Background
Human cytomegalovirus (HCMV) is a ubiquitous beta-herpesvirus that affects 60–80% of the human population [
1]. The lytic replication cycle of HCMV is a temporally regulated cascade of events that is initiated when the virus binds to host cell receptors. Upon entry into the cell, the viral DNA translocates to the nucleus, where expression of viral immediate early (IE), early and late genes occurs in a stepwise fashion [
2]. While generally asymptomatic in immunocompetent individuals, primary HCMV infection may cause infectious mononucleosis and has been associated with atherosclerosis and coronary restenosis [
3,
4]. Furthermore, HCMV is the leading contributor of congenital viral infections in the United States and Europe, causing cytomegalic inclusion disease, pneumonia and severe neurological anomalies in infected neonates [
5‐
7].
Like other herpesviruses, HCMV establishes lifelong latency in its host from which reactivation can occur and cause severe and fatal disease in immunocompromised individuals [
8]. Cellular immune responses (MHC class I-restricted T-cells and natural killer (NK) cells) appear to be an important factor in both the control of acute infections and the establishment and maintenance of viral latency in the host [
9‐
14]; however, the mechanisms by which T-cells affect HCMV replication are currently undefined. While cytotoxic T-cell activity has been shown to correlate with recovery from HCMV infection in patients [
15,
16], recent studies suggest that immune cytokines such as tumor necrosis factor-α and interferons (IFNs) may have direct inhibitory effects on HCMV replication [
17,
18]. In particular, the involvement of IFNs as a means of curtailing viral replication without cellular elimination is consistent with the hypothesis that cytokines produced by activated immune cells play a direct role in the control of viral infections [
19‐
21].
Type I IFNs (IFN-α and IFN-β) and type II IFN (IFN-γ) are important components of the host immune response to viral infections. IFN-α and IFN-β are produced by most cells as a direct response to viral infection [
22‐
24], while IFN-γ is synthesized almost exclusively by activated NK cells and activated T-cells in response to virus-infected cells [
25]. Both types of IFNs achieve their antiviral effects by binding to their respective receptors (IFN-α/β or IFN-γ receptors), resulting in the activation of distinct but related Janus kinase/signal transducer and activator of transcription (Jak/STAT) pathways. The result is the transcriptional activation of IFN target genes and the synthesis of a number of proteins that interfere with viral replication (reviewed in [
26]). Although IFNs are effective inhibitors of viruses such as vesicular stomatitis virus and encephalomyocarditis virus [
26], almost all RNA and DNA viruses have evolved mechanisms to subvert the host IFN response [
21,
26,
27]. For example, HCMV inhibits IFN-stimulated antiviral and immunoregulatory responses at multiple steps [
24,
28‐
32]. Likewise, the herpes simplex virus (HSV-1) protein ICP34.5 [
33], the influenza A virus NS1 protein [
34], the simian virus-5 V protein [
35], the Sendai virus C protein [
36], the hepatitis C virus (HCV) NS5A and E2 proteins [
37] and the Ebola virus VP35 protein [
38] have all been shown to block IFN-mediated responses in infected cells. However, several studies have shown that viruses normally resistant to the effects of type I or type II IFNs separately, are susceptible to IFNs when used in combination. For example, IFN-α/β and IFN-γ synergistically inhibit the replication of HSV-1 both
in vitro and
in vivo [
20]. In addition, recent reports have indicated that IFNs used in combination have a synergistic antiviral activity against severe acute respiratory syndrome-associated coronavirus (SARS-CoV) [
39], HCV [
40] and Lassa virus [
41].
In the present study, we examined the effects of IFN-α, IFN-β and/or IFN-γ on HCMV replication in human foreskin fibroblasts (HFFs). Treatment of HFFs with IFN-α, IFN-β or IFN-γ separately inhibited HCMV replication by ≤ 40-fold in both plaque reduction and viral growth assays. In contrast, treatment with IFN-α and IFN-γ or IFN-β and IFN-γ inhibited HCMV replication 10–20 times greater than that achieved by each IFN separately. This effect was synergistic in nature and the mechanism of inhibition may involve, at least in part, the regulation of IE gene expression. As with HSV-1 [
20], we have found that when used in combination, both type I and type II IFNs potently inhibit the replication of HCMV
in vitro.
Discussion
The immune response to viral infection is responsible for preventing viral dissemination and uncontrolled replication within the host. Following viral infection, type I IFNs are secreted by infected cells and function to induce an antiviral state in neighboring uninfected cells. Infiltrating immune cells, such as NK cells and macrophages, secrete numerous chemokines and cytokines that contribute to the overall antiviral response. Upon activation of the adaptive immune response, T-cells can further add to the milieu of immune cytokines present at the site of viral infection by secreting additional cytokines, including IFN-γ. Although several studies have examined the effects of proinflammatory cytokines on HCMV replication
in vitro, these studies are limited as they only examine the effect of one type of cytokine on viral replication rather than examining cytokines in combination. In support of the latter, recent studies have shown that type I and type II IFNs function, in synergy, to inhibit both RNA and DNA viruses, including HCV [
41], SARS-CoV [
39], Lassa virus [
40] and HSV-1 [
20]. These studies may more accurately represent the
in vivo inflammatory response that results after viral infection. The results presented herein are consistent with this hypothesis and establish that type I (IFN-α and IFN-β) and type II (IFN-γ) IFNs synergistically inhibit the replication of HCMV.
In the present study we have demonstrated that combination treatment with type I and type II IFNs renders cells non-permissive to HCMV replication
in vitro. The inhibitory effect by IFN-α/β and IFN-γ was synergistic in nature (Table
2, Figure
2C, 2D) and the degree of inhibition was not matched by increasing the concentrations of each individual IFN (Table
1, Figure
2A). These results indicate that the observed IFN-induced antiviral effects are a direct result of the presence of two distinct types of IFNs. Moreover, inhibition of HCMV replication in cells treated with IFN-α/β and IFN-γ was observed in both HFF and embryonic lung fibroblasts (MRC5) (data not shown) infected with either Towne-GFP (see Methods) or another laboratory strain, AD169 (data not shown). The mechanism(s) by which HCMV replication is inhibited remains unclear. Type I and type II IFNs may synergize by acting on one or more different stages of the HCMV lytic cycle such as (1) viral attachment, (2) viral entry, (3) IE gene expression, (4) early gene expression, (5) DNA replication, (6) late gene expression, (7) virus assembly or (8) viral egress and maturation. To address the question of attachment and entry, PCR was used to amplify viral DNA from IFN-treated and vehicle-treated cultures shortly after infection. As previously observed [
20,
46], IFN treatment did not prevent viral entry into cells as indicated by equal PCR product yield from all treatment groups (Figure
4). These data indicate that IFNs exert their inhibitory effects at a step after viral attachment and entry.
Previously, Yamamoto,
et al. (1987) demonstrated that treatment of cells with both IFN-α and IFN-γ potently inhibits HCMV replication; however, this study neither determined whether the effect was synergistic nor identified the mechanism of inhibition. However, the authors suggested that IFN-mediated inhibition of HCMV might occur at or prior to early gene expression [
48]. Similarly, over the course of our experiments utilizing the Towne-GFP strain, it was noticed that very few cells expressed green fluorescent protein (GFP) when treated with IFN-α/β and IFN-γ together (data not shown). In this recombinant Towne strain, GFP expression is driven by the early promoter UL127. The lack of GFP-positive cells in IFN-α/β and IFN-γ-treated groups suggested to us that the synergistic antiviral activities mediated by type I and type II IFNs occurred at a stage prior to early gene expression. Previous, studies have shown that type I or type II IFN treatment can inhibit HCMV IE mRNA expression [
46] and/or HCMV IE protein expression [
45,
46]. Using real-time PCR, we showed that while IFN-α, IFN-β or IFN-γ treatment inhibited IE mRNA expression by 2–6 fold at 6 h p.i., combination IFN-α and IFN-γ or IFN-β and IFN-γ treatment inhibited IE mRNA expression by 6–11 fold. Of note, this inhibitory effect was abolished by 24 h p.i. (data not shown), suggesting that IE mRNA expression is delayed by IFN treatment. The observed decrease in viral IE mRNA expression was accompanied by a decrease in IE protein expression, as viral IE protein expression was reduced in HFFs treated with both type I and type II IFNs (Figure
6A). Furthermore, immunofluorescent microscopy of IE protein expression revealed that nearly 100% of vehicle- and individual IFN-treated cells expressed IE72/86 5 d p.i., as compared to 46% or 21% of cells treated with IFN-α and IFN-γ or IFN-β and IFN-γ, respectively (Figure
6B–6G). It appears that although individual IFN treatment results in a marginal inhibition in IE expression early in infection, the effect is not maintained as demonstrated by high viral titers at 4 d p.i. (Figure
3) and increased IE protein expression at 5 d p.i. (Figure
6A–6E). Additionally, HCMV cytopathic effect, characterized by enlarged cells containing intranuclear and cytoplasmic inclusions, increased over time in vehicle- and individual IFN-treated groups, while morphology was unchanged in cells treated with IFN-α/β and IFN-γ (data not shown). Collectively, these data suggest that the synergistic inhibition of HCMV replication by IFN-α/β and IFN-γ may involve, at least in part, the regulation of IE gene expression. The significance of an inhibitory block at this level is evident when the phenotype of IE1 mutant viruses is considered. Greaves and colleagues have demonstrated that HCMV IE1 mutants exhibit a diminished replication efficiency and a reduced ability to form plaques, as well as defective early gene expression [
47,
49,
50]. Interestingly, in the presence of both type I and type II IFNs, HCMV shows similar replication and gene expression defects. Although our data suggest that IE gene regulation contributes to the synergistic inhibition of HCMV replication by IFN-α/β and IFN-γ, other mechanisms may also affect this dramatic response. Accordingly, the decrease in IE protein levels exceeds that in IE mRNA levels in response to IFN-α/β and IFN-γ, suggesting that additional regulation at the level of translation, post-translational processing and/or protein stability may be involved. Delineating the other putative regulatory mechanisms that contribute to IFN-α/β and IFN-γ synergistic inhibition of HCMV replication is the focus of ongoing studies.
Type I IFNs (IFN-α and IFN-β) and type II IFN (IFN-γ) activate distinct but related Jak/STAT signal cascades resulting in the transcription of several hundred IFN-stimulated genes [
26]. Although similar genes are activated by all three IFNs, Der,
et al. (1998) have identified numerous genes differentially regulated by IFN-α, IFN-β or IFN-γ [
51]. In particular, IFN-β stimulation induces twice as many genes as compared to IFN-α. This differential regulation of IFN-induced genes may explain in part the fact that the level of inhibition observed in HFFs treated with both IFN-β and IFN-γ was consistently greater than that observed in cells treated with both IFN-α and IFN-γ, although both IFN-α and IFN-β bind to the same receptor. Similarly, when compared individually, IFN-β consistently inhibited HCMV replication and IE gene expression to levels greater than IFN-α. Therefore, to better understand the cellular factors involved in the synergistic inhibition of HCMV, the profile of IFN-stimulated genes present in cells treated with both type I and type II IFNs should be further examined.
Methods
Cells, viruses and interferons
HFFs (Viromed, Minneapolis, MN) were maintained in minimal essential medium (MEM) supplemented with 10% fetal bovine serum, penicillin G (100 U/ml), streptomycin (100 mg/ml), 2 mM L-glutamine, 1 mM sodium pyruvate and 100 μM non-essential amino acids at 37°C in 5% CO
2. HCMV strain RVdlMwt-GFP was propagated in HFFs as previously described [
52]. RVdlMwt-GFP, referred to as Towne-GFP throughout this manuscript, is a recombinant of HCMV strain Towne that expresses GFP under the control of the early promoter UL127. This virus was kindly donated by Mark F. Stinski and has been previously described [
53].
Recombinant human universal IFN-α, IFN-β and IFN-γ (PBL Biomedical Laboratories, New Brunswick, NJ) were added to cell cultures 12 h prior to HCMV infection and maintained after viral infection. Concentrations of 100 IU/ml of each IFN were used in all experiments unless stated otherwise.
Plaque reduction and viral replication assays
For plaque reduction assays, vehicle- and IFN-treated HFFs were infected with a fixed inoculum of Towne-GFP. After 2 h adsorption, the inoculum was removed and medium containing 1.0% methylcellulose (Fisher Scientific, Houston, TX) and the respective IFN(s) was added to the cells. Plaque numbers were determined 14 d later by fluorescent microscopy (Nikon TE300 inverted epifluorescent microscope, Nikon USA, Lewisville, TX).
For viral replication assays, vehicle- and IFN-treated HFFs were infected with Towne-GFP at a MOI of 2.5. After 2 h adsorption, the inoculum was removed, monolayers were washed twice with 1X PBS, and fresh IFN-containing medium was returned to each well. For GCV-treated groups, 100 μM GCV (Sigma, St. Louis, MO) was added to culture medium immediately following infection. One, 2, 3 or 4 d p.i. cells and medium were harvested and titers of infectious virus were determined by a microtiter plaque assay on HFFs [
20].
Synergy assays
To determine the degree of antiviral interaction between type I and type II IFNs, interaction indexes were calculated using the inequalities: d
a/D
a+d
b/D
b > 1 and d
a/D
a+d
b/D
b <1, where d
a and d
b are the IFN concentrations needed to jointly produce the effect under consideration, and D
a and D
b are the IFN concentrations capable of producing the effect on their own, termed isoeffective doses [
42]. Interaction index values of less than 1 indicate synergism, interaction index values greater than 1 indicate antagonism and interaction index values equal to 1 indicate additivity. Isobolograms were also generated to geometrically assess the degree of antiviral interaction between type I and type II IFNs, as previously described [
43]. Using the guidelines described by Berenbaum [
43], isoboles were generated for IC
95 values at various concentrations of IFN-α or IFN-β in the presence of various concentrations of IFN-γ. Concave isoboles are indicative of synergy while convex isoboles are indicative of an antagonistic effect (Figure
2B). For all synergy experiments, HCMV plaque reduction assays were conducted as described above.
Viral entry assay
Vehicle- and IFN-treated HFFs were inoculated with Towne-GFP at MOIs of 0.3, 1, 3, 10 or 30. After 2 h adsorption, the inoculi were removed, cells were washed twice with 1X PBS, and subsequently treated with 0.05% trypsin for 5 minutes to ensure the release of virus that had adhered but had not entered the cells. Cells were pelleted and washed twice with 1X PBS to remove trypsin and non-adhered virus. DNA was isolated from each sample by a standard phenol:chloroform DNA extraction procedure [
54], and HCMV-specific oligonucleotide primers were used to amplify a 373 bp product corresponding to exon 4 of the HCMV IE gene, as described previously [
55]. PCR products were resolved in a 2% agarose gel and imaged using an Alpha Innotech gel documentation system (Alpha Innotech, Corp., San Leandro, CA).
Real-time PCR
Vehicle- and IFN-treated HFFs were infected with Towne-GFP at a MOI of 2.5. Six h p.i., total RNA was prepared using a RNeasy Mini Prep kit (Qiagen, Inc., Valencia, CA) according to the manufacturer's instructions. Samples were treated with DNase I (Ambion, Inc., Austin, TX), RNA concentration and purity were determined spectrophotometrically (A
260/A
280) and 250 ng was reverse transcribed in a total volume of 20 μl using the iScript cDNA Synthesis Kit (Biorad, Hercules, CA) according to the manufacturer's instructions. For real-time PCR, 1 μl of cDNA was amplified in 1X iQ SYBR Green Supermix containing specific primer pairs using the iCycler iQ Real-Time PCR Detection System (Biorad). The optimal primer concentrations and sequences were as follows: 200 nM IE1, sense 5' CAAGTGACCGAGGATTGCAA 3', antisense 5' CACCATGTCCACTCGAACCTT 3' ; 200 nM IE2, sense 5' TGACCGAGGATTGCAACGA 3', antisense 5' CGGCATGATTGACAGCCTG 3' [
56]; 100 nM 18S rRNA, sense 5' GAGGGAGCCTGAGAAACGG 3', antisense 5' GTCGGGAGTGGGTAATTTGC 3'. All samples were run on the same plate where those for the internal control (18S rRNA) and those for the genes of interest were each run in triplicate, for each of 3 independent RNA preparations. PCR parameters were as follows: an initial step to denature at 95°C for 30 seconds followed by 40 cycles at 95°C for 15 seconds and anneal/extend at 60°C for 45 seconds. Following amplification, melt curves were generated to confirm the specificity of each primer pair with 80 cycles of increasing increments of 0.5°C beginning with 55°C for 30 seconds. Relative quantification of the target genes in comparison to the 18S reference gene was determined by calculating the relative expression ratio (R) of each target gene as follows: R = (E
target)ΔCT(vehicle-sample)/(E
18S)ΔCT(vehicle-sample) [
57]. Differences in gene expression between the IFN-treated cells and the vehicle-treated control cells were expressed as fold-inhibition.
Western blotting
Vehicle- and IFN-treated HFFs were infected with Towne-GFP at a MOI of 2.5. Twelve h p.i., the cells were harvested in 500 μl of 1X RIPA buffer containing a protease inhibitor cocktail (Roche Applied Science, Indianapolis, IN) and 1 mM PMSF. Lysates were sheared 3X with a 27G 1/2 needle and cell debris was pelleted by centrifugation at 14,000 r.p.m. at 4°C. Total protein concentrations from cleared supernatants were estimated with a Micro BCA™ Protein Assay Kit (Pierce, Rockford, IL), 50 μg of total protein were resolved on 10% SDS-polyacrylamide gels and transferred by blotting to PVDF membranes (Amersham Biosciences, Piscataway, NJ). Non-specific reactivity was blocked with 5% nonfat dried milk in Tris-buffered saline containing 0.1% Tween-20 (TBST) for 1 h at room temperature and blots were incubated for 1 h at room temperature with a polyclonal antibody that recognizes the HCMV IE proteins (IE72/86), kindly provided by Daniel N. Streblow [
58]. The blots were then washed in TBST and incubated with donkey anti-rabbit IgG conjugated to horseradish peroxidase (1:5000; Amersham Biosciences) for 1 h at room temperature. Antigen-antibody complexes were detected using an enhanced chemiluminescence system (Amersham Biosciences). Blots were subsequently washed in TBST and tested for immunoreactivity to a rabbit polyclonal antibody to human β-actin (Sigma; loading control).
Indirect immunofluorescence
Vehicle- and IFN-treated HFFs were infected with Towne-GFP at a MOI of 1.0. Five d p.i., cells were washed 3X with 1X PBS, fixed with 1:1 methanol/acetone for 10 minutes at room temperature, washed again with 1X PBS, and blocked with 4% BSA/PBS for 15 minutes at room temperature. Cells were incubated for 1 h at 37°C with a HCMV IE antibody (IE72/86 kD; Chemicon #MAB810, Temecula, CA) diluted 1:200 in 0.5% BSA/PBS. Cells were then stained with 1:50 Alexa Fluor 568-conjugated goat anti-mouse IgG F(ab')2 (Molecular Probes, Eugene, OR) for 30 minutes at 37°C, followed by a 2 minute incubation with 1 μM 4',6-diamidino-2-phenylindole, dihydrochloride (DAPI; Molecular Probes) at room temperature. Cells were coverslipped and mounted in Prolong Antifade mounting medium (Molecular Probes), visualized on a Zeiss Axio Plan II microscope (Thornwood, NY) and images were analyzed with deconvolution SlideBook™ 4.0 Intelligent Imaging software (Intelligent Imaging Innovations, Denver, CO). To determine the number of HCMV-infected cells, three fields of view (100X) for each treatment group were considered and the percent of IE-positive cells was calculated as: (average number of IE-stained cells/average number of DAPI-stained cells)×100.
Statistics
Data are presented as the means ± standard error of the means (sem). Data from IFN-treated groups were compared to vehicle-treated groups and significant differences were determined by one-way analysis of variance (ANOVA) followed by Tukey's post hoc t test (GraphPad Prism© Home, San Diego, CA).
Acknowledgements
This work was supported by the National Institutes of Health (AI054626, AI054238, RR018229, and CA08921; R.F.G.) and (HD045768; C.A.M.). Bruno Sainz is a recipient of a National Research Service Award from the NIH (AI0543818). The authors would like to thank Dr. Mark F. Stinski (University of Iowa, Iowa City, Iowa) for kindly supplying the recombinant virus Towne-GFP and Dr. Daniel N. Streblow (Oregon Health Sciences University, Portland, OR) for kindly donating the HCMV IE antibody. We also thank Dr. Aline Scandurro for critical review of this manuscript and Dr. Joseph Vaccaro and Joshua Costin for their expertise in statistical analyses. We are also indebted to Dr. David Woodhall for his expertise and assistance with HCMV propagation and plaque assays.
Competing interests
The author(s) declare that they have no competing interests.
Authors' contributions
BS and HL conceived of the study, participated in the experimental design, performed all experiments and drafted the manuscript. RG and CM participated in the coordination and design of the study. All authors read and approved the final manuscript.