Journal of General Virology (2015), 96, 2693–2696
Short
Communication
DOI 10.1099/jgv.0.000226
Semliki Forest virus and Sindbis virus, but not
vaccinia virus, require glycolysis for optimal
replication
James S. Findlay and David Ulaeto
Correspondence
CBR Division, Dstl Porton Down, Salisbury SP4 0JQ, UK
James S. Findlay
[email protected]
Received 25 March 2015
Accepted 22 June 2015
Viruses are obligate intracellular pathogens which rely on the cell’s machinery to produce the
energy and macromolecules required for replication. Infection is associated with a modified
metabolic profile and one pathway which can be modified is glycolysis. In this study, we
investigated if the glycolysis pathway is required for alphavirus replication. Pre-treatment of Vero
cells with three different glycolysis inhibitors (2-deoxyglucose, lonidamine and oxamate) resulted
in a significant reduction (but not abrogation) of Semliki Forest virus and Sindbis virus
replication, but not of the unrelated virus, vaccinia virus. Reduced virus yield was not associated
with any significant cytotoxic effect and delayed treatment up to 3 h post-infection still resulted
in a significant reduction. This suggested that glycolysis is required for optimal replication of
alphaviruses by supporting post-entry life cycle steps.
Viruses are obligate intracellular pathogens that rely on the
cell’s machinery to produce the energy and macromolecules
required for replication. As viruses are known to activate/
suppress transcription of host genes, it is unsurprising that
some viruses alter the metabolic state of the cell. Studies
have found that a range of viruses, including hepatitis C
virus, influenza virus, vaccinia virus and human cytomegalovirus, all alter the metabolic state of cells they infect
(Diamond et al., 2010; Fontaine et al., 2014; Munger et al.,
2006; Ritter et al., 2010). Under aerobic conditions, cells
produce ATP, NADH and certain amino acid precursors
from the tricarboxylic acid (TCA) cycle, which relies on
acetyl-CoA. One of the major sources of acetyl-CoA is
glycolysis, where the end-product pyruvate is decarboxylated to produce acetyl-CoA. As well as producing pyruvate
to support the TCA cycle, glycolysis also produces two ATPs
and NADHs from each glucose molecule. It has been
reported that glycolysis is required for optimal replication
of dengue virus and that disruption of glycolysis can prevent
influenza A(H1N1) virus entry into cells (Fontaine et al.,
2015; Kohio & Adamson, 2013). Therefore, glycolysis appears
to be important for multiple steps in virus life cycles.
Alphaviruses are zoonotic arboviruses that can infect
humans and other vertebrate hosts. Pathogenic members
of this group include Venezuelan equine encephalitis virus
(VEEV), western equine encephalitis virus, eastern equine
encephalitis virus and chikungunya virus (CHIKV). Epizootic strains of VEEV cause sporadic outbreaks of encephalitis
in humans in the Americas (Go et al., 2014). In addition,
these alphaviruses are listed as Category B agents by the
US Centers for Disease Control and Prevention. CHIKV
causes severe joint pain and fever in humans, and has
000226 G 2015 The Authors
historically been present in Asia, Africa, and has recently
spread to Europe, the Caribbean, and Central and South
America (Morrison, 2014; Rezza et al., 2007). It has been
reported that infection by two alphaviruses, Mayaro virus
and Sindbis virus (SINV), results in increased glucose
uptake and glycolytic flux in cell lines (El-Bacha et al.,
2004; Silva da Costa et al., 2012). This suggests that glycolysis
may be important in the alphavirus life cycle. If so, this may
open up the use of glycolysis inhibitors that have been
designed for cancer treatment as potential antivirals for
alphavirus infections. In this study, we investigated if the glycolysis pathway is required for alphavirus replication and if
disruption of glycolysis could be an antiviral strategy.
Vero cells (African Green monkey kidney cells) were incubated for 16 h in minimal Dulbecco’s modified Eagle’s
medium (lacking glutamine and pyruvate, and supplemented
with 1 mg glucose ml21 and 2 % FCS) either in the absence
or presence of two concentrations of glycolysis inhibitors. All
incubations were performed at 37 8C in a 5 % CO2, humidified atmosphere. The experiments were performed in the
absence of glutamine and pyruvate as these can be used to
produce acetyl-CoA, and therefore could overcome the
imposed block by glycolysis inhibitors. Three glycolysis
inhibitors (Sigma-Aldrich) were used: 2-deoxyglucose
(2-DG), an analogue of glucose, which is converted by
hexokinase II to an intermediate that cannot be further
used in glycolysis, was used at 2 and 10 mM; lonidamine
(LON), a hexokinase II inhibitor, which prevents the first
step of glycolysis, was used at 62.5 and 312.5 mM; and
oxamate (OX), a lactate dehydrogenase (LDH) inhibitor,
which prevents the recycling of NADH, was used at 16 and
80 mM. The concentrations of inhibitors used were based
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 18:23:37
Printed in Great Britain
2693
J. S. Findlay and D. Ulaeto
(a)
SFV
Virus titre (TCID50 ml–1)
108
107
*
***
106
105
104
*
103
102
***
0
2
10
2-DG (mM)
**
0
(b)
62.5 312.5
LON (µM)
0
16
80
OX (mM)
SINV
108
Virus titre (TCID50 ml–1)
on their reported inhibitory concentrations of glycolysis in
cancer cells (Elwood, 1968; Hulleman et al., 2009). After
16 h incubation, the cells were infected with Semliki Forest
virus (SFV) at m.o.i. 3 TCID50 per cell. Cultures were incubated for 24 h post-infection (p.i.) to ensure maximum yield
of progeny in untreated cultures, after which the released
virus was titrated by TCID50 assay (Butcher & Ulaeto,
2005). All three inhibitors resulted in a significant decrease
in yield, with 2-DG and OX causing .3 log drop in
virus yield, whereas LON caused a 1 log drop (Fig. 1a).
To determine if this was specific to SFV, these experiments
were repeated using SINV at m.o.i. 3 (Fig. 1b). Treatment
of the cells with all three inhibitors significantly reduced
SINV yield at the highest concentration, with 2-DG and
OX causing a 2 log drop and LON causing a 1 log
drop. To determine if this was specific to alphaviruses, the
experiment was repeated using 2-DG and an unrelated
virus, the orthopoxvirus vaccinia virus (VACV). Treatment
of the cells with 10 mM 2-DG did not result in a significant
decrease in VACV replication (Fig. 1c). Therefore, a requirement for glycolytic ATP production (to be used for energy
generation or as substrate for polynucleotide synthesis) is
not a universal feature of virus replicative strategies.
107
*
**
106
**
105
104
103
102
0
2
10
2-DG (mM)
0
62.5 312.5
LON (µM)
0
16
80
OX (mM)
In order to exclude drug cytotoxicity as the mediator of
inhibition, viability assays were undertaken with the three
compounds. Uninfected cells were incubated with either
10 mM 2-DG, 312.5 mM LON or 80 mM OX for 40 h
(the timespan between addition of inhibitor and harvesting
of virus) and the culture supernatant tested for LDH
activity. Treatment with the glycolysis inhibitors did not
result in increased LDH release relative to untreated controls and in fact appeared to reduce LDH release from
the cultures (Fig. 2a). However, OX is a LDH inhibitor,
which could thus confound the LDH assay for this drug.
Consequently the experiment was repeated using the highest concentration of inhibitors, and the cells were analysed
for membrane integrity by incubation with SYTOX Orange
membrane-impermeable dye according to the manufacturer’s instructions and analysis on a FACSCanto flow cytometer (Fig. 2b). Despite a twofold increase in membrane
permeability for cells treated with OX, the inhibitors did
not result in a statistically significant increase in the percentage of cells with damaged membranes compared with
the control. Therefore, the reduction in virus yield was
not due to cellular toxicity of the drugs.
Fig. 1. Glycolysis inhibitors reduce alphavirus but not orthopoxvirus yield. Vero cells pre-treated with 2-DG, LON and OX were
then infected with (a) SFV, (b) SINV or (c) VACV. At 24 h p.i., the
released virus was titrated by TCID50. The data were analysed
using a paired t-test (with Bonferroni’s post-test correction where
appropriate). *P,0.05; **P,0.01; ***P,0.001.
It has been reported that inhibition of glycolysis with 2-DG
can prevent influenza A virus entry by a mechanism
involving prevention of endosomal acidification (Kohio &
Adamson, 2013). To examine if disruption of glycolysis
was affecting the entry of SFV and SINV, Vero cells were
either: pre-treated with 2-DG for 16 h and then infected by
SFV or SINV; treated with 2-DG at the same time as infection
(T0); or infected and then treated with 2-DG at 3 h p.i. (T3).
At each treatment time point, there was a significant
reduction in both SFV and SINV yield compared with controls (Fig. 3). Although no SFV was detected in these experiments, the assay used has a limit of detection of
16102 TCID50 ml21. Consequently progeny virus may
have been produced at ,16102 TCID50 ml21 in these
samples. Nonetheless, addition of 2-DG 3 h after infection
clearly resulted in a significant reduction, indicating that
2-DG-mediated inhibition of replication appears to operate
after the phase entry for SFV and SINV. In these experiments,
the inhibitor was added to wells without prior removal of the
virus inoculum. Although it is possible that the inhibitor was
acting to prevent binding or entry of virus after the 3 h incubation period, the published kinetics of SFV and SINV
2694
(c)
VACV
Virus titre (TCID50 ml–1)
107
106
105
104
103
102
0
10
2-DG (mM)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 18:23:37
Journal of General Virology 96
Requirement of glycolysis for alphavirus replication
SFV
(a)
1.0
0.8
0.6
**
0.4
**
**
0.2
0
10
2-DG (mM)
0
312.5
LON (µM)
0
80
OX (mM)
No. of cells with membrane
damage relative to control
(b)
Virus titre (TCID50 ml–1)
Relative LDH levels
1.2
0
SINV
108
107
106
***
105
***
***
104
103
102
***
No Pre-treat
inhibitor
***
***
T0
T3
No Pre-treat
inhibitor
T0
T3
Fig. 3. Glycolysis is required for a post-entry process. Vero cells
were either: pre-treated with 2-DG for 16 h and then infected by
SFV or SINV (Pre-treat); treated with 2-DG at the same time as
infection by SFV or SINV (T0); or infected by SFV or SINV and
then 3 h p.i. treated with 2-DG (T3). At T24, released virus was
titrated by TCID50. The data were analysed by using a paired ttest with Bonferroni’s post-test correction ***P,0.001.
4
3
2
1
0
0
10
2-DG (mM)
0
312.5
LON (µM)
0
80
OX (mM)
Fig. 2. Glycolysis inhibitors do not affect cell viability. Vero cells
were incubated with each inhibitor for 40 h and then the effect
on cell viability quantified by (a) released LDH levels (using a
LDH-Cytotoxicity Assay kit; Abcam) or (b) flow cytometry (using
SYTOX Orange membrane-impermeable dye; Invitrogen). The
data were analysed compared with the control (no inhibitor) using
a one-sample t-test. **P,0.01.
binding and entry indicate that the majority of virus is
expected to have entered the cell after 3 h incubation at
37 8C at the m.o.i. used in this study (Byrnes & Griffin,
1998; Helenius et al., 1980; Jan et al., 2000).
Viruses are dependent on cell metabolism for replication
and infection by a range of viruses, including alphaviruses,
is associated with an increase in glycolysis/glycolytic flux.
Pre-treatment of Vero cells with three different glycolysis
inhibitors, targeting different steps/enzymes in the glycolysis pathway, resulted in a significant reduction of released
SFV and SINV (Fig. 1). Reduced virus yield was not associated with any significant cytotoxic effect (Fig. 2) and
delayed treatment up to 3 h p.i. still resulted in a significant
reduction that was, however, not significantly different
from pre-treatment regimens (Fig. 3). This suggests that
glycolysis is required for a post-entry step, unlike the situation with influenza A (Kohio & Adamson, 2013). Treatment with the inhibitors reduced, but did not completely
abrogate, replication of SFV and SINV. Therefore, glycolysis appears to be required for optimal SFV and SINV replication. Whilst it is not clear precisely what step(s) in virus
http://vir.sgmjournals.org
replication the glycolysis inhibitors is disrupting, it is likely
to be one with a significant energy requirement. However,
treatment with 2-DG did not reduce VACV replication
(Fig. 1c). This supports a recent finding that glycolysis
was not induced in VACV-infected cells and that the
removal of glutamine, but not glucose, from culture
medium resulted in reduced replication (Fontaine et al.,
2014). As the same cell line was used for each virus, this
could suggest that the inhibitors are affecting biosynthetic
and/or bioenergetic pathways that are required for alphaviruses, but not for orthopoxviruses. Further work is
required to examine this issue.
Interestingly, 2-DG and OX had a more profound effect than
LON on alphaviruses (Fig. 1). As these inhibitors all target the
glycolysis pathway, either LON is less efficient at inhibition of
glycolysis than the other two or 2-DG and OX are exhibiting
‘off-target’ effects. LON targets hexokinase II, an enzyme
which has been shown to be upregulated by several viruses,
including dengue virus (Fontaine et al., 2015). Upregulation
could offset the effect of LON. Furthermore, 2-DG, as well
as being a glycolysis inhibitor, can affect protein glycosylation
and activate the unfolded protein response (Kurtoglu et al.,
2007; Xi et al., 2011). As alphaviruses rely on glycoproteins
for entry and the endoplasmic reticulum stress response can
inhibit translation, these off-target activities could explain
the increased antiviral activity of 2-DG. As a LDH inhibitor,
OX leads to the accumulation of lactate, which is known to
cause a detrimental reduction in intracellular pH and reduced
cell metabolism (Patel et al., 2000). Therefore, OX could also
have a secondary effect that might potentiate direct glycolysisassociated inhibition of virus replication. Whilst the glycolysis
inhibitors affected both SFV and SINV, SINV was less sensitive
to these inhibitors than SFV (Fig. 1). It seems likely that this
difference must have a genetic basis, e.g. a species-specific
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 18:23:37
2695
J. S. Findlay and D. Ulaeto
interaction with a cellular pathway. In addition, phylogenetic
studies divide the alphaviruses into three clades, with SINV in
a separate group to SFV (Levinson et al., 1990). Therefore,
further studies using a range of alphaviruses would be
required to determine if the differential inhibition is based
on the phylogenetic distinction between the different clades
of alphaviruses. Such analysis, coupled with mutation studies,
could provide additional insight into the strategies employed
by viruses to meet their metabolic requirements in host cells.
Inhibition of glycolysis modulates prednisolone resistance in acute
lymphoblastic leukemia cells. Blood 113, 2014–2021.
Jan, J. T., Chatterjee, S. & Griffin, D. E. (2000). Sindbis virus entry into
cells triggers apoptosis by activating sphingomyelinase, leading to the
release of ceramide. J Virol 74, 6425–6432.
Kohio, H. P. & Adamson, A. L. (2013). Glycolytic control of
vacuolar-type ATPase activity: a mechanism to regulate influenza
viral infection. Virology 444, 301–309.
Kurtoglu, M., Gao, N., Shang, J., Maher, J. C., Lehrman, M. A.,
Wangpaichitr, M., Savaraj, N., Lane, A. N. & Lampidis, T. J. (2007).
Under normoxia, 2-deoxy-D -glucose elicits cell death in select
tumor types not by inhibition of glycolysis but by interfering with
N-linked glycosylation. Mol Cancer Ther 6, 3049–3058.
References
Butcher, W. & Ulaeto, D. (2005). Contact inactivation of orthopoxviruses by household disinfectants. J Appl Microbiol 99, 279–284.
Levinson, R. S., Strauss, J. H. & Strauss, E. G. (1990). Complete
Byrnes, A. P. & Griffin, D. E. (1998). Binding of Sindbis virus to cell
sequence of the genomic RNA of O’nyong-nyong virus and its use in
the construction of alphavirus phylogenetic trees. Virology 175,
110–123.
Diamond, D. L., Syder, A. J., Jacobs, J. M., Sorensen, C. M., Walters,
K. A., Proll, S. C., McDermott, J. E., Gritsenko, M. A., Zhang, Q. &
other authors (2010). Temporal proteome and lipidome profiles
11644–11647.
surface heparan sulfate. J Virol 72, 7349–7356.
reveal hepatitis C virus-associated reprogramming of hepatocellular
metabolism and bioenergetics. PLoS Pathog 6.
El-Bacha, T., Menezes, M. M., Azevedo e Silva, M. C., Sola-Penna, M.
& Da Poian, A. T. (2004). Mayaro virus infection alters glucose
Morrison, T. E. (2014). Reemergence of chikungunya virus. J Virol 88,
Munger, J., Bajad, S. U., Coller, H. A., Shenk, T. & Rabinowitz, J. D.
(2006). Dynamics of the cellular metabolome during human
cytomegalovirus infection. PLoS Pathog 2, e132.
Patel, S. D., Papoutsakis, E. T., Winter, J. N. & Miller, W. M. (2000). The
metabolism in cultured cells through activation of the enzyme 6phosphofructo 1-kinase. Mol Cell Biochem 266, 191–198.
lactate issue revisited: novel feeding protocols to examine inhibition of
cell proliferation and glucose metabolism in hematopoietic cell
cultures. Biotechnol Prog 16, 885–892.
Elwood, J. C. (1968). Effect of oxamate on glycolysis and respiration
in sarcoma 37 ascites cells. Cancer Res 28, 2056–2060.
Rezza, G., Nicoletti, L., Angelini, R., Romi, R., Finarelli, A. C., Panning,
M., Cordioli, P., Fortuna, C., Boros, S. & other authors (2007).
Fontaine, K. A., Camarda, R. & Lagunoff, M. (2014). Vaccinia virus
Infection with chikungunya virus in Italy: an outbreak in a
temperate region. Lancet 370, 1840–1846.
requires glutamine but not glucose for efficient replication. J Virol
88, 4366–4374.
Fontaine, K. A., Sanchez, E. L., Camarda, R. & Lagunoff, M. (2015).
Dengue virus induces and requires glycolysis for optimal replication.
J Virol 89, 2358–2366.
Go, Y. Y., Balasuriya, U. B. & Lee, C. K. (2014). Zoonotic encephalitides
caused by arboviruses: transmission and epidemiology of alphaviruses
and flaviviruses. Clin Exp Vaccine Res 3, 58–77.
Ritter, J. B., Wahl, A. S., Freund, S., Genzel, Y. & Reichl, U. (2010).
Metabolic effects of influenza virus infection in cultured animal
cells: intra- and extracellular metabolite profiling. BMC Syst Biol 4, 61.
Silva da Costa, L., Pereira da Silva, A. P., Da Poian, A. T. & El-Bacha, T.
(2012). Mitochondrial bioenergetic alterations in mouse neuroblastoma
cells infected with Sindbis virus: implications to viral replication and
neuronal death. PLoS One 7.
Helenius, A., Kartenbeck, J., Simons, K. & Fries, E. (1980). On the
entry of Semliki forest virus into BHK-21 cells. J Cell Biol 84, 404–420.
Xi, H., Kurtoglu, M., Liu, H., Wangpaichitr, M., You, M., Liu, X., Savaraj,
N. & Lampidis, T. J. (2011). 2-Deoxy-D -glucose activates autophagy
Hulleman, E., Kazemier, K. M., Holleman, A., VanderWeele, D. J., Rudin,
C. M., Broekhuis, M. J., Evans, W. E., Pieters, R. & Den Boer, M. L. (2009).
via endoplasmic reticulum stress rather than ATP depletion. Cancer
Chemother Pharmacol 67, 899–910.
2696
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Wed, 14 Jun 2017 18:23:37
Journal of General Virology 96