The Antiviral Cytomegalovirus Inducible Gene 5/Viperin Is Expressed in Atherosclerosis and Regulated by Proinflammatory Agents
Objective— Inflammatory processes play an important role in atherosclerosis, and increasing evidence implies that microbial pathogens and proinflammatory cytokines are involved in the development and activation of atherosclerotic lesions. To find new inflammatory genes, we explored the vascular transcriptional response to an activator of innate immunity bacterial lipopolysaccharides (LPSs).
Methods and Results— Gene arrays identified the cytomegalovirus-inducible gene 5 (cig5)/viperin among the genes most potently induced by LPS in human vascular biopsies. Viperin was expressed by endothelial cells in atherosclerotic arteries and significantly elevated in atherosclerotic compared with normal arteries. In culture, cytomegalovirus infection, interferon-γ, and LPS induced viperin expression.
Conclusion— Viperin is expressed in atherosclerosis and induced in vascular cells by inflammatory stimuli and cytomegalovirus infection. The putative functions of viperin in atherosclerosis may relate to disease-associated microbes.
Development of atherosclerosis is associated with vascular inflammation. Recruited macrophages produce proinflammatory cytokines such as tumor necrosis factor-α (TNF-α) and interleukin-1β (IL-1β), whereas T cells activated by disease-associated antigens in lesions secrete interferon-γ (IFN-γ).1 The inflammatory milieu in the vessel exerts strong effects on the vascular endothelial and smooth muscle cells. They respond by expressing adhesion molecules, cytokines, and chemokines, which contribute to the atherosclerotic process.1 Clinical manifestations of atherosclerosis, such as myocardial infarction, appear to be triggered by this inflammatory process.2
Although it is well established that atherosclerosis is an inflammatory disease, the initiating mechanisms are not fully understood. Several contributing factors have been proposed, including oxidized lipoproteins, bacteria, and viruses. Chlamydia pneumoniae and cytomegalovirus (CMV) have been detected in human atherosclerotic lesions, and epidemiological and animal studies suggest an association between these pathogens and disease.3–6 Microbial products ligating pattern recognition receptors may also contribute to plaque inflammation.7
To investigate the vascular response to inflammation, we used gene arrays followed by single-gene studies. In this way, the CMV inducible gene 5 (cig5)/viperin, a virus-inducible antiviral protein, was identified as a putative culprit molecule in vascular inflammation and atherosclerosis.
The studies were approved by the ethical committee at the Karolinska Hospital and the ethics committee for animal experiments in Stockholm. Patients were included after informed consent.
A total of 53 patients scheduled for carotid endarterectomy and 13 patients scheduled for nephrectomy were included. Nine biopsies from the atherosclerotic carotid and 7 from the renal artery were placed in Dulbecco’s modified Eagle’s medium (D-MEM)/F12 (Gibco) enriched with 30 mg/mL human albumin (Biovitrum AB), and incubated with or without lipopolysaccharide (LPS) from Escherichia coli, serotype O55:B5 (Sigma Chemical) at 100 ng/mL for 6 hours. A portion of 3 of the renal artery biopsies was incubated with or without LPS for 24 hours. The incubated biopsies and unincubated 44 atherosclerotic and 6 renal artery biopsies were frozen. Serum was also obtained from 44 random reference patients free from medication and evidence of cardiovascular disease.
Fourteen apolipoprotein E-deficient (apoE−/−) mice were euthanized using CO2 anesthesia at 20 weeks of age. The aorta was freed from connective tissue under a dissection microscope. An atherosclerotic segment from the ascending aorta and a macroscopically normal segment from the descending aorta from each vessel were frozen. From 7 apoE−/− and 7 C57BL/6 mice euthanized at 25 weeks of age, the aorta was frozen.
Human coronary artery smooth muscle cells (CVSMCs; Clonetics) were cultured in smooth muscle medium-2 (CC-3181 and CC-4149; Clonetics). Human umbilical vein endothelial cells (HUVECs; Clonetics) were maintained in endothelial cell medium-2 (CC-3156 and CC-4176; Clonetics). The following incubations were performed: medium only, 100 ng/mL LPS (Sigma), 10 ng/mL TNF-α, IL-1β, and IFN-γ (PeproTech), a mix of TNF-α, IL-1β, and IFN-γ (each 10 ng/mL), infection at a multiplicity of infection between 0.1 and 1 with the endothelial-adapted CMV strain VR1814 (a generous gift from Dr Giuseppe Gerna, University of Pavia, Italy), the laboratory strain AD169, Towne, or mock infection with supernatants from uninfected cells. Viral titers were determined as described previously.8
RNA was isolated using the RNeasy extraction kit including a deoxyribonuclease step (Qiagen), quality controlled using an Agilent Bioanalyzer 2100, and reverse-transcribed to cDNA. mRNA levels were assessed by real-time RT-PCR. Primers and probes for human TNF-α, IL-1β, viperin, and murine viperin were designed using PrimerExpress software (Applied Biosystems; sequences available on request) and for β-actin9 and cyclophilin10 as described. Monocyte chemoattractant protein-1 and IL-8 reagents were purchased from Applied Biosystems. Samples were analyzed using an ABI Prism 7700 Sequence Detector. Results were normalized to cyclophilin in the human and β-actin in the murine experiments.
RNA with a 28S/18S ratio >2.0 from 1 renal artery and from 8 pooled carotid lesions were analyzed using Affymetrix gene expression arrays U95A according to manufacturer instructions.
Enzyme-Linked Immunosorbent Assay
Serum from patients included in this study was tested for human CMV (HCMV)–specific IgG and IgM in an enzygnost anti-HCMV/IgG ELISA and an enzygnost anti-HCMV/IgM ELISA (Behring).
Acetone-fixed cryostat sections were incubated with 2.5% normal horse serum for 30 minutes, followed by rabbit anti-viperin,12 diluted 1:3000 at 4°C overnight. Subsequently, the ImmPRESS Universal Antibody anti-rabbit Immunoglobulin Kit (Vector Laboratories) was used according to manufacturer instructions. Staining for von Willebrand factor was performed as described previously.7
The Mann-Whitney U test was used and P<0.05 considered significant. Values are expressed as mean±SEM. For CMV prevalence, 2-group χ2 tests were used.
Upregulation of CMV Induced Genes on LPS Stimulation of Arterial Tissue
The transcriptional response to LPS of a nonatherosclerotic human renal artery and pooled carotid atherosclerotic lesions were evaluated using Affymetrix global expression arrays. For the renal artery, 6828 of 12558 (54%) transcripts gave a detectable signal. The expression of 276 transcripts increased and 258 decreased after LPS stimulation. For the atherosclerotic biopsies, 5580 of 12558 (44%) transcripts were detected. The expression of 283 transcripts increased and 182 decreased after LPS stimulation. Cig5 and cig49 were among the 3 most strongly induced in the normal and atherosclerotic biopsies (Table).
Viperin Expressed in Human Atherosclerotic Lesions and Upregulated by LPS in Human Arteries
Real-time RT-PCR showed 3.8-fold higher viperin mRNA levels in human atherosclerotic lesions than in normal arteries (4.09±0.621 versus 1.06±0.199; P=0.012). After LPS stimulation, viperin mRNA increased in renal arteries (19.8±5.14 versus 1.19±0.271; P=0.009) and carotid lesions (20.1±5.37 versus 2.42±0.357; P=0.006). The viperin protein was expressed in human atherosclerosis but not in normal arteries (Figure 1) and induced by in vitro treatment of renal arteries with LPS (Figure 1). Immunohistochemistry revealed viperin in the endothelium of human atherosclerotic lesions, which were of type V–VI, but not in nonatherosclerotic renal arteries (Figure 2). The prevalence of CMV infection was higher in patients (40 of 44; mean age 71.9 years) compared with a reference group (31 of 44; mean age 62.0 years; P=0.0012). No significant difference in viperin expression was observed between anti-CMV IgG-positive and IgG-negative individuals in the patient group (P=0.49). No patients were positive for anti-CMV IgM.
Murine Viperin Expressed in Atherosclerotic Lesions of ApoE−/− Mice
RNA was isolated from atherosclerotic and nonatherosclerotic segments of 20-week-old apoE−/− mice. Lesions in apoE−/− mice of this age in our colony are usually advanced atheromata without evidence of plaque rupture. Murine viperin mRNA was substantially higher in the atherosclerotic segments compared with nonatherosclerotic ones from the same group of mice (18.2±2.3 versus 4.5±0.83 arbitrary units; P<0.0001). A similar difference in murine viperin expression was seen between whole aortas from 25-week-old apoE−/− mice compared with C57BL/6 mice (data not shown).
LPS, IFN-γ, and CMV Infection Induced Viperin in HUVECs and Coronary Artery Smooth Muscle Cells
Viperin protein was induced in HUVECs and coronary artery smooth muscle cells (CVSMCs) in response to LPS, CMV, and IFN-γ but not in response to TNF-α or IL-1β (Figure 1). Both cell types also displayed strong mRNA induction in response to LPS, a cytokine mix (Figure 3a), or CMV (Figure 3b). No difference was observed between mock and uninfected cells. After addition of CMV to cultures, changes in cell morphology became visible in the light microscope after 72 to 96 hours (ie, ≈48 hours after viperin induction [data not shown]).
In the present report, we used gene array to explore the human vascular transcriptional response to an activator of innate immunity: LPS. The antiviral protein viperin was identified as one of the most profoundly induced genes. We subsequently detected viperin expression in human and murine atherosclerosis. mRNA levels were significantly higher in lesions than in normal arteries, and the protein was detected in the endothelium of human atherosclerotic lesions but not in normal arteries. Viperin expression could be induced by LPS, IFN-γ, and CMV in human endothelial and smooth muscle cells and was also induced by LPS in normal human arteries.
The strong induction of cig5/viperin and cig49 in response to LPS in human vessels suggests that inflammatory stimulation may promote a similar response in the artery as CMV infection. Although cig49 is yet poorly characterized, cig5/viperin is conserved between species12,13 and known to efficiently inhibit CMV infection when overexpressed in human fibroblasts.12 Viperin contains a motif associated with protein radical formation14 and biosynthesis of cofactors,15 which may influence the antimicrobial defense. Hence, viperin may be important in the local defense against CMV, a putative atherosclerosis-related pathogen.
In our experiments, viperin expression could be induced by several different stimuli (ie, by LPS, CMV, and IFN-γ but not by TNF-α or IL-1β). Multiple ways of induction have also been demonstrated by others.12,13,16,17 The induction by different viruses and by LPS16,18 suggests an involvement of pathogen-associated molecular pattern receptors capable of initiating responses against evolutionarily distant stimuli.13 Interestingly, Fas-dependent cell death also promotes activation of an antiviral response. In this way, viperin induction may synergize with cytotoxic attack in the vascular defense against CMV and other viral infections that may play a role in vascular disease. However, the induction pathway of viperin is not known in detail. Because TNF-α and IL-1β did not induce viperin, it is not likely that nuclear factor κB activation alone upregulates viperin. Although human data are limited, given that multiple agents are capable of viperin induction, most likely, viperin is also induced during inflammatory processes other than CMV defense. In support of this, the viperin levels in biopsies from atherosclerotic lesions did not correlate to CMV seropositivity. However, the prevalence of anti-CMV IgG was higher in individuals with symptomatic atherosclerotic disease than in a healthy reference group. Importantly, a primary CMV infection was not detected in any patients.
We detected viperin in the endothelium of human atherosclerotic lesions but not in other cell types or in normal arteries. Viperin induction after CMV infection was 100-fold stronger in cultured vascular endothelial than in smooth muscle cells (Figure 3b). In contrast, the response to LPS or proinflammatory cytokines was similar in HUVECs as in CASMCs (Figure 3a). We therefore conjecture that the pathways of induction can differ between cell types, with CMV infection being a strong activator in the endothelium. The detection of viperin in atherosclerotic lesions may thus be linked to CMV but also to an active immune status in the lesion. Consequently, the role of viperin in vascular inflammation needs to be further evaluated.
In conclusion, this study shows that viperin, an evolutionarily conserved antiviral component of the innate immune response, is expressed in atherosclerotic arteries and strongly induced in vascular cells by inflammatory stimuli and CMV infection. The putative functions of viperin in atherosclerosis may relate to disease-associated microbes. Further studies are needed to determine its pathogenic importance.
This study was supported by the Swedish Medical Research Council (grants 6816 and 2042), the Swedish Heart-Lung Foundation, the AFA Health Fund, and the Swedish Health Care Sciences Postgraduate School at Karolinska Institutet. We thank Dr Afsar Rahbar, Center for Molecular Medicine, Karolinska Institute, for help with the ELISA.
Consulting Editor for this article was Peter Libby, MD, Brigham and Women’s Hospital, Boston, Mass
- Received October 15, 2004.
- Accepted April 25, 2005.
Hansson GK. Immune mechanisms in atherosclerosis. Arterioscler Thromb Vasc Biol. 2001; 21: 1876–1890.
van der Wal AC, Becker AE, van der Loos CM, Das PK. Site of intimal rupture or erosion of thrombosed coronary atherosclerotic plaques is characterized by an inflammatory process irrespective of the dominant plaque morphology. Circulation. 1994; 89: 36–44.
Streblow DN, Orloff SL, Nelson JA. Do pathogens accelerate atherosclerosis? J Nutr. 2001; 131: 2798S–2804S.
Nicholson AC, Hajjar DP. Herpes virus in atherosclerosis and thrombosis: etiologic agents or ubiquitous bystanders? Arterioscler Thromb Vasc Biol. 1998; 18: 339–348.
Edfeldt K, Swedenborg J, Hansson GK, Yan ZQ. Expression of toll-like receptors in human atherosclerotic lesions: a possible pathway for plaque activation. Circulation. 2002; 105: 1158–1161.
Wentworth BB, French L. Plaque assay of cytomegalovirus strains of human origin. Proc Soc Exp Biol Med. 1970; 135: 253–258.
Wuttge DM, Zhou X, Sheikine Y, Wagsater D, Stemme V, Hedin U, Stemme S, Hansson GK, Sirsjo A. CXCL16/SR-PSOX is an interferon-gamma-regulated chemokine and scavenger receptor expressed in atherosclerotic lesions. Arterioscler Thromb Vasc Biol. 2004; 24: 750–755.
Chin KC, Cresswell P. Viperin (viperin), an IFN-inducible antiviral protein directly induced by human cytomegalovirus. Proc Natl Acad Sci USA. 2001; 98: 15125–15130.
Boudinot P, Riffault S, Salhi S, Carrat C, Sedlik C, Mahmoudi N, Charley B, Benmansour A. Vesicular stomatitis virus and pseudorabies virus induce a vig1/cig5 homologue in mouse dendritic cells via different pathways. J Gen Virol. 2000; 81: 2675–2682.
Sofia HJ, Chen G, Hetzler BG, Reyes-Spindola JF, Miller NE. Radical SAM, a novel protein superfamily linking unresolved steps in familiar biosynthetic pathways with radical mechanisms: functional characterization using new analysis and information visualization methods. Nucleic Acids Res. 2001; 29: 1097–1106.
Boudinot P, Massin P, Blanco M, Riffault S, Benmansour A. vig-1, a new fish gene induced by the rhabdovirus glycoprotein, has a virus-induced homologue in humans and shares conserved motifs with the MoaA family. J Virol. 1999; 73: 1846–1852.
Zhu H, Cong JP, Shenk T. Use of differential display analysis to assess the effect of human cytomegalovirus infection on the accumulation of cellular RNAs: induction of interferon-responsive RNAs. Proc Natl Acad Sci USA. 1997; 94: 13985–13990.