies and enhanced
chemiluminescence, with densitometry in ImageJ
software.
ACKNOWLEDGEMENTS
This work was funded by grants from the
US National Institutes of Health to PSB (RO1 HL-
071158) and JM (R01 AI-081773).
All experiments were performed
minimum of 3 times. In purified enzyme studies
Figures 3-5 and Figure 6), “N” represents a single
a
CONFLICT OF INTEREST
The authors declare no conflicts of interest
with regard to the content of this paper.
(
measurement, with multiple measurements made
on each of multiple batches of enzyme (i.e.
technical replicates, with the number of replicates
overall exceeding the number of independent
batches of enzyme / biological replicates). In all
other Figures, “N” refers to independent biological
preparations. Statistical significance between
groups was calculated using one-way or two-way
ANOVA with post-hoc Student’s t-testing.
CONTRIBUTIONS
SMN, JM and XS performed experiments.
PSB, SMN, KWN, JM and DF conceived the
study. PSB, KWN and JM provided funding and
resources. SMN prepared the figures. PSB wrote
the paper, with input and editing by KWN, JM and
DF.
REFERENCES
1
2
3
.
.
.
Warburg, O., Wind, F., and Negelein, E. (1927) The metabolism of tumors in the body. J. Gen.
Physiol. 8, 519-530
Pavlova, N. N. and Thompson, C. B. (2016) The Emerging Hallmarks of Cancer Metabolism. Cell
Metab 23, 27-47
Ward, P. S., Patel, J., Wise, D. R., Abdel-Wahab, O., Bennett, B. D., Coller, H. A., Cross, J. R.,
Fantin, V. R., Hedvat, C. V., Perl, A. E., Rabinowitz, J. D., Carroll, M., Su, S. M., Sharp, K. A.,
Levine, R. L., and Thompson, C. B. (2010) The common feature of leukemia-associated IDH1 and
IDH2 mutations is a neomorphic enzyme activity converting alpha-ketoglutarate to 2-
hydroxyglutarate. Cancer Cell 17, 225-234
4
.
Dang, L., White, D. W., Gross, S., Bennett, B. D., Bittinger, M. A., Driggers, E. M., Fantin, V. R.,
Jang, H. G., Jin, S., Keenan, M. C., Marks, K. M., Prins, R. M., Ward, P. S., Yen, K. E., Liau, L. M.,
Rabinowitz, J. D., Cantley, L. C., Thompson, C. B., Vander Heiden, M. G., and Su, S. M. (2010)
Cancer-associated IDH1 mutations produce 2-hydroxyglutarate. Nature 465, 966
Intlekofer, A. M., Dematteo, R. G., Venneti, S., Finley, L. W., Lu, C., Judkins, A. R., Rustenburg, A.
S., Grinaway, P. B., Chodera, J. D., Cross, J. R., and Thompson, C. B. (2015) Hypoxia Induces
Production of L-2-Hydroxyglutarate. Cell Metab 22, 304-311
5
6
.
.
Oldham, W. M., Clish, C. B., Yang, Y., and Loscalzo, J. (2015) Hypoxia-Mediated Increases in L-2-
hydroxyglutarate Coordinate the Metabolic Response to Reductive Stress. Cell Metab 22, 291-303
6

Acid and 2‐Hydroxyglutarate
7
.
Nadtochiy, S. M., Urciuoli, W., Zhang, J., Schafer, X., Munger, J., and Brookes, P. S. (2015)
Metabolomic profiling of the heart during acute ischemic preconditioning reveals a role for SIRT1 in
rapid cardioprotective metabolic adaptation. J. Mol. Cell Cardiol. 88, 64-72
8
9
1
.
.
Struys, E. A. (2006) D-2-Hydroxyglutaric aciduria: unravelling the biochemical pathway and the
genetic defect. J. Inherit. Metab Dis. 29, 21-29
Van Schaftingen, E., Rzem, R., and Veiga-da-Cunha, M. (2009) L: -2-Hydroxyglutaric aciduria, a
disorder of metabolite repair. J. Inherit. Metab Dis. 32, 135-142
0. Rzem, R., Veiga-da-Cunha, M., Noel, G., Goffette, S., Nassogne, M. C., Tabarki, B., Scholler, C.,
Marquardt, T., Vikkula, M., and Van, S. E. (2004) A gene encoding a putative FAD-dependent L-2-
hydroxyglutarate dehydrogenase is mutated in L-2-hydroxyglutaric aciduria. Proc. Natl. Acad. Sci.
U. S. A 101, 16849-16854
1
1. Rzem, R., Achouri, Y., Marbaix, E., Schakman, O., Wiame, E., Marie, S., Gailly, P., Vincent, M. F.,
Veiga-da-Cunha, M., and Van, S. E. (2015) A mouse model of L-2-hydroxyglutaric aciduria, a
disorder of metabolite repair. PLoS. One. 10, e0119540
1
1
2. Venneti, S. and Thompson, C. B. (2013) Metabolic modulation of epigenetics in gliomas. Brain
Pathol. 23, 217-221
3. Jin, Y., Elalaf, H., Watanabe, M., Tamaki, S., Hineno, S., Matsunaga, K., Woltjen, K., Kobayashi,
Y., Nagata, S., Ikeya, M., Kato, T., Jr., Okamoto, T., Matsuda, S., and Toguchida, J. (2015) Mutant
IDH1 Dysregulates the Differentiation of Mesenchymal Stem Cells in Association with Gene-
Specific Histone Modifications to Cartilage- and Bone-Related Genes. PLoS. One. 10, e0131998
4. Su, X., Wellen, K. E., and Rabinowitz, J. D. (2016) Metabolic control of methylation and
acetylation. Curr. Opin. Chem. Biol. 30, 52-60
5. Koivunen, P., Lee, S., Duncan, C. G., Lopez, G., Lu, G., Ramkissoon, S., Losman, J. A., Joensuu, P.,
Bergmann, U., Gross, S., Travins, J., Weiss, S., Looper, R., Ligon, K. L., Verhaak, R. G., Yan, H.,
and Kaelin, W. G., Jr. (2012) Transformation by the (R)-enantiomer of 2-hydroxyglutarate linked to
EGLN activation. Nature 483, 484-488
1
1
1
6. Xu, W., Yang, H., Liu, Y., Yang, Y., Wang, P., Kim, S. H., Ito, S., Yang, C., Wang, P., Xiao, M. T.,
Liu, L. X., Jiang, W. Q., Liu, J., Zhang, J. Y., Wang, B., Frye, S., Zhang, Y., Xu, Y. H., Lei, Q. Y.,
Guan, K. L., Zhao, S. M., and Xiong, Y. (2011) Oncometabolite 2-hydroxyglutarate is a competitive
inhibitor of alpha-ketoglutarate-dependent dioxygenases. Cancer Cell 19, 17-30
1
1
7. Losman, J. A. and Kaelin, W. G., Jr. (2013) What a difference a hydroxyl makes: mutant IDH, (R)-
2
-hydroxyglutarate, and cancer. Genes Dev. 27, 836-852
8. Wang, P., Wu, J., Ma, S., Zhang, L., Yao, J., Hoadley, K. A., Wilkerson, M. D., Perou, C. M., Guan,
K. L., Ye, D., and Xiong, Y. (2015) Oncometabolite D-2-Hydroxyglutarate Inhibits ALKBH DNA
Repair Enzymes and Sensitizes IDH Mutant Cells to Alkylating Agents. Cell Rep. 13, 2353-2361
9. Sandvig, K., Olsnes, S., Petersen, O. W., and van Deurs, B. (1987) Acidification of the cytosol
inhibits endocytosis from coated pits. J. Cell Biol. 105, 679-689
0. Gasbarrini, A., Borle, A. B., Farghali, H., Bender, C., Francavilla, A., and Van, T. D. (1992) Effect
of anoxia on intracellular ATP, Na+i, Ca2+i, Mg2+i, and cytotoxicity in rat hepatocytes. J. Biol.
Chem. 267, 6654-6663
1
2
2
2
2
1. Ladilov, Y. V., Balser, C., and Piper, H. M. (1998) Protection of rat cardiomyocytes against
simulated ischemia and reoxygenation by treatment with protein kinase C activator. Circ. Res. 82,
4
51-457
2. Rose, U. M., Abrahamse, S. L., Jansen, J. W., Bindels, R. J., and van Os, C. H. (1995) Cellular
acidification occurs during anoxia in cultured, but not in freshly isolated, rabbit proximal tubular
cells. Pflugers Arch. 429, 722-728
3. Wise, D. R., Ward, P. S., Shay, J. E., Cross, J. R., Gruber, J. J., Sachdeva, U. M., Platt, J. M.,
Dematteo, R. G., Simon, M. C., and Thompson, C. B. (2011) Hypoxia promotes isocitrate
dehydrogenase-dependent carboxylation of alpha-ketoglutarate to citrate to support cell growth and
viability. Proc. Natl. Acad. Sci. U. S. A 108, 19611-19616
7

Acid and 2‐Hydroxyglutarate
2
2
2
4. Read, J. A., Winter, V. J., Eszes, C. M., Sessions, R. B., and Brady, R. L. (2001) Structural basis for
altered activity of M- and H-isozyme forms of human lactate dehydrogenase. Proteins 43, 175-185
5. Butler, J. A., Mishur, R. J., Bhaskaran, S., and Rea, S. L. (2013) A metabolic signature for long life
in the Caenorhabditis elegans Mit mutants. Aging Cell 12, 130-138
6. Thompson, L. J., Strittmatter, L., Tardif, J., Sharma, R., Tremblay-Vaillancourt, V., Aubut, C.,
Boucher, G., Clish, C. B., Cyr, D., Daneault, C., Waters, P. J., Vachon, L., Morin, C., Laprise, C.,
Rioux, J. D., Mootha, V. K., and Des, R. C. (2015) A Metabolic Signature of Mitochondrial
Dysfunction Revealed through a Monogenic Form of Leigh Syndrome. Cell Rep. 13, 981-989
7. Mekhail, K., Gunaratnam, L., Bonicalzi, M. E., and Lee, S. (2004) HIF activation by pH-dependent
nucleolar sequestration of VHL. Nat. Cell Biol. 6, 642-647
8. Chiche, J., Brahimi-Horn, M. C., and Pouyssegur, J. (2010) Tumour hypoxia induces a metabolic
shift causing acidosis: a common feature in cancer. J. Cell Mol. Med. 14, 771-794
9. Webb, B. A., Chimenti, M., Jacobson, M. P., and Barber, D. L. (2011) Dysregulated pH: a perfect
storm for cancer progression. Nat. Rev. Cancer 11, 671-677
0. Hjelmeland, A. B., Wu, Q., Heddleston, J. M., Choudhary, G. S., MacSwords, J., Lathia, J. D.,
McLendon, R., Lindner, D., Sloan, A., and Rich, J. N. (2011) Acidic stress promotes a glioma stem
cell phenotype. Cell Death. Differ. 18, 829-840
2
2
2
3
3
1. De Los, A. A., Ferrari, F., Fujiwara, Y., Mathieu, R., Lee, S., Lee, S., Tu, H. C., Ross, S., Chou, S.,
Nguyen, M., Wu, Z., Theunissen, T. W., Powell, B. E., Imsoonthornruksa, S., Chen, J., Borkent, M.,
Krupalnik, V., Lujan, E., Wernig, M., Hanna, J. H., Hochedlinger, K., Pei, D., Jaenisch, R., Deng,
H., Orkin, S. H., Park, P. J., and Daley, G. Q. (2015) Failure to replicate the STAP cell phenomenon.
Nature 525, E6-E9
3
3
3
3
3
2. Barth, P. G., Wanders, R. J., Scholte, H. R., Abeling, N., Jakobs, C., Schutgens, R. B., and Vreken,
P. (1998) L-2-hydroxyglutaric aciduria and lactic acidosis. J. Inherit. Metab Dis. 21, 251-254
3. Stacpoole, P. W., Nagaraja, N. V., and Hutson, A. D. (2003) Efficacy of dichloroacetate as a lactate-
lowering drug. J. Clin. Pharmacol. 43, 683-691
4. Hoffman, D. L. and Brookes, P. S. (2009) Oxygen sensitivity of mitochondrial reactive oxygen
species generation depends on metabolic conditions. J. Biol. Chem. 284, 16236-16245
5. Nadtochiy, S. M., Madukwe, J., Hagen, F., and Brookes, P. S. (2014) Mitochondrially targeted nitro-
linoleate: a new tool for the study of cardioprotection. Br. J. Pharmacol. 171, 2091-2098
6. Thomas, J. A., Buchsbaum, R. N., Zimniak, A., and Racker, E. (1979) Intracellular pH
measurements in Ehrlich ascites tumor cells utilizing spectroscopic probes generated in situ.
Biochemistry 18, 2210-2218
3
7. Teng, X., Emmett, M. J., Lazar, M. A., Goldberg, E., and Rabinowitz, J. D. (2016) Lactate
Dehydrogenase C Produces S-2-Hydroxyglutarate in Mouse Testis. ACS Chem. Biol. In Press.
8

Acid and 2‐Hydroxyglutarate
Figure 1. Acidic pH drives 2-HG production in cells. HEK293 cells were maintained at pH 7.4, or
incubated under conditions to acidify the cytosol (see methods). (A/B): Pseudocolored 490nm/440nm
dual excitation ratiometric images of BCECF fluorescence, showing intracellular pH for cells maintained
at pH 7.4 (panel A), or pH 6.5 with EIPA (panel B). (C): Quantitative time course of cellular
acidification. Data are means ± SD, N=3. (D): Cellular 2-HG levels measured by LC-MS/MS following
incubation at different cellular pH levels in the absence or presence of the α-KGDH inhibitor KMV. (E):
Cellular α-KG levels measured by LC-MS/MS following incubation at different cellular pH levels in the
#
absence or presence of KMV. Data are means ± SEM, N=6. Different symbols above bars (* †) indicate
statistically significant differences (two-way ANOVA and Student’s t-test) between groups – i.e. bars
with the same symbol are not different.
9

Acid and 2‐Hydroxyglutarate
Figure 2. Enzyme purity and Assay Validation. (A): SDS-PAGE analysis of commercially obtained
LDH (Sigma L3916), MDH (Sigma M2634) and ICDH (Sigma #I2002). Following electrophoresis on
1
2.5% SDS-PAGE, gels were transferred to nitrocellulose membranes and stained with Ponceau S.
Numbers to the left of images show molecular weight markers, in kDa. Data are shown for 2 separate
batches of material (2 lanes). Arrows to the right of images show expected position of band, for published
molecular weight of the protein. (B): Comparison of NADH spectrophotometric assay vs. 2-HG LC-
MS/MS assay, for the 2-HG generating activity of LDH (i.e., data from Figures 3A and 3B of main
manuscript). Reaction is shown in the inset, with red/blue colored parameters matching appropriate axes
on the graph. Data are means ± SEM. Dotted line shows linear regression curve fit with correlation
coefficient shown alongside inset. (C): Comparison of NADH spectrophotometric assay vs. 2-HG LC-
MS/MS assay, for the 2-HG generating activity of MDH (i.e., data from Figures 3D and 3E of main
manuscript). Reaction is shown in the inset, with red/blue colored parameters matching appropriate axes
on the graph. Data are means ± SEM. Dotted line shows linear regression curve fit with correlation
coefficient shown alongside inset. (D): Comparison of Formazan spectrophotometric assay vs. α-KG LC-
MS/MS assay, for the L-2-HG consuming activity of L-2-HGDH (i.e., data from Figures 6A and 6B of
main manuscript). Reaction is shown in the inset, with red/blue colored parameters matching appropriate
axes on the graph. Data are means ± SEM, with N for each measure corresponding to the parent data sets
in Figures 2 and 3 of the main manuscript. Dotted line shows linear regression curve fit with correlation
coefficient shown alongside inset.
1
0

Acid and 2‐Hydroxyglutarate
Figure 3. pH dependency of 2-HG generation by lactic or malic dehydrogenases. The 2-HG synthetic
activity of isolated LDH was assayed spectrophotometrically as NADH consumption (panel A) or by
direct LC-MS/MS assay of 2-HG formation (panel B). Insets to each panel show reaction schemes, with
the measured parameter (y-axis of graph) highlighted in red. (A): NADH consumption at various pH
values, for LDH in the presence of 1 mM (white circles) or 10 mM (black circles) α-KG. (B): 2-HG
production at various pH values, for LDH in the presence of 1 mM α-KG. (C): Native pyruvate to lactate
converting activity of LDH at various pH values, assayed spectrophotometrically as NADH consumption.
Similar experiments were also performed with isolated MDH. (D): NADH consumption at various pH
1
1

Acid and 2‐Hydroxyglutarate
values, for MDH in the presence of 1 mM (white symbols) or 10 mM (black symbols) α-KG. (E): 2-HG
production at various pH values, for MDH in the presence of 10 mM α-KG. (Note – with MDH, the high
concentration of α-KG was necessary due to the low sensitivity of LC-MS/MS based 2-HG assay). (F):
Native MDH activity (oxaloacetate to malate) at various pH values, assayed spectrophotometrically as
NADH consumption. (Note – MDH activity was measured in the reverse direction, i.e., oxaloacetate to
malate, due to thermodynamic constraints as detailed in the supplemental methods). All data are means ±
SEM, N>3. *p<0.05 (two-way ANOVA followed by Student’s t-test) compared to corresponding value at
pH 7.4. Where error bars appear absent, they are sufficiently small as to be wholly contained within the
data symbols.
1
2

Acid and 2‐Hydroxyglutarate
Figure 4. Effect of Hypoxia on 2-HG Generation
by LDH. Isolated LDH was incubated in an open
flow spectrophotometer cell, as detailed in the
methods. A parallel incubation contained 100 μM
hemoglobin, to confirm deoxygenation of the sample.
(A): Absorbance spectra of purified Hb under
normoxic (gray line) or hypoxic (black line)
conditions, the latter comprising a 1 hr. of argon
purge. (B): NADH consumption by LDH under
normoxic or hypoxic conditions. Data are expressed
normalized to the normoxia values, and are means ±
SEM. (C): 2-HG generation by LDH under normoxic
or hypoxic conditions. Samples were removed from
the same incubations as panel B and rapidly
deproteinized for LC-MS/MS analysis of 2-HG levels.
Data are expressed normalized to the normoxia
values, and are means ± SEM.
1
3

Acid and 2‐Hydroxyglutarate
Figure 5. pH dependency of isocitrate
dehydrogenase activity. The ability of
commercial native ICDH (Figure 2A) to catalyze
the generation of 2-HG from α-KG was
monitored spectrophotometrically as NADPH
consumption (panel A) or by direct LC-MS/MS
assay of 2-HG formation (panel B). Insets to each
panel show reaction schemes, with the measured
parameter (y-axis of graph) highlighted in red.
(
A): NADPH consumption at various pH values,
for ICDH in the presence of 10 mM α-KG. (B):
-HG production at various pH values, for ICDH
2
in the presence of 10 mM α-KG. Although
NADPH was consumed, 2-HG was not generated.
(C): Reverse activity of ICDH to generate
isocitrate from α-KG, monitored by LC-MS/MS
assay of isocitrate formation (see panel inset for
reaction scheme). Data are means ± SEM. N=7-8.
*
p<0.05 (two-way ANOVA followed by
Student’s t-test) compared to corresponding value
at pH 7.4.
1
4

Acid and 2‐Hydroxyglutarate
Figure 6. pH dependency of L-2-HG dehydrogenase (L-2-HGDH). The L-2-HGDH activity of
isolated permeabilized mouse heart mitochondria was measured spectrophotometrically using either an
iodonitrotetrazolium (INT) to formazan coupled assay (panel A) or by direct LC-MS/MS assay of α-KG
formation (panel B). (A): Formazan production at various pH values, for mitochondria supplied with 500
μM L-2-HG. (B): α-KG production at various pH values, for mitochondria supplied with 500 μM L-2-
HG. Data are means ± SEM, N=4. *p<0.05 (two-way ANOVA followed by Student’s t-test) compared to
corresponding value at pH 7.4.
1
5

Acid and 2‐Hydroxyglutarate
Figure 7. Acid induced substrate promiscuity in α-keto acid dehydrogenases. (A): Generalized
reaction scheme for α-keto acid dehydrogenases, with the table showing specific examples and the
corresponding –R group for each substrate/product pair. (B): The ability of LDH to catalyze the native
back reaction of MDH (i.e. OAA to malate) was measured at various pH values, assayed
spectrophotometrically as NADH consumption. (C): Generation of malate in this system was also assayed
by LC-MS/MS. Data are means ± SEM, N=4. *p<0.05 (two-way ANOVA followed by Student’s t-test)
compared to corresponding value at pH 7.4.
1
6

Acid and 2‐Hydroxyglutarate
Figure 8. Characterization of L-2-HGDH transgenic cells. HEK293 cells were transfected with mock
(
mCherry alone) or V5:L-2-HGDH plus mCherry containing vectors, and stable clones selected as
described in the methods. (A): Monitoring of successful transfection by western blotting for V5 and
mCherry. Clone #2 exhibited highest levels of transfection. Lower panel shows Ponceau S stained
membrane indicating equal protein loading. (B): L-2-HGDH enzymatic activity was measured by the
Formazan assay as described in the methods (see also Figure 3 main manuscript). Graph shows calculated
rates, with inset showing raw spectrophotometric traces. In agreement with panel A, clone #2 showed
highest L-2-HGDH activity. (C): Quantitation of 2-HG levels in mock or L-2-HGDH transgenic cells, in
response to acid plus KMV exposure as in Figure 1. All data in this Figure are N=1, representing merely
rationale for choice of clone #2 for subsequent cell signaling experiments.
1
7

Acid and 2‐Hydroxyglutarate
Figure 9. Requirement of 2-HG for acid-induced HIF induction. Mock and L-2-HGDH transfected
cells (see Figure 8) were exposed to acid plus KMV as in Figure 1, or to hypoxia, followed by western
blotting for HIF-1α. (A): Representative HIF-1α western blots at high exposure (upper panel) or low
exposure (center panel), with corresponding Ponceau S stained membrane to indicate protein loading
(
lower panel). Arrows indicate position of HIF-1α at expected molecular weight. (B): Quantitative
densitometry from 4 independent experiments of the type shown in panel A. For normoxia and acid plus
KMV conditions, high exposure blots were used. However, due to saturation of the HIF signal, low
exposure blots were used for the hypoxia condition. Data are means ± SEM, N=4. *p<0.05 (paired
Student’s t-test) between mock and L-2-HGDH-Tg cells under the same condition.
1
8

Acidic pH is a Metabolic Switch for 2-Hydroxyglutarate Generation and Signaling
Sergiy M. Nadtochiy, Xenia Schafer, Dragony Fu, Keith Nehrke, Joshua Munger and Paul
S. Brookes
J. Biol. Chem. published online August 10, 2016
Access the most updated version of this article at doi: 10.1074/jbc.M116.738799
Alerts:
•
•
When this article is cited
When a correction for this article is posted
Click here to choose from all of JBC's e-mail alerts
This article cites 0 references, 0 of which can be accessed free at
http://www.jbc.org/content/early/2016/08/10/jbc.M116.738799.full.html#ref-list-1