Background
Dengue is an important public health problem: Using updated mapping approaches, there are 390 million estimated dengue infections per year worldwide, of which 96 million manifest clinically at some level of severity [
1]. This leads to half a million hospitalizations and 25,000 deaths, mainly in children [
2]. The etiological agent of dengue disease is the dengue virus (DENV), which is responsible for dengue fever (DF), dengue hemorrhagic fever (DHF) and dengue shock syndrome (DSS).
Dengue virus, a member of the
Flaviviridae family, is an enveloped virus containing a ~11 kb genome of positive single-stranded RNA which encodes three structural proteins (C, pr-M, E) and seven nonstructural proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, NS5) [
3]. Four serotypes of dengue virus (DENV-1, DENV-2, DENV-3, and DENV-4) cause DF and more severe manifestations like DHF and DSS [
4].
Dengue virus is transmitted by female mosquitoes mainly of the species
Aedes aegypti and
A. albopictus. The disease is widespread throughout the tropics, with local spatial risk variations influenced strongly by rainfall, temperature and unplanned rapid urbanization [
5].
In many viral diseases, infection is controlled at the cellular level by type I interferon (IFN-α/β) produced in response of viral RNA. Recognition of IFN-I by its specific cell receptors activates the JAK-STAT signaling pathway, which leads to the establishment of an antiviral state in the cell. However, viruses possess diverse strategies to circumvent IFN effects [
6].
DENV, like other viruses, has the ability to inhibit this innate response, which may contribute to its pathogenesis. DENV virus induces the activation and release of high levels of chemokines and pro-inflammatory cytokines in human dendritic cells, although it is a weak inducer of IFN-I, as it blocks IRF-3 phosphorylation, impairing subsequent IFN-α/β production [
7].
There are three types of IFN. Type I comprises several subtypes of IFN-α and several other isoforms such as IFN-β,-ε,-κ,-δ,-ω,-τ and -ζ [
8,
9]. Type II contains only one member, IFN-γ. The latest identified group, type III, comprises IFN-λ1, IFN-λ2, IFN-λ3 (also named interleukin [IL]-29, IL-28A and IL-28B, respectively) and IFN-λ4 [
10‐
12], but there is limited information about their expression and biological functions. Furthermore, the role of IFN-III during DENV infection is unexplored.
IFN-λ specific signaling occurs through its receptor on the cell membrane, which consists of two subunits: IL28R1 (also named IFN-λR1, IL28Rα) and IL-10 receptor 2 (IL-10R2) [
8,
10,
11]. Most cells express the IL-10R2 subunit, but IL28R1 distribution appears to be much more restricted. The receptor for IFN-λ is expressed preferentially in cells of epithelial origin and in some cells of the immune system [
8,
10,
13].
Binding of IFN-λ to its receptor leads to the activation of JAK1 and Tyk2, which mediate the phosphorylation of STAT1 and STAT2 proteins. STAT phosphorylation leads to their dimerization, binding to IRF9 and translocation into the nucleus where they regulate the expression of specific genes. IFN-III has similar functions to IFN-I, inducing an antiviral state in cells [
8,
10,
13] and mediates the antiviral response against several human pathogens including hepatitis B virus, hepatitis C virus and human immunodeficiency virus [
13‐
15]. The restricted expression of its receptor suggests that the IFN-λ could exert an antiviral role on certain cell types [
16,
17].
There is no commercial vaccine or specific antiviral treatment for dengue; thus, the development of efficient therapeutic strategies against this common and sometimes severe infection is urgent. The antiviral activity of interferons has been observed in diverse clinical studies showing a relative efficiency against different viral infections. In 1984, Limonta Vidal et al. [
18] introduced the use of leukocyte interferon to treat children with dengue, showing a significant decrease in the number of complications and deaths when interferon was applied. However, the clinical use of interferon to treat dengue is not common, possibly due to IFN-α side effects associated with its systemic activity. IFN- λ is an attractive treatment alternative because it elicits a satisfactory antiviral activity with only a minor secondary response [
19,
20]. Nevertheless, DENV has developed mechanisms to evade the immune response. Therefore, the precise analysis of IFN- λ response during dengue infection is crucial to evaluate/predict its potential clinical uses in this disease.
The present work focuses on the effect in epithelial cells of IFN-λ (IFN-III) on the DENV-2 infection. We found higher levels of IFN-λ in dengue fever patients compared with healthy blood donors and we provide evidence that endogenous and exogenous IFN-λ inhibit DENV-2 replication in the C33-A cell line.
Discussion
IFN-III is the most recently identified interferon group [
10‐
12], and there is limited information about its expression and biological functions. The particular role of IFN-III during DENV infection is unexplored. Although most cells respond to IFN-I, certain cell types, especially epithelial cells, can respond to IFN-III due to the restriction in receptor IL28R1 expression [
23,
24].
In the present work we demonstrated that dengue fever patients had ~1.8 times higher IFN-λ levels compared to healthy blood donors (
p < 0.001). We are conducting further studies to answer whether IFN-λ plays a role in viral clearance or in the generation of specific clinical outcomes. For IFN-α, for instance, a recent study found higher levels in DF than DHF patients infected by DENV-1 or 2, that declined rapidly at day 3 after fever onset [
25]. Such studies require the analysis of many patients, the determination of disease severity and the inclusion of patients infected with different DENV serotypes, among other variables.
Epithelial cells are IFN-III’s main target. An analysis of IL28R1 receptor expression in mouse cells showed that vaginal epithelium and keratinocytes but not fibroblasts respond to IFN-λ [
23]. Several studies have been conducted in cell lines and primary cell cultures to study IFN-λ response [
26]. In the present study we found that the C33-A cervical, epithelial cell line is permissive to DENV-2 infection and that this cell line both produces and responds to IFN-λ. IL28R1, which is essential to support the antiviral action of IFN-λ, is expressed in C33-A cells. IL28R1 expression is inducible under viral infections, concurrent with the expression of IFN-λ [
16]. Accordingly, IL28R1 expression increased during DENV-2 infection in C33-A cells, which could potentiate the effect of IFN-λ signalling, leading to stimulation of gene transcription for the antiviral response and to inhibition of DENV-2 replication; as has been demonstrated for HCV, HIV and HBV infections [
15,
27].
The present study demonstrated the ability of IFN-λ in the inhibition of DENV-2 replication. IFN-λ reduced the amount of infectious particles secreted into the supernatant of C33-A cultures, quantified by plaque assays. Treatment with IFN-λ or IFN-α decreased the viral titer. Dose-response experiments showed that the IC50 values for IFN-λ1 and IFN-λ2 were 8.24 ng/ml and 8.705 ng/ml, respectively (Fig.
5). In view of these IC50 values, we tested whether an IFN- λ concentration of 10 ng/ml, which produces optimal gene induction in other viral infections [
28,
29], could induce the expression of antiviral response genes during dengue infection. In a study involving West Nile virus (WNV), another member of the
Flaviviridae family, induction of IFN-λ and JAK/STAT pathway activation was demonstrated in infected cells. IFN-λ was able to prevent WNV infection of naïve cells, but failed to inhibit viral replication in cells once infection had been established, likely due to IFN-λ inhibition [
30].
Furthermore, we report the IFN-λ1-mediated inhibition of DENV-2 replication in Vero cells, which are devoid of IFN-I, but have functional IFN-III responses. These data suggest that dengue virus and the IFN-λ treatment can active ISGs without type I IFNs. However, pre-treatment with IFN-λ1 before dengue infection was not able to inhibit viral replication to the same degree in Vero as in C33-A cells, which shows that dengue virus can block activated signalling pathways that induce both type I and III interferons. Since viral titers did not decrease markedly with the highest IFN-λ1 doses used, the data suggest that viral replication can be inhibited, albeit weakly, by endogenous and exogenous IFN-λ1 in complete absence of IFN-I. Furthermore, the reduction in viral progeny observed with IFN-λ1+ IFN-α, suggests that both interferons work synergistically. Importantly, the MX1 gene can be induced in Vero cells stimulated with recombinant type I interferon, thus demonstrating that although these cells don’t produce IFN-I they respond to it [
21,
31,
32].
The antiviral effect shown by IFN-III could occur through the stimulation of antiviral genes as cells were first treated with IFN-λ1 and then infected. It is noteworthy that DENV-2 infection induced ISG transcripts but that these did not reach the levels observed when cells were stimulated with IFN-λ1. However, when cells were pre-treated with IFN-λ1 and then infected, ISG transcript levels were significantly higher and were maintained longer than in the DENV-2-infected group. SOCS1 and SOCS3 were induced after ISGs reached their highest peaks, which may denote the negative action exerted on the IFN-activated JAK/STAT pathway. However, the analyzed ISGs in the presence of IFN-λ1 stimulation remained at low levels but were sufficient to reach a second expression peak capable of inhibiting viral replication at late stages, which explains the decrease in viral titers observed between 18 and 48 h (Fig.
6).
The antiviral genes MX1 and OAS1 showed an early activation at 6 h and peaked at 18 and 48 h, respectively, in the IFN-λ1/DENV-2 group. OAS1 mRNA was higher when cells were pre-treated with IFN-λ1 and subsequently infected compared to cells that were stimulated only with IFN-λ1. This outcome was likely due to increased activation of the gene in the presence of infection plus gene activation mediated by interferon.
OAS1 has antiviral activity against DENV-2 infection [
33]. The OAS/RNase L antiviral system is induced via IFN and plays a critical role in innate immunity, controlling the production of viral particles. RNase L possesses antiviral action, cleaving the viral RNA, inducing apoptosis in infected cells, and increasing the production of IFN-β through MDA5/RIG-I/IPS-I [
34].
Higher activation of the MX1 gene occurs through stimulation of IFN-λ1 because it is a gene-specific interferon response, which indicates the proper functioning of the pathway in the cell line. In the case of the IFN-λ1/DENV-2 experiments, increased expression was most likely caused by pre-treatment with IFN-λ1 because the DENV-2-infected group showed no transcript increase for this gene. It is significant that although the transcript levels were as high as when cells were stimulated only with IFN-λ1, the observed levels may be sufficient to inhibit viral replication even with the activation of the SOCS genes, whose levels appear not be sufficient to completely inhibit the transcription of MX1 at later times.
Transcription levels of ISGs in different groups showed a clear relationship between treatment with IFN-λ and its role in inhibition of DENV-2 replication. It is possible to observe a process of evasion of the immune response, most likely mediated by viral infection, showing a decrease in the transcription of evaluated ISGs in DENV-2 infected cells compared with cells stimulated with IFN-λ and an increase in SOCS1 and SOCS3 genes.
However, the IFN-λ1 gene was activated initially and then in the last times evaluated (6 and 48 h post-infection), suggesting its escape from the suppressor activity of SOCS proteins. When cells were stimulated with IFN-λ1 and then infected, there was a slight and steady increase in the evaluated ISGs, which peaked at 12 or 18 and 48 h post-infection, suggesting that activity of the pathway is only partially affected by viral evasion. These findings suggest that IFN-I and IFN-III are not always concomitantly expressed and support the idea that there are differences in the regulatory mechanisms of their expression and in the group of genes that can be individually or jointly activated [
35].
It is noteworthy that the highest peaks of the evaluated ISGs (12 and 18 h) corresponded to the time when the viral titer was reduced in culture supernatants; thus, greater inhibition of viral titer was achieved when ISGs reached their peaks (48 h). These findings suggest that inhibition of viral titer was caused by IFN-λ treatment and by the activation of genes in response to interferon, whose antiviral action in the cells inhibited the replication of DENV-2.
IFN-λ production increases after treatment with IFN-α prior to viral infection [
36,
37]. Our results suggest that IFN-III enhances the expression of IFN-I as pre-treatment with IFN-λ1 increased IFN-β expression during infection with dengue virus. This leads us to propose that the crosstalk between interferon type I and III pathways may exist in some cells, like mononuclear or plasmacytoid dendritic cells, to produce and respond to both types of interferon [
38,
39]. However, these are not necessarily redundant systems as shown in our experiments, wherein the expression of IFN-λ1 exists despite the evasion of the immune response mounted by DENV-2.
In the absence of antiviral drugs against DENV, this paper suggests that interferon lambda could be a valuable candidate to explore in clinical studies. In addition, the immunomodulatory properties reported for IFN-λ [
13] can contribute to the regulation of the inflammatory process characteristic of dengue. Studies with IFN-α in Japanese encephalitis models have shown decreased inflammation [
40], but treatment with IFN-α has side effects associated with the ubiquitous expression of its receptor [
19,
20,
41]. In contrast, the narrow expression of type III interferon receptor could considerably diminish side effects of IFN-λ clinical use.
Conclusions
This is the first report of increased levels of IFN-III in dengue fever patients in vivo and of the antiviral effect of IFN-III on DENV-2 infection in vitro; however, aspects of the activation of this pathway and its crosstalk with IFN-I remain unanswered.
The epithelial cell line C33-A is a useful model for in vitro mechanistic studies on DENV replication and its interaction with the innate response led by IFN-III. In infection studies, IFN-λ1 and IFN-λ2 inhibited DENV-2 replication in a dose-dependent manner. In Vero cells, which do not produce IFN-α, viral replication inhibition by exogenous IFN-λ1 was also observed, although at higher concentrations, suggesting that IFN-λ1 itself can inhibit the infection, however the potential to decrease the viral load increases with the presence of IFN-α. DENV-2 infection induces IFN-λ1 expression but not of IFN-β. Exogenous IFN-λ1 induced the transcription of several ISGs whose expression peaks corresponded to the time when viral titer was reduced in culture supernatants (12, 18, and 48 h post-infection). The present work opens the possibility for the study of other interactions and mechanisms of DENV infection with IFN-III in this cell model and presents an opportunity to consider this type of IFN as a feasible candidate for the inhibition of infection in dengue patients.
Methods
Cells
Human cervical carcinoma C33-A (ATCC: CRM-HTB-31), African green monkey kidney Vero (ATCC: CCL81, kindly donated by Dr. Judith González, Universidad Autónoma del Estado de Morelos, Mexico) and baby hamster kidney BHK-21 (C-13) (ATCC: CCL-10) cell lines were maintained in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10 % fetal calf serum, 100 U/mL penicillin and 100 μg/mL streptomycin at 37 °C with 5 % CO2. Aedes albopictus C6/36 cells (kindly donated by Dr. Celso Ramos, Instituto Nacional de Salud Pública, Cuernavaca, Mexico) were maintained in Minimum Essential Medium (MEM) supplemented with 10 % fetal calf serum, 100 U/mL penicillin and 100 μg/mL at 27 °C.
Virus
Dengue virus serotype 2 (Thailand/16681/1984) (DENV-2) used in this study was kindly provided by Dr. Alvaro Aguilar-Setien. DENV-2 was replicated in C6/36 cells and titrated by plaque assay in BHK-21 cells [
42].
Clinical samples
Serum samples from dengue patients collected from June to September 2013, were analyzed to determine the IFN-λ. All serum samples were maintained at -70 °C, thawed onto ice and immediately subjected to ELISA. Patients were recruited in the Regional Hospital at Mepetec, Puebla, Mexico. All patients had signs of dengue fever and DENV infection was confirmed by IgM antibodies and viral NS1 antigen detection. The samples from included patients were obtained between days 3 and 5 after fever onset. Sampling and patient information management were performed following ethical regulations approved by Local Research and Health Ethics Committee #2101 of Mexican Institute for Social Security, IMSS (Protocol: R-2011-2103-24) and in accordance with the Declaration of Helsinki. Written informed consent was obtained from each participant. Controls were serum samples of healthy blood donors submitted to conventional studies at the hospital blood bank and were used to compare IFN-λ levels.
Infection assays
C33-A or Vero cells were seeded onto 12- or 24-well plates and infected with different virus doses [quantified by multiplicity of infection (MOI)]. Viral inocula were incubated for 1 h and then discarded. Cells were washed twice to eliminate non-adsorbed virus. Fresh medium (SFB 2.5 %) was added to the cells and then incubated for hours or days depending on the specific assay.
Cell stimulation
Cells were treated prior to infection for 6 h with 100–500 UI/ml of IFN-α (Probiomed, Mexico city, Mexico.), 10–50 ng/ml of IFN-λ1 (PeproTech, Rocky Hill, NJ, USA) or 10–50 ng/ml IFN-λ2 (PeproTech). A poly(I:C) (InvivoGen, San Diego, CA, USA) control treatment was performed on cells using DMEM without SFB and 100 μg/ml of poly(I:C).
Western blot
Infected cells or mock-infected cells, previously treated or mock-treated with IFNs for 6 h, were analyzed in order to determine the induction of IFN-λ1 and IFN-λ2. Total extracts of cells were obtained 48 h post-infection and subjected to SDS-PAGE using a 15 % resolving gel. Proteins in gel were transferred to PVDF membranes and detected by anti-IFN-λ1 (Santa Cruz Biotechnology, Santa Cruz, CA, USA, sc-67642), anti-IFN-λ2 (Abcam, Cambridge, MA, USA, ab109820) and anti-β-actin specific antibodies (Abcam, mAbcam8226). Primary antibodies were revealed with a goat anti-rabbit IgG-HRP (Santa Cruz Biotechnology, sc-2004), rabbit anti-goat IgG-HRP (Abcam, ab6741) or goat anti-mouse IgG-HRP (ab97023) and the Immobilon Western chemiluminescence system (Millipore-Merck, Darmstadt, Germany).
Fluorescence microscopy
C33-A cells were seeded onto glass chamber slides (Lab-Tek, Nalgene Nunc, Rochester, NY, USA) and infected as mentioned above. Infected or non-infected controls were washed with phosphate-buffered saline (PBS) and fixed in ethanol/acetone 50 % v/v for 10 min at room temperature. Cells were permeabilized with 0.2 % Triton-X100 for 2 min in order to detect the viral pr-M protein inside infected cells and incubated in blocking solution (PBS with 10 % FBS and 0.2 % Tween 20). Cells were labelled with anti-dengue virus pr-M antibody (Abcam, ab41473) or anti-IL28R (Santa Cruz Biotechnology, sc-66541), washed and incubated with rabbit anti-mouse IgG-Texas red (Abcam, ab6726) or bovine anti-goat IgG-CFL-647 (Santa Cruz Biotechnology, sc-362284) for 1 h at room temperature. SYBR-14 dye (Molecular Probes, Eugene, OR, USA) was used to stain the nuclei. The stained cells were visualized by laser confocal microscopy system E600-C1 using the image analysis software EZ-C1 (Nikon).
Real time PCR
Total RNA of DENV-2 infected and/or stimulated cells was extracted using Trizol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions, treated with DNase I (Thermo Scientific, Waltham, MA, USA) and quantified at 260/280 nm. Reverse transcription was performed using RevertAid H Minus Reverse Transcriptase kit (Thermo Scientific) and random hexamers. The reaction program was 25 °C for 10 min, 42 °C for 60 min and 70 °C for 10 min. The obtained cDNA was used to amplify the IFN-β, OAS1 and MX1 genes using the Taqman Universal PCR Master Mix, no AmpErase UNG kit (Applied Biosystems, Foster City, CA, USA). IFN-λ1, SOCS1 and SOCS3 genes were amplified using Maxima SYBR Green/ROX qPCR Master Mix (Thermo Scientific). The reaction mixture contained 5 μl 2X Master Mix, 200 nM of each sense and antisense primer, 200 nM of Taqman probe and 35–50 ng RNA in a final volume of 10 μl. Forty PCR cycles were performed at 50 °C for 2 min, 94 °C for 10 min, 95 °C for 15 s, 60 °C for 30 s and 72 °C for 30 s in a StepOne Real-Time PCR System (Applied Biosystems). Specific primers for IFN-λ1, IFN-β, OAS1 and MX1 genes were designed using ProbeFinder Assay Design Software (Roche, Basel, Switzerland,
www.universalprobelibrary.com) in order to use the Taqman probes of Universal ProbeLibrary Set, Human (Roche). In all cases, HPRT1 (Hypoxanthine Guanine Phosphoribosyl Transferase) gene was used as an internal control to determine the relative expression of each gene. All primers were purchased from Integrated DNA Technologies (Coralville, IA, USA). Obtained Ct values were analyzed by the 2 − ΔΔCt method, and the expression levels were reported as proportions with respect to controls. Sequences of primers and probes are shown in Table
1.
Table 1
Primers used in the present study
MX1 | TCCAGCCACCATTCCAAG | TTCTCCTG (2) | 60 | 868–885 |
NM_001144925.2 | CAACAAGTTAAATGGTATCACAGAGC | 927–902 |
OAS1 | GGTGGAGTTCGATGTGCTG | CCAGGGCA (37) | 64 | 538–556 |
XM_006719434.1 | AGGTTTATAGCCGCCAGTCA | 601–582 |
IFN-β | CTTTGCTATTTTCAGACAAGATTCA | CTGGCTGG (20) | 69 | 336–360 |
NM_002176.2 | GCCAGGAGGTTCTCAAGAAT | 404–389 |
IFN-λ1 | TCCTAGACCAGCCCCTTCA |
c
| 73 | 435–453 |
NM_172140.1 | GTGGGCTGAGGCTGGATA | 507–490 |
IFN-λ2 | CCTGGTGGACGTCTTGGA |
c
| 84 | 406–423 |
NM_172138.1 | GTGGGCTGAGGCTGGATA | 489–472 |
IFN-λ3 | AGGGCCAAAGATGCCTTAG |
c
| 123 | 167–185 |
NM_172139.2 | CAGCTCAGCCTCCAAAGC | 298–281 |
HPRT1 | – | – | 127 | – |
NM_000194.2d
|
ELISA
Levels of IFN- λ in the serum of patients were measured using enzyme-linked immunosorbent assay (ELISA) kit specific for IFN-λ1 and λ3 (DY1598B, R&D systems, Inc. Minneapolis, USA). The ELISA was performed according to the manufacturer’s instructions.
Statistical analysis
Quantitative analyses were performed in triplicate in at least two different assays. Two-way ANOVA and Bonferroni test were performed using GraphPad Prism 5 (GraphPad Software, La Jolla, CA, USA). Differences with p < 0.05 were considered statistically significant.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
HKPO, NHRM, VVR and GSL participated in the study design; HKPO performed the infection and qPCR experiments. JRL and JCFA performed the confocal microscopy analysis; LFM and HKPO handled the samples and clinical data of patients and performed the ELISA assays; HKPO, VVR and NHRM analyzed data, performed the statistical analyses and reviewed the manuscript. NHRM and GSL coordinated the study and the writing of the manuscript. All authors read and approved the final manuscript.