Glycogen synthase kinase 3ß participates in late stages of Dengue virus-2 infection

 The glycogen synthase kinase-3 (GSK-3) is a mul­tifunctional monomeric protein that participates as an intermediary in several signalling pathways, including the Wnt/?-catenin, Hedgehog and PI3K/Akt pathways. Several mechanisms and molecules can activate GSK-3, including activation of cytokine receptor, heterotrimeric G protein-coupled receptors and tyrosine kinase recep­tors. The role of GSK was identified in the metabolism of glucose through phosphorylation and subsequent in­hibition of the glycogen synthetase enzyme and insulin signalling. However, GSK-3 was later identified as a pro­tein having serine-threonine kinase activity that regu­lates different cellular processes such as gene transcrip­tion, embryonic development, translation, cytoskeletal organisation, cell cycle progression and regulation of pro and anti-apoptotic pathways. Therefore, GSK-3 activity is tightly modulated by cells.⁽¹⁾

GSK3 is highly conserved and plants, fungi, flies, helminths, and vertebrates exhibit orthologous proteins. In mammals, there are two homologous forms of GSK3 gene product: GSK3? of 51 kDa (located on chromo­some 19) and GSK3? of 47 kDa (located on chromosome 3), which possess 85% similar and 98% homologous se­quences within their kinase domains.⁽²⁾ However, these proteins are not functionally homologous or redundant. GSK3?, better studied, is constitutively active in resting cells and is inhibited upon activation of any signalling pathways in which it is involved.⁽¹⁾ This kinase is mainly found in cytoplasm and nucleus, but it can also be found in mitochondria where its activity is regulated. Regula­tion of GSK3? by phosphorylation has been extensively studied. Phosphorylation at serine 9 (Ser9) and tyrosine 216 (Tyr216) lead to GSK3? inactivation and activation, respectively. In addition, formation of protein com­plexes, intracellular localisation, and certain stabilising drugs influence GSK3? modulation.⁽³⁾

Impairment of GSK3? function have been described in several disorders and diseases including cancer, car­diovascular disease and neurological disorders such as Alzheimer’s disease, bipolar disorders, and Huntington’s disease. GSK3? is also involved in neoplastic transfor­mation and development of hepatocellular cancer.⁽⁴⁾

A few investigations have described participation of GSK3? in inducing apoptosis in viral infections includ­ing those caused by varicella-zoster virus (VZV), hepa­titis C virus (HCV), human immunodeficiency virus-1 (HIV-1), Venezuelan equine encephalitis virus (VEEV), coxsackievirus and enterovirus 71 (EV71).^{(5, 6, 7, 8)}

In infections caused by Dengue virus (DENV), GSK3? regulates transcription factor NF-?B, lead­ing to production of nitric oxide (NO) and induction of apoptosis. This signal is triggered by binding of DENV anti-NS1 antibodies to cells.⁽⁹⁾ DENV-2 inhibits GSK-3 activity to induce expression of MHC Class-1-related chain (MIC) A and MIC-B, and IL-12 production in monocyte-derived dendritic cells (Mo-DCs).⁽¹⁰⁾

Considering that PI3K/Akt kinase pathway is in­volved in the infection of epithelial cells, Huh-7 and Vero, by DENV-2⁽¹¹⁾ a nd t hat GSK3? is downstream of this cascade, it is intriguing to evaluate the role of GSK3? in the infective process of DENV-2. Current reports on the participation of GSK3? activity in DENV-2 infection process has been contrasting. In some settings, GSK3? activation leads to apoptosis, while in other conditions it seemed to induce cell proliferation.⁽¹²⁾ Furthermore, GSK3? pathway has been hypothetically postulated as crucial in modulating Drosha microprocessor activity and microRNA biogenesis that could be the trigger of important events involving microRNAs at early stages of DENV infection.⁽¹³⁾ Likewise, several families of miR­NAs including miR-34, miR-15, and miR-517 families have been reported to inhibit multiple flaviviruses such as DENV, West Nile virus (WNV) and Japanese encepha­litis virus. Members of miR-34 family can repress Wnt/?- catenin signalling with antiflaviviral effects, modulating type I interferon (IFN) signalling pathways by binding of GSK3? to TANK-binding kinase 1(TBK1).⁽¹⁴⁾

In this work, the role of the protein GSK3? during Dengue virus infection was investigated in Huh7 and Vero cells. Importantly, we compared GSK3? activation during the stages of infection and assessed its influence on cellular responses and viral release.

MATERIALS AND METHODS

Cell culture - Viruses were cultured in C6/36 HT (high temperature) cells from Aedes albopictus. Virus cultures were titrated in Vero cells (from African green monkey kidney, ATCC number CCL-81); these cells and Huh7 cells (human hepatoblastoma, donated by Dr Pris­cilla Yang, Harvard Medical School, Boston, MA, USA) were used for evaluation assays of GSK3? pathway. Spe­cific monoclonal antibodies against DENV E protein (?E) were obtained from culture supernatants of 4G2 hybridoma cells (ATCC number: HB-112). Vero, Huh7, and C6/36 HT were maintained in Dulbecco’s Modified Eagle Medium (DMEM) (Gibco) supplemented with 1-10% FBS (Gibco); 4G2 cells were grown in Hybry-care medium (ATCC), all supplemented with 1% peni­cillin/streptomycin (Sigma-Aldrich, St. Louis, MO). All cells were maintained in 5% CO₂ atmosphere at 37ºC, except for C6/36 HT, which was maintained at 34ºC.

Pharmacological inhibitors and antibodies - GSK3? small molecule inhibitor Kin-001-184 was donated by Dr Priscilla Yang (Harvard Medical School). CT 99021 (Kin-001-157) was obtained from Axon (cat # 1386 Groningen - The Netherlands). Mycophenolic acid (MPA), obtained from Sigma-Aldrich (Ref. M3536- 250G), was used as positive control for the inhibition of DENV replication. GSK3? inhibitors were dissolved in dimethyl sulfoxide (DMSO, Sigma) and MPA was dissolved in methanol (50 mg/mL). The primary anti­bodies used were rabbit ?-GSK3? (cat # 9369), rabbit ?-phospho-GSK3?-Ser9 (cat # 9323), rabbit ?-Akt (cat # 9272), rabbit ?-phospho-Akt-Ser473 (cat # 9271S), rab­bit ?-GADPH (cat # 2118), and rabbit ?-?-catenin (cat # 9587) (Cell Signalling, Danvers, MA). For immunoflu­orescence, secondary antibodies conjugated to fluoro­phores Alexa 488 and Alexa 594 (Molecular Probes, Eugene, OR) were used, and Hoechst 33258 (Thermo Fisher Scientific, cat # H3569) was used for nuclear labelling. The secondary antibodies used were IRDye 800CW goat anti-mouse and IRDye 680 goat anti-rabbit (1:15000) (Li-COR, Lincoln, NE). Protein quantification was performed using BCA Protein Assay kit (Pierce, Thermo Scientific ref 23225).

Cytotoxicity assay - Following treatments with inhibi­tors, the viability of Huh7 cells was tested using the MTT (3- (4,5-Dimethylthiazol-2-yl) -2,5-Diphenyltetrazolium bromide) assay. Cells were seeded onto 96-well plates and incubated for 24 h. The culture medium was replaced with DMEM-containing GSK3? or MPA inhibitors at concentrations of 5, 10, 20, and 40 ?M, prepared by serial dilution. After 24 h incubation, the medium was replaced with 50 ?L of MTT [0.5 mg/mL in phosphate-buffered sa­line (PBS)], followed by 3 h of incubation at 37ºC. DMSO (100 ?L) was added to solubilise formazan crystals and incubated for 15 min. Absorbance at 450 nm measured using a microplate reader (Benchmark, Bio-Rad Labora­tories, Hercules, CA, USA). Three independent experi­ments were performed with each treatment in triplicates.

Virus growth and titration - The prototype strain DENV-2 New Guinea C (NGC) donated by Maria Elena Peñaranda and Eva Harris (Sustainable Sciences Insti­tute and the University of California) was used in all in­fection experiments. Virus stocks were used for infection of C6/36 HT cells at low multiplicity of infection (MOI) (0.01 PFU/cell). Once infected, cells were incubated for seven days and supernatants were aliquoted and stored at -80ºC until titration. Viral titre determination was performed by diluting virus (10_-1-10_-5) in serum-free me­dium. Vero cell monolayers grown to 90% confluence in 48-well plates were inoculated with diluted virus. Af­ter 1 h adsorption at 37ºC, viral inoculum was removed. Cells were washed with PBS and covered with 2% car­boxymethyl cellulose (medium viscosity carboxymethyl cellulose, Sigma-Aldrich) in DMEM containing 2% foe­tal bovine serum (FBS). After seven days of incubation, cells were fixed with 4% paraformaldehyde and stained with 0.5% violet crystal prepared in 20% methanol. Vi­ral titre calculations were done by counting two replica plates from three independent experiments (n = 6).

Flow cytometry of infected cells - Huh7 cells (2 × 10⁵) were seeded onto 6-well plates for 24 h. The cells were washed once with warm trypsin-supplemented PBS and twice with PBS. Cells were resuspended in 500 ?L of PBS and labelled with DIOC6 (to measure the mitochon­drial membrane potential) and propidium iodide (PI3-A, to assess cell membrane damage).

Assessment of GSK3? phosphorylation using In-Cell Western - Activation kinetics of GSK3? was done in situ using In-Cell Western. Briefly, 2.5 X 10⁴ Huh7 cells were seeded into each well of 96-well plates and incubated in 2% FBS-containing medium. To cease activation of sig­nalling pathways by growth factors, culture medium was replaced with serum-free medium 24 h later, followed by 2 h of incubation at 37°C. The medium was subsequently removed and the wells were washed once with warm pre­heated PBS. Cells were infected with DENV-2 at a MOI of 5 in a final volume of 25 ?L/well for indicated times (1 min to 2 h); cells were washed with cold PBS, fixed with 4% paraformaldehyde (PFA), and incubated at room temperature for 20 min with gentle agitation. After five 5-min washes with wash solution (Triton 0.1% in PBS) with gentle agitation, cells were incubated with 150 ?Lof blocking solution (LICOR ODYSSEY blocking buffer) and incubated for 90 min at room temperature under moderate agitation. Subsequently, the blocking solution was removed and cells were incubated for 2 h at room temperature with either rabbit ?-pGSK3?-Ser-9 (1:100) or mouse ?-GSK3? (1:100). After thorough washes, cells were incubated secondary antibodies IRDye goat ?-rabbit 800D or IRDye 680RD goat ?-mouse diluted 1: 500 (di­luted in ODYSSEY LICOR blocking buffer) at room temperature for 1 h with gentle agitation. Cells were then washed thoroughly with wash solution. All wash solution were completely removed from wells and cells were ana­lysed using Odyssey Infrared Imaging System, software version 3.0 (Li-COR). ?-pGSK3?-Ser-9 values were nor­malised to baseline GSK3? protein levels.

Evaluation of GSK3? small molecule inhibitors - Two small molecule inhibitors of GSK3? were evalu­ated in Huh7 cells prior to or following infection with DENV-2 for three different time points (0-24 h) and the more effective inhibitor was chosen for further experi­ments. (1) Pre-infection treatment: 3 h before infection, cells were treated with Kin-001-157 inhibitor (iGSK3?) at concentrations of 20 or 40 ?M. Prior to infection, me­dium was removed and cells washed with pre-warmed PBS. Viral inoculum was added (DENV-2 MOI = 10) and infection maintained for 1 h. Cells were washed and subsequently incubated in 2% FBS, drug-free DMEM for 24 h. (2) Early infection treatment [0-12 h post-infec­tion (hpi)]: cells were infected with DENV-2 (MOI = 10) diluted in DMEM containing iGSK3?, and incubated for 1 h. Cells were washed and medium replaced with 2% FBS-DMEM containing inhibitor and incubated for 11 h. Medium was replaced using inhibitor-free, 2% FBS medium and cells were incubated for 12 h. (3) Late in­fection treatment (12-24 hpi): cells were infected with DENV-2 (MOI = 10) diluted in serum-free medium and incubated for 1 h. Cells were washed with PBS and in­cubated for 11 h in 2% DMEM Medium was replaced with 2% FBS-DMEM containing iGSK3? and cells were incubated for 12 h. The concentrations of iGSK3? tested were 20 and 40 ?M. Supernatants were collected after a total of 24 h post-infection or treatment and monolayers fixed with 4% PFA for immunofluorescence assays or lysed for western blotting.

GSK3? silencing with siRNA and shRNA - S ilenc­ing of GSK3? was carried out using two methodologies: cells were transfected with plasmids with gene sequenc­es expressing short hairpin RNAs (shRNAs) or a com­mercial pool of small interfering RNAs (siRNAs).

Reverse transfection of shRNAs - Three different versions of lentiviral vector pCMV-GIN-ZEO-GSK3? expressing green fluorescent protein (GFP) were used. ⁽¹⁵⁾ Versions 1 and 3 (Ver-1 and Ver-3) express shRNAs targeting GSK3? and have been previously validated,⁽¹⁶⁾ and the scrambled version (Ver-2) contained nontarget­ing specific sequences. Briefly, 4.0 ?g of lentiviral DNA (quantified using the Nanodrop system) was dissolved in 500 ?L of Opti-DMEM medium (serum-free medium in 6-well plates). Four ?L of Lipofectamine (Invitrogen) was added to the DNA, gently mixed, and incubated for 20 min at room temperature. Huh7 or Vero cells suspen­sion (in 2% FBS-DMEM, 2 x 10⁵ cells/well) was added into the DNA/Lipofectamine mixture in the wells, and incubated at 37ºC. After 48 h of incubation, the trans­fection efficiency was confirmed by GFP expression for fluorescence using the TYPHOON 9400 imager. Cells expressing ? 50% GFP efficiency were infected 48 h post-transfection (hpi). The supernatant was removed from cells 24 hpi (72 hpi) before lysing. Cell lysates were stored at -70ºC until analysis.

Reverse transfection of siRNAs - A pool of six dif­ferent siRNAs directed against GSK3? was used. For the negative control, nontargeting siRNA (NT Pool) was used. For transfection, 6 pmol siRNA/well was dissolved in 200 ?L Opti-MEM in 12-well plates and mixed gently; 1 ?L of Lipofectamine RNAiMAX was added to each well, mixed, and incubated at room temperature for 20 min. Huh7 or Vero cell suspension (in 2% FBS-DMEM, 2 × 105 cells/well) was added to the DNA/Lipofectamine mixture in wells and incubated for 24 h at 37ºC, prior to infection, for the indicated times. The supernatant was removed from cells 24 hpi (72 hpi) before lysing. Cell lysates were stored at -70ºC until analysis.

Western blotting - Cells were lysed with lysis buffer (150 mM NaCl, 20 mM Tris pH 7.4, 10% glycerol, 1 mM EDTA, 1% NP40, and 1 mg/mL protease inhibitor cock­tail). Twenty ?g of protein in the loading buffer (0.375 M Tris, pH 6.8, 50% glycerol, 10% SDS, 0.5 M DTT, and 0.002% bromophenol blue) was denatured by heating at 100ºC for 5 min before gel electrophoresis [10% sodium dodecyl sulphate-polyacrylamide gel electrophoresis polyacrylamide gels (SDS-PAGE)] using a Mini-Protein System (Bio-Rad). Separated proteins were transferred onto nitrocellulose membranes (Amersham, GE, Bos­ton, MA) in a Mini Trans-Blot electrophoretic transfer cell at 250 mA for 2 h. Membranes were washed using wash buffer, T-TBS (20 mM Tris-HCl pH 7.5, 500 mM NaCl, 0.05% Tween-20 in buffered saline, pH 7.4), and blocked with 5% of skimmed milk for 1 h. Membranes were incubated overnight at 4ºC with the appropriate primary antibodies: Rabbit ?-pAkt Ser-473 (1:500), rab­bit ?-p-GSK3?-Ser9 (1:500), undiluted 4G2 ?-Envelope antibody (?-ENV), or mouse ?-GADPH (1:1000). Mem­branes were thoroughly washed and incubated with peroxidase-coupled anti-rabbit or anti-mouse secondary antibodies (1: 5000, Pierce). Signals were developed us­ing electrochemiluminescence (ECL, Thermo Scientif­ic) and imaged with autoradiographic films (Hyperfilm ECL, Amersham or AGFA RP2 plus films).

Fluorescence microscopy - H uh7 c ells w ere p re­pared for fluorescence microscopy according to Cuar­tas et al.⁽¹¹⁾ Briefly, cells were seeded on coverslips in 24-well plates at a density of 5 x 10⁴ cells per well. At 24 hpi, cells were washed with cytoskeleton buffer (CB) and fixed with 3.8% PFA at 37ºC for 30 min. Cells were permeabilised with 0.5% Triton X-100 in CB. Cells were blocked with 5% FBS in CB and subsequently incubated with undiluted primary ?E antibody. After thorough washes, cells were simultaneously incubated with anti-mouse secondary antibody conjugated to Alexa 594, phalloidin Alexa 488 (for actin labelling) and Hoechst 33258 (for core labelling, 1: 5000) followed by washes with CB. Fluorescence were evaluated using an epi­fluorescence microscope (IX-81 Olympus), and images captured by software (Media Cybernetics, Image-Pro Plus). Confocal imaging was obtained using a FluoView FV1000 Confocal Microscope (Olympus).

Image analysis - Quantification of RGB images ob­tained by fluorescence microscopy was performed in Fiji (Distribution of ImageJ 2.0.0.). For contrast enhancing of images (gray value histogram-based approach), pixels saturated at 0% were used to define intensity thresholds. Measurements of integrated density and mean of area gray values for each cell and its background were used to estimate fluorescence response of DENV E protein in cells. The DENV E protein fluorescence response is defined as the mean intensity of the gray values assigned to every pixel within a defined cell area whose value is higher than the intensity of the background pixels.

Statistical analysis - Analysis of variance (ANOVA) was performed. Error bars correspond to 95% confi­dence interval. Analyses were carried out using PRISM 8 statistical package. Results were considered signifi­cant if type II statistical error was 95%.

RESULTS

Infection of Huh7 cells with DENV-2 caused damage to cell membranes - Effect of DENV-2 infection on cell membrane and mitochondria activity was tested in Huh7 cells. Damage to cell membrane following infection with DENV-2 at different MOIs was assessed by flow cytom­etry measurement of PI3-A. Levels of PI3-A increased in cells infected at MOI 1 and 10, compared to unin­fected cells (Fig. 1A-C). A decrease in mitochondrial ac­tivity of infected cells was noted at both MOI (Fig. 1D). However, at MOI of 10, fluorescence intensity of DIOC6 in infected cells increased (Fig. 1E).

DENV-2 induces inhibitory phosphorylation of GSK3? (Ser9) in Huh7 cells - To assess changes to GSK3? activities during infection with DENV-2, we performed dose dependent infection experiments for up to 2 h and evaluated GSK3? phosphorylation status in situ using In- Cell Western. An inhibitory phosphorylation of GSK3?- Ser9 was observed in Huh7 cells after 1 min of infection with DENV-2. p-GSK3? remained sustained through 50 min post-infection (Fig. 2). This suggests that GSK3? be­comes inactivated very early during DENV-2 infection.

Continuous inhibition of GSK3? modulated DENV-2 activities - The effect of two small molecule inhibitors of GSK3? on DENV-2 infection was assessed. According to previous reports, Kin-001-184 and Kin-001-157 inhibit GSK3? with high specificity.⁽¹⁷⁾ A decrease in the intra­cellular DENV E protein was detected when Vero and Huh7 cells were treated with non-cytotoxic concentra­tions (5, 10, 20, and 40 ?M) of inhibitors over infection period (0-24 hpi), compared to untreated infected cells (Fig. 3A). Culture supernatant exhibit dose-dependent reduction in viral titre following continuous treatment of cells with Kin-001-157 at 20 and 40 ?M resulted in 0.9- and 1.8-log reduction in viral titre was observed, respec­tively. However, treatment with Kin-001-184 resulted in only 0.7 and 0.5 Log reduction in viral titre at 40 and 20 ?M concentrations, respectively (Fig. 3B). Consequently, Kin-001-157 was chosen for subsequent experiments giv­en its better inhibitory actions on viral activities.

Inhibition of GSK3? selectively affected the late but not the early stages of DENV-2 infection - Since viral in­fection is a multi-stage process, we employed three strat­egies to delineate the role of GSK3? at infection stages in the Vero and Huh7 cells: pre-infection treatment (3 h prior) at the early (0-12 hpi) and late post-infection time­points (12-24 hpi). Inhibiting GSK3? did not affect viral titres or the amount of intracellular DENV E protein in pre-treated cells (Fig. 4A-C) or at early infection (0-12 hpi) (Fig. 4D-F). In contrast, treatment of cells with iGSK3? during the later stages of infection (12-24 hpi) resulted in a reduction of viral titres. In Vero cells, 1.1- log reduction in the viral titre at inhibitor concentrations of 40 and 20 ?M (Fig. 4G) was observed. In Huh7 cells, viral titres decreased by 1.4 and 0.8 Log at 40 ?M and 20 ?M, respectively (Fig. 4H); there were no changes in the DENV E levels as detected by western blotting (Fig. 4I). Mycophenolic acid (MPA) and ribavirin treatment were used as positive controls due to their demonstrated inhibitory effects on DENV virus replication.

Subcellular distribution of viral envelope protein remained unaffected by GSK3? inhibition in DENV- 2-infected Huh7 cells - We i nvestigated t he d istribution of viral proteins at various time points following infec­tion with DENV-2 in the presence or absence of iGSK3? in Huh7 cells. A homogeneous intracellular distribution pattern was observed for DENV E in infected, treated or untreated Huh7 cells (Fig. 5A). Neither treatment with GSK3? inhibitor at early infection stages 0-12 hpi (Fig. 5B) nor at late infection stages 12-24 hpi (Fig. 5C), had significant effects on DENV E distribution pattern. Sta­tistical analysis indicated no significant change (Fig. 5D) in fluorescent intensity of DENV E protein as infection con­trol and GSK3? treatments were compared (p = 0.3898).

Knockdown of GSK3? had no effect on DENV-2 in­fection of Vero and Huh7 Cells - L oss-of-function ex­periments were carried out in Vero and Huh7 cells using shRNAs and siRNAs that targeted GSK3?. Silencing of GSK3? was evaluated by GFP protein expression from vectors bearing interfering shRNAs pCMV-GIN-ZEO-GSK3? (Ver1 and Ver3), which were detected 48 hpi us­ing TYPHOON 9400 scanner for Huh7 cells (Fig. 6A), and fluorescence microscopy for Vero cells (Fig. 6C). The amount of GSK3? protein in cells expressing GFP at baseline (arrowheads, Fig. 6E) was also tested. An ex­pression efficiency was observed at 48 h, the time point at which transfected cells were infected. Supernatants of infected and knocked-down cells were subsequently titrated 24 hpi and 72 hpi (Fig. 6B, D). In addition, potent knockdown of GSK3? with siRNAs in Huh7 and Vero cells was confirmed by western blotting using infrared detection of Odyssey System (Fig. 6E, G). However, no decrease in viral titres was observed in both Vero and Huh7 cells (Fig. 6F, H).

Fig. 7 is a schematic of the main findings of the role of GSK3? in the infection by Dengue virus.

DISCUSSION

The involvement of PI3K/Akt signalling pathway proteins including GSK3? and several other downstream effectors in viral infections has been described. GSK3? participates in the infection cycle of some viruses such as enterovirus,⁽¹⁸⁾ human papillomavirus (HPV),⁽¹⁹⁾ vari­cella-zoster virus (VZV),⁽⁵⁾ hepatitis C virus (HCV),^{(8, 20)} among other. PI3K/Akt signalling pathway is activated during infection cycle resulting in apoptosis.^{(21, 22, 23, 24)} Nonetheless, the role of GSK3? in this process is not ful­ly understood.

We previously reported that DENV-2 infection caus­es activation of Akt in Huh7 and Vero cells.⁽¹¹⁾ Activation of Akt pathway during the infection with DENV and the Japanese encephalitis virus is associated with apoptosis inhibition.⁽²¹⁾ However, proteins in signalling pathway downstream of active DENV infections remain uniden­tified. Activation of Akt and downstream inactivation of GSK3? inhibit cell death and modulate cell cycle regula­tion by cyclin-D1,⁽²⁵⁾ implicating inactivation of GSK3? as potential requirement for the inhibition of extrinsic pathway-triggered apoptosis during early viral infection. This hypothesis is consistent with findings from the cur­rent study in which we observed Ser9 phosphorylation and inactivation of GSK3? at early-infection time points.

The inactivation of GSK3? would explain the lack of effect on virus production upon chemical inhibition of GSK3? at this stage of infection.

We did not investigate activation of PI3K/Akt and downstream inactivation of GSK3? using UV-inactivat­ed viruses, as our focus was on delineating the specific role of DENV-2 infection with active virions. Howev­er, Hilde M van der Schaar et al.⁽²⁶⁾ suggested that ac­tivation of Akt pathway occurs upon engagement of cell receptors by the virus. The study presented DENV tracking in living cells, where authors detected that sin­gle DENV particles are able to bind membrane regions enriched with clathrin-coated pits only 48 s after infec­tion. Whereas at 94 seconds, the clathrin signal rapidly disappears indicating disassembling of the clathrin shell required for the subsequent internalisation of the vesicle. Fusion of viral membrane with late endosomes occurred 512 s post-infection. Based on this work and our find­ings, we presume that activation of PI3K/Akt pathway and subsequent phosphorylation of GSK3? as early as 1 min post-infection occur upon virus binding to the cell receptor involved in activation of the pathway even be­fore viral endocytosis begins.

Our findings on the treatment of infected cells with iGSK3? later in the replication cycle (12-24 hpi) were also consistent with what is expected on apoptosis in­duction during viral infections for the release of new enveloped viral particles of viruses such as DENV. If GSK3? plays a role in the induction of cell death during DENV-2 infection, a late inhibition would affect mito­chondria-dependent apoptosis, which can be regulated by GSK3?,(27) and thus influence viral release or intra­ cellular trafficking of viral particles. Our flow cytom­etry experiment data suggest that this phenomena may occur by means of mitochondrial intrinsic apoptosis, considering a statistically significant reduction in mito­chondrial activity (DIOC6) in cells infected at MOI = 1 and MOI = 10, compared with uninfected cells.

Although results obtained from small molecule in­hibitors of GSK3? and interfering RNAs (shRNAs and siRNAs) did not show similar results related to a decreased viral infection, the lack of a GSK3? silenc­ing effect on the infection could likely be explained by the activity of non-silenced protein. Although the use of interfering RNAs (siRNAs) resulted in a remarkable decrease in the amount of GSK3?, as seen via western blotting, this reduced protein level does not affect nor­mal functioning during DENV-2 infection (12-24 hpi). Similar results on the efficacy of pharmacological in­hibitors, compared to genetic inhibitors, have been ob­served in studies involving other viruses.⁽²⁸⁾ In our study, this might be explained by limited silencing of a single protein isoform (GSK3B1). Recently, studies conducted in Huh7.5 cells using a HCV replicon showed that treat­ment with a GSK3? inhibitor affected viral replication cycle late during infection, very likely at the assembly and release of viral particles,⁽⁸⁾ which was confirmed by the findings in the current study.

Future studies that demonstrate a participation of cellular proteins such as GSK3? in viral infections may allow potential use as a specific therapeutic target for the treatment of infections, capitalising on participation of the kinase in later steps of the signalling pathway. The role of GSK3? in the development of the Dengue disease and the immune-pathogenic mechanisms responsible for severe Dengue fever has been described. Since there is no vaccine or drugs currently available for the treatment of Dengue fever, GSK3B inhibition could counteract or reduce complications from the disease.

Therapeutic use of PI3K/Akt inhibitors has been ap­plied in patients with different types of cancer,⁽²⁹⁾ whereas GSK3? protein inhibitors are used in the treatment of neurodegenerative diseases such as Alzheimer’s disease. ⁽³⁰⁾ The availability of pharmacological inhibitors against proteins involved in this signalling pathway for the treat­ment of chronic diseases would provide opportunities for rapid evaluation of their potential use in treatments of viral diseases such as Dengue fever. However, given the broad spectrum of metabolic pathways that may be impacted and the regulatory role of these proteins in some essential cellular processes, such as regulation of glucose metabo­lism described for GSK3?, the identification of possible side effects of these inhibitors would be necessary.

In conclusions - I n t his w ork, w e d escribe t he i n­volvement of the GSK3? during DENV-2 infection of Huh7 and Vero cells, in which the kinase specifically modulates late stages of infection, during possible ac­tivation of apoptosis to promote the viral release from infected cells. These findings indicate the potential role of GSK3? during DENV-2 infection process and to some extent, elucidate the complex network of intracellular in­teractions triggered by the virus in infected cells, aimed at maximising the viral replication process. Although the involvement of the PI3K/Akt signalling pathway in Dengue virus infection has already been described, participation of downstream effectors is very diverse, and little is known about these cellular proteins. Aside well-described roles of GSK3? in process of glucose metabolism and different cellular processes, growing evidence supports its participation in induction of apop­tosis in some viral infections such as HIV-1, VZV, HCV, among others. Further studies are required to advance our knowledge and fully describe the participation of cellular proteins such as GSK3? in viral pathogenesis.

ACKNOWLEDGEMENTS

To Dr Priscilla Yang and Dr Michael Vetter (Harvard Medical School, Boston, MA), for their support in providing reagents, technical, and scientific advice in their laboratory; Dr Gloria Patricia Cardona (Group of Neurosciences, Univer­sidad de of Antioquia) for her support with reagents, conduc­tion of experiments, advice, and critical review of the manu­script. We are also grateful to Vicky C Roa Linares (Group of Molecular and Translational Medicine, Universidad de An­tioquia) for the critical review and contributions to the final version of the manuscript.

AUTHORS’ CONTRIBUTION

ACL performed the cellular, molecular, and virology ex­periments, and wrote the first draft of the paper; JCGG as PI of the Colciencias Grant conceived the study and critically re­viewed/corrected this manuscript. All the authors have read and approved the final version of the manuscript.