Dengue virus induces and requires glycolysis for optimal replication

Abstract

Viruses rely on host cellular metabolism to provide the energy and biosynthetic building blocks required for their replication. Dengue virus (DENV), a member of the Flaviviridae family, is one of the most important arthropod-borne human pathogens worldwide. We analyzed global intracellular metabolic changes associated with DENV infection of primary human cells. Our metabolic profiling data suggested that central carbon metabolism, particularly glycolysis, is strikingly altered during a time course of DENV infection. Glucose consumption is increased during DENV infection and depriving DENV-infected cells of exogenous glucose had a pronounced impact on viral replication. Furthermore, the expression of both glucose transporter 1 and hexokinase 2, the first enzyme of glycolysis, is upregulated in DENV-infected cells. Pharmacologically inhibiting the glycolytic pathway dramatically reduced DENV RNA synthesis and infectious virion production, revealing a requirement for glycolysis during DENV infection. Thus, these experiments suggest that DENV induces the glycolytic pathway to support efficient viral replication. This study raises the possibility that metabolic inhibitors, such as those that target glycolysis, could be used to treat DENV infection in the future.

Importance: Approximately 400 million people are infected with dengue virus (DENV) annually, and more than one-third of the global population is at risk of infection. As there are currently no effective vaccines or specific antiviral therapies for DENV, we investigated the impact DENV has on the host cellular metabolome to identify metabolic pathways that are critical for the virus life cycle. We report an essential role for glycolysis during DENV infection. DENV activates the glycolytic pathway, and inhibition of glycolysis significantly blocks infectious DENV production. This study provides further evidence that viral metabolomic analyses can lead to the discovery of novel therapeutic targets to block the replication of medically important human pathogens.

FIG 1

Carbon metabolism is altered during DENV infection. HFFs were mock or DENV infected (MOI of 9) and harvested at 10, 24, and 48 hpi for intracellular metabolic analysis by Metabolon. (A) Heat map visualization of fold changes in levels of glycolytic intermediates profiled during a time course of DENV infection. For paired comparisons, shaded cells indicate P ≤ 0.05 (dark gray indicates that the mean values are significantly higher in DENV-infected cells for that comparison, and light gray indicates that the values are significantly lower). Italicized values indicate 0.05 < P < 0.1. (B and C) Line plots of selected glycolytic intermediate levels (B) or glutaminolytic intermediate levels (C) in mock- and DENV-infected cells measured at the time points indicated.

FIG 2

Exogenous glucose is necessary for efficient infectious DENV production. HFFs were infected with DENV at an MOI of 3 and fed replete, glucose-free, or glutamine-free medium at 2 hpi. At 24 hpi, released infectious virus was quantified by focus-forming-unit reduction assays on Vero cells (A) or intracellular viral RNA from DENV-infected cells fed replete or glucose-free medium was measured by real-time RT-qPCR (B). ***, P < 0.0001.

FIG 3

Glycolysis is induced during DENV infection. (A) Glucose uptake is increased during DENV infection. At 24 hpi, mock- and DENV-infected (MOI of 9) HFFs were exposed to a radiolabeled glucose analog for 5 min and then intracellular radioactivity was quantified. (B, C, and D) Expression levels of GLUT1 and HK2 are elevated in DENV-infected cells. HFFs were mock or DENV (DV) infected (MOI of 9) and treated with 0 or 150 μM DFO at 2 hpi. Cell lysates harvested at 24 hpi were subjected to immunoblot analysis with the antibodies indicated. The values below the lanes are the relative intensities of the major bands, and β-actin served as a loading control. (C) Real-time RT-qPCR analysis of HK2 transcript levels in mock- and DENV-infected cells (MOI of 5) harvested at 24 hpi. Relative abundance of HK2 mRNA was normalized to the abundance of GAPDH mRNA by the delta threshold cycle method. *, P ≤ 0.05.

FIG 4

Glycolysis is required for maximal DENV replication. (A, B, and C) HFFs were infected with DENV at an MOI of 3 and fed replete medium (0 mM) or replete medium supplemented with oxamate (50 or 100 mM) or 2DG (10 or 50 mM) at 2 hpi. At 24 hpi, released infectious virus (A) and viral RNA (B and C) were quantified by focus-forming unit reduction assays or real-time RT-qPCR, respectively. (D and E) TIME cells were infected with DENV at an MOI of 1 and fed replete medium supplemented with 0, 50, or 100 mM oxamate at 2 hpi. Infectious extracellular virus (D) and viral RNA (E) were quantified at 24 hpi as described above. **, P ≤ 0.01; ***, P < 0.0001; ns, not significant.