MAJOR ARTICLE
Combination of RNA Interference and Virus Receptor Trap Exerts Additive Antiviral Activity in Coxsackievirus B3-induced Myocarditis in Mice Elisabeth A. Stein,1 Sandra Pinkert,1 Peter Moritz Becher,2 Anja Geisler,1 Heinz Zeichhardt,3 Robert Klopfleisch,4 Wolfgang Poller,5 Carsten Tschöpe,5 Dirk Lassner,6 Henry Fechner,1,a and Jens Kurreck1,a Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
1
Department of Applied Biochemistry, Technische Universität Berlin, Institute of Biotechnology, 2Department of General and Interventional Cardiology, University Heart Center Hamburg Eppendorf, 3Campus Benjamin Franklin, Institute for Virology, Charité-Universitätsmedizin Berlin, 4Institute of Veterinary Pathology, Freie Universität Berlin, 5Campus Benjamin Franklin, Department of Cardiology and Pneumology, Charité-Universitätsmedizin Berlin, and 6IKDT Institut Kardiale Diagnostik und Therapie GmbH, Berlin, Germany
Background. Coxsackievirus B3 (CVB3) is a major heart pathogen against which no therapy exists to date. The potential of a combination treatment consisting of a proteinaceous virus receptor trap and an RNA interferencebased component to prevent CVB3-induced myocarditis was investigated. Methods and Results. A soluble variant of the extracellular domain of the coxsackievirus-adenovirus receptor (sCAR-Fc) was expressed from an adenoviral vector and 2 short hairpin RNAs (shRdRp2.4) directed against CVB3 were delivered by an adeno-associated virus (AAV) vector. Cell culture experiments revealed additive antiviral activity of the combined application. In a CVB3-induced mouse myocarditis model, both components applied individually significantly reduced inflammation and viral load in the heart. The combination exerted an additive antiviral effect and reduced heart pathology. Hemodynamic measurement revealed that infection with CVB3 resulted in impaired heart function, as illustrated by a drastically reduced cardiac output and impaired contractility and relaxation. Treatment with either sCAR-Fc or shRdRp2.4 significantly improved these parameters. Importantly, the combination of both components led to a further significant improvement of heart function. Conclusions. Combination of sCAR-Fc and shRdRp2.4 exerted additive effects and was significantly more effective than either of the single treatments in inhibiting CVB3-induced myocarditis and preventing cardiac dysfunction. Keywords.
viruses; myocarditis; RNA interference; gene therapy; coxsackievirus.
Coxsackievirus B3 (CVB3) is a plus-stranded RNA virus belonging to the picornavirus family, commonly identified as the infectious agent associated with acute and chronic myocarditis [1]. Strong evidence suggests that dilated cardiomyopathy can develop upon chronic
Received 17 June 2014; accepted 27 August 2014; electronically published 5 September 2014. Presented in part: European Society of Gene and Cell Therapy, Madrid, Spain, 25–28 October 2014. a H. F. and J. K. have contributed equally to this work and thus share last authorship. Correspondence: Jens Kurreck, PhD, Technische Universität Berlin, Institute of Biotechnology, Department of Applied Biochemistry, TIB 4/3-2, Gustav-MeyerAllee 25, 13355 Berlin, Germany ([email protected]) The Journal of Infectious Diseases® 2015;211:613–22 © The Author 2014. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: [email protected]. DOI: 10.1093/infdis/jiu504
persistence of the infection. This disease may finally progress to terminal heart failure and frequently requires heart transplantation. Although treatment with interferon β was shown to eliminate cardiotropic viruses and improve heart function in patients with myocardial persistence of viral genomes [2], no virus-specific treatment for CVB3 infections exists to date. In recent years, the focus in the development of antiviral drugs has shifted from pharmacologically active small molecule drugs to biological therapeutics [3]. In the case of CVB3, soluble variants of the coxsackievirusadenovirus receptor (sCAR) as a decoy were found to inhibit CVB3 infection in vitro and in vivo [4–8]. Dimeric sCAR fused to the immunoglobulin Fc-region (sCAR-Fc) was shown to neutralize the virus in the absence of undesirable side effects [5, 9]. We recently demonstrated that the in vivo expression of sCAR-Fc RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
613
MATERIAL AND METHODS Coxsackievirus B3
The cardio-virulent, genetically characterized Nancy strain of CVB3 was used for all in vitro and in vivo experiments. Virus propagation and titration on HeLa cells were described previously [14]. Vectors
For the expression of sCAR-Fc, the doxycycline (Dox)-inducible adenoviral vector AdG12 was used [7]. The same vector without Dox was used as control. The expression of shRdRp2 and shRdRp4 (shRNAs directed against the CVB3 RNA dependent RNA polymerase) was accomplished via a self-complementary AAV vector of serotype 2 (AAV2-shRdRp2.4) or 9 (AAV2.9shRdRp2.4) [14]. Vector maps are shown in Figure 1. As a control, the same vector type for the expression of shRNA directed against GFP (shGFP), was used (AAV2-shGFP / AAV2.9shGFP) [14]. Cell Cultures, Plaque and XTT Cell Viability Assays, RNA Isolation and cDNA Synthesis, Quantitative RT-PCR, and IgG ELISA
HeLa cells were cultured in Dulbecco modified Eagle medium (Gibco BRL, Karlsruhe, Germany) supplemented with 5% FCS, 1% MEM nonessential amino acids, 2% HEPES buffer, 0.1% Gentamycin, and 1% penicillin/streptomycin (all from 614
•
JID 2015:211 (15 February)
•
Stein et al
Figure 1. Schematic representation of the vectors used for the expression of sCAR-Fc, shRdRp2, and shRdRp4. A, Structure of the sCAR-Fcexpressing adenoviral vector, AdG12. Two expression cassettes for the constitutive expression of the reverse tetracycline transactivator (rtTA) and the doxycycline-dependent expression of sCAR-Fc, respectively, were inserted into the E1 region between nucleotide positions 453 and 3333 of an E1- and E3-deleted adenovirus 5 backbone. The same vector, without addition of Dox, was used as a control. CMV IE prom indicates the immediate-early CMV promoter; rtTA, reverse tetracycline-controlled transactivator rtTA-M2; tight1, doxycycline-dependent response promoter; sCAR-Fc, fusion protein of the soluble extracellular domain of human CAR and the human IgG Fc region; SV40 and bGH pA, polyadenylation signal of SV40 and bovine growth hormone, respectively; 5′ITR, left inverted terminal repeat of adenovirus type 5; 3′ITR, right inverted terminal repeat of adenovirus 5. B, Structure of the AAV vector used for the U6 promoter (U6 Pro) driven expression of shRdRp2 and shRdRp4 (AAV-shRdRp). Two expression cassettes were inserted in a head to head orientation. The inverted terminal repeats (ITRs) are derived from AAV2. The 3′ITR carries a deletion of the terminal resolution site (trs) and the D-sequence (positions 118–141), resulting in the packaging of self-complementary AAV vector genomes. Capsid proteins of serotype 2 were used for all in vitro experiments, whereas serotype 9 was used for in vivo studies. As a control an AAV vector expressing a shRNA against the nonrelated target GFP (AAVshGFP) was used. Abbreviations: AAV, adeno-associated virus; CAR, coxsackievirus-adenovirus receptor; CMV IE, cytomegalovirus immediateearly; GFP, green fluorescent protein; IgG, immunoglobulin G; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
GIBCO). Plaque assays were carried out as described elsewhere [7]. Cell viability was determined using Roche’s Cell Proliferation Kit II (XTT). RNA was isolated using Life Technologie’s TRIZOL reagent. Thermo Scientific’s RevertAid H Minus First Strand cDNA Synthesis Kit (Waltham, Massachusetts) was used for cDNA synthesis. Expression of inflammatory cytokines was analyzed utilizing the TaqMan Gene expression master mix and specific FAM-tagged TaqMan gene expression assays both from Life Technologies: IL6-Mm00446190_m1, IFNγ-Mm00801778_m1, TNFα-Mm0043258_m1. Detection of CVB3 specific RNA was carried out using Life Technologie’s TaqMan Gene expression master mix and a CVB3 specific primer/probe set. Human immunoglobulin G (IgG) enzymelinked immuosorbent assay (ELISA; Bethyl Laboratories Inc.) was used for detection of sCAR-Fc. In Vivo Mouse Myocarditis Model
Each group consisted of 6 mice. AAV2.9-shRdRp or AAV2.9shGFP vector was injected into the right jugular vein of
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
reduced CVB3 infection, inflammation, and dysfunction of the heart in mice [7]. However, although sCAR-Fc was highly efficient to abrogate CVB3-induced myocarditis when applied prior to infection, the inhibitory effect after onset of the CVB3 infection was limited. Another antiviral approach that gained importance in recent years is the inhibition of viruses by means of RNA interference (RNAi) [10]. We and others have shown that CVB3 can efficiently be inhibited by RNAi in vitro and in vivo [11–14]. However, complete elimination of the virus was not achieved. As both approaches were insufficient when applied individually, we reasoned that the combination of sCAR-Fc as an extracellular trap of CVB3 and RNAi as a mechanism that degrades viral RNA inside cells might exert additive antiviral effects. In a proof-of-concept study we demonstrated that the concurrent application of chemically synthesized siRNAs and sCAR-Fc expressed from an adenoviral vector (AdV) acted synergistically in a cell culture model [15]. In the present study, we translated this strategy to a preclinical in vivo setting of CVB3-induced myocarditis. Both antiviral components were delivered by viral vectors and the efficiency of the double treatment was studied in a CVB3-induced mouse myocarditis model. Our approach demonstrates that the combination treatment results in additive protective effects by inhibiting virus replication, preventing inflammation of heart tissue and improving cardiac function.
6-week-old male BALB/c mice at the dose of 1 × 1012 vector genomes (vg)/ mouse. Eight days later, mice were injected with 1 × 1010 particles of AdG12 into the left jugular vein. Another 2 days later, mice were infected with 5 × 104 plaque-
forming units (PFU) of CVB3 intraperitoneally. Dox (200 µg/ mL) for the expression of sCAR-Fc was given orally by drinking water starting 1 day after CVB3 infection. After 1 week, animals were anesthetized (0.8–1.2 g/kg urethane; 0.05 mg/kg
RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
615
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
Figure 2. Effect of sCAR-Fc and shRdRp on CVB3 replication in vitro. A, Plaque assay. HeLa cells were transduced with AdG12 (MOI 5) and the respective AAV2 vector at 2 × 104 vg/cell. Where indicated, Dox was added to the media at a final concentration of 1 µg/mL. 48 hours later, cell supernatants were preincubated with CVB3 (MOI 0.1) at 4°C for 30 minutes and added to the transduced cells for 30 minutes at 37°C. Subsequently, cells were overlaid with agarcontaining media, cultured at 37°C and 5% CO2 and stained with crystal violet 48 hours later. B, Two-step Plaque Assay. HeLa cells were transduced, cultured, and infected as in A. After infection, virus-containing media was replaced by fresh media. Virus replication was measured by viral plaque assay after another 24 hours of culture. The data presented in this figure as mean ± SD were obtained from three independent experiments, each performed in duplicate. **P < .01. C, XTT cell viability assay. HeLa cells were transduced with the indicated vectors and infected with CVB3 (MOI 0.1) as in A. Following infection, cells were cultured at 37°C and 5% CO2. Cell viability was determined at the indicated time points using the Roche Cell Proliferation Kit II (XTT) according to the manufacturer’s instructions. Data obtained in 2 independent experiments, each performed in quadruplicate, are shown as mean ± SD. Abbreviations: AAV, adeno-associated virus; CVB3, coxsackievirus B3; MOI, multiplicity of infection; PFU, plaque-forming unit; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; SD, standard deviation; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
were transduced and infected with CVB3 as described before. After infection, cells were cultured for another 24 hours and subsequently lysed. The number of infectious viral particles was then determined by plaque assay. Both sCAR-Fc and shRdRp reduced CVB3 replication by 1 to 2 log10 steps. The reduction of the CVB3 titer was significantly higher (approximately 4 log 10 steps) when both components were applied simultaneously (Figure 2B, Supplementary Table 3). The additive protective effect of shRdRp and sCAR-Fc was further underlined by the increase in cell viability as measured by XTT assays (Figure 2C). CVB3-infected cells that received no antiviral treatment were completely lysed 48 hours after infection. Treatment of cells with either of the 2 components resulted in a 50% reduction of cell viability compared to uninfected cells. The combination approach again exerted an additive antiviral effect and cell viability remained virtually at the same level as determined for the uninfected cells for up to 2 days.
Statistical Analysis
To prove the specificity of our approach, dose dependency assays were performed. The multiplicity of infection (MOI) of AdG12 was kept constant at 5 as higher doses completely prevented cell lysis. The dose of AAV2-shRdRp and its control vector, AAV2-shGFP, ranged from 103 to 2 × 104 vg/cell. As expected, the antiviral effect of AdG12 in the presence of Dox was independent of the doses of the added control vector AAV2-shGFP (Figure 3). Unspecific effects of the AAV vector expressing a shRNA can therefore be excluded for the concentration range under investigation. In contrast, the antiviral activity of AAV2-shRdRp increased with higher vector doses. This observation was made for the single RNAi-treatment (ie, sCAR-Fc expression was not induced by Dox), as well as for the combination treatment for which Dox was added to induce sCAR-Fc expression. Importantly, as found above for the singledose experiments, the additive antiviral effect was again clearly detectable for all AAV vector doses under investigation.
Graph Pad Prism 6.0 software was used for statistical analysis. In vivo data were statistically analyzed applying unpaired, 2-tailed Mann–Whitney U testing. For the in vitro data analysis the unpaired student t test was applied. Differences were considered significant at P < .05 (with * referring to .01 < P < .05 and ** referring to P < .01). RESULTS Additive Anti-CVB3 Effect Caused by sCAR-Fc and shRdRp In Vitro
In a first attempt to test for possible additive antiviral effects of sCAR-Fc and shRdRp2.4, the ability of both components to inhibit the generation of new virus progeny was investigated in cell culture by plaque assays (Figure 2A). HeLa cells were transduced with AdG12 (MOI 5) and AAV2-shRdRp or AAV2shGFP (2 × 104 vg/cell) and cultured in the presence or absence of Dox (1 µg/mL). In sum, 48 hours after transduction, cells were infected with CVB3 (MOI 0.1) and overlaid with agar containing media. Another 48 hours later, cells were stained with crystal violet and evaluated for number and size of plaques. Cells treated with noninduced AdG12 or AAV2-shGFP and infected with CVB3 were completely lysed comparable to the mock-treated, CVB3-infected cells. Both sCAR-Fc and shRdRp alone were capable of inhibiting CVB3 replication as approximately 50% of the surface remained covered by viable cells. The lowest number of virus plaques was observed for cells treated with both components simultaneously, illustrating the additive antiviral effect. To quantify the extent of the observed additive effect, 2-step plaque assays were conducted (Figure 2B). To this end, cells 616
•
JID 2015:211 (15 February)
•
Stein et al
The Additive Antiviral Effect Observed is Directly Dependent on the AAV Vector Dose Used In Vitro
Combined Application of sCAR-Fc and shRdRp Significantly Reduces Myocardial Inflammation and Virus Load in the Heart Compared to the Single Component Approach in a CVB3induced Mouse Myocarditis Model
After having observed a strong additive effect of the double treatment in vitro, the combination approach was tested in a mouse myocarditis model in vivo. In our previous study, induction of sCAR-Fc inhibited CVB3 virtually to completion when applied prior to the infection [7]. We therefore chose to induce the expression of sCAR-Fc postinfection to be able to investigate possible additive effects of the double treatment. Antiviral shRNAs were previously found to lead to a partial inhibition of CVB3 when applied prior to infection [14] and were employed in the same manner in the present study.
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
buprenorphine intraperitoneally), intubated, and artificially ventilated with a rodent ventilator type 7025 (Ugo Basile). Hemodynamic parameters were recorded via a microconductance catheter (1.2F) system (Scisense Inc.) in closed chest animals as described elsewhere [16]. Afterward, animals were killed, blood samples were taken, and organs harvested. For histopathological analysis, tissue samples were fixed in 4% neutral buffered formalin for 24 hours and embedded in paraffin wax. Sections (2 µm) were stained with hematoxylin and eosin (HE) according to standard protocols [17]. Evaluation of myocardial integrity and determination of myocarditis score were performed as described elsewhere in a double-blind way [18]. For molecular analysis and plaque assays, organ samples were frozen in liquid nitrogen and stored at −80°C. Plaque assays for determination of CVB3 titer in the heart were carried out as described elsewhere [14]. The investigation was performed in accordance with the principles of laboratory animal care and the German law of animal protection. Approval was obtained by the local ethics review board.
Animals were divided into 5 groups (Figure 4A): Shamoperated and mock-infected animals served as negative control, whereas sham-operated, CVB3-infected mice were used as positive control for the CVB3 myocarditis model. To analyze both, the single components as well as combined treatment, groups received either sCAR-Fc (AdG12 (+Dox) + AAV2.9-shGFP) or the antiviral shRNAs (AdG12 (−Dox) + AAV2.9-shRdRp) or both (AdG12 (+Dox) + AAV2.9-shRdRp). AAV2.9-shRdRp or AAV2.9-shGFP was intravenously injected into mice at the dose of 1 × 1012 vector copies per mouse. Eight days later, 1 × 1010 particles of AdG12 were applied via the same delivery route. Another 2 days later, mice were infected intraperitoneally with 5 × 104 PFU of CVB3. Mice were killed 8 days after the infection. In the animal groups, in which expression of sCAR-Fc was induced by oral administration of Dox, the level of sCAR-Fc was analyzed and found to be comparable for both groups (Figure 5A). The expression of comparable levels of active antiCVB3 shRNAs was confirmed by quantitative stemloop reverse transcription polymerase chain reaction (RT-PCR; Figure 5B). To investigate the inflammation level, the cardiac tissue of all animals was fixed and stained with HE 7 days postinfection, and the level of inflammation was graded (Figure 4B left; Supplementary Table 2C). For the negative control group, staining revealed no signs of inflammation or necrosis in any of the 6 animals. In contrast, all CVB3-infected, untreated mice were
Heart Function is Significantly Improved in the sCAR-Fc and shRdRp Treated Group Compared to Either of the Single Component Groups
Cardiac contractility is the most important clinical parameter that characterizes the function of the heart. Hemodynamic measurements for the determination of systolic and diastolic
RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
617
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
Figure 3. Dose-response relationship of the additive anti-CVB3 effect. HeLa cells were transduced with AdG12 (MOI 5) and AAV2-shRdRp or AAV2-shGFP at the indicated vg/cell and cultured in the presence or absence of Dox (1 µg/mL) as indicated. After 48 hours of culture, cell supernatants were preincubated with CVB3 (MOI 0.1) at 4°C for 30 minutes. Afterward, transduced cells were infected with these supernatants for 30 minutes. After infection, virus-containing media was replaced by fresh media. Virus replication was measured by viral plaque assay after another 24 hours of culture. Data shown were obtained from 2 independent experiments, each performed in duplicate and are presented as mean ± SD. Abbreviations: AAV, adeno-associated virus; CVB3, coxsackievirus B3; MOI, multiplicity of infection; PFU, plaque-forming unit; SD, standard deviation; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
classified with the highest myocarditis score of 4, which is characterized by the presence of mononuclear cell infiltration and myocyte necrosis in >20% of the heart tissue [18]. Single treatment with either sCAR-Fc or shRdRp2.4 reduced the myocarditis score to 3 or 2, in 3 out of 6 mice in each group. Most importantly, animals treated with both antiviral components showed substantial less inflammation in the heart compared to either of the single element approaches. In contrast to any other group, cardiac integrity was completely maintained in 4 out of 6 animals. Exemplary HE-stained heart sections are shown for each group in the right part of Figure 4B. Proinflammatory cytokines trigger the development of inflammation, which is a major event in the development of virus-induced myocarditis [19]. We investigated the levels of inflammatory cytokines in the heart tissue by RT quantitative PCR (qPCR; Figure 4C, Supplementary Table 2B). Following CVB3-infection, the proinflammatory cytokines interleukin 6 (IL-6), tumor necrosis factor α (TNF-α), and interferon γ (IFN-γ) were significantly up-regulated 39-fold, 6-fold, and 134-fold, respectively. Treatment of the animals with either sCAR-Fc or shRdRp led to reduced levels of all 3 cytokines. The decrease of the cytokine levels was even more pronounced in the double-treatment mice, supporting the data obtained from myocardial inflammation scoring. These data were further underlined by the viral load in cardiac tissue determined by plaque assay. For the sham-operated, CVB3-infected group, an average of 4.7 × 104 PFU/mg heart tissue was determined (Figure 4D left, Supplementary Table 2C). This titer was reduced by approximately 70% for the sCAR-Fc (1.35 × 104 PFU/mg heart tissue) and shRdRp (1.47 × 104 PFU/ mg heart tissue) treated mice, respectively. The reduction in viral load was significantly higher (approximately 95%) when both antiviral components were used simultaneously (2.54 × 10 3 PFU/mg heart tissue). CVB3-specific RT qPCR also confirmed these results (Figure 4D right). The highest level of genomic RNA was detected in cardiac tissue of shamoperated, CVB3-infected animals. The level was approximately 50% lower for the sCAR-Fc or shRdRp treated groups. For the combination group, the level of CVB3 genomes was further decreased to approximately 25% compared to the untreated group. Taken together, differences in viral load correlated well with pathohistological and inflammatory differences among the groups. Hence, the additive protective effect of sCAR-Fc and shRdRp compared to single treatment was further underlined by these results.
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
Figure 4. Influence of sCAR-Fc and shRdRp on heart inflammation and virus load in murine CVB3-induced myocarditis. A, Timeline, application scheme and group distribution of the in vivo approach. Ten days prior to infection, 1 × 1012 vector genomes of the respective AAV2.9 vector were injected intravenously via the jugular vein. Eight days later, 1 × 1010 particles of AdG12 were injected via the same delivery route. Two days thereafter, mice were infected intraperitoneally with 5 × 104 PFU of CVB3 and sCAR-Fc expression was induced as indicated 1 day later by addition of Dox (200 µg/mL) to the drinking
618
•
JID 2015:211 (15 February)
•
Stein et al
DISCUSSION
Figure 5. Serum level of sCAR Fc and expression of shRdRp2 and shRdRp4 in the heart 7 days postinfection. A, 7 days after infection, blood samples of all mice were taken, and serum was prepared. Serum level of sCAR Fc was measured using a human IgG specific sandwich ELISA kit. Signals obtained from sham-operated and CVB3-infected animals were subtracted as background. Values shown are mean ± SEM. B, Analysis of shRdRp2 and shRdRp4 expression in heart tissue of mice transduced with the AAV2.9 shRdRp vector in the myocarditis model. The processed siRNAs were detected via quantitative stemloop RT PCR 7 days after infection. Expression data were normalized to the expression of 18 S rRNA and the siRNA level in the group AdG12 (Dox) + shRdRp. Levels of the siRNAs were similar between this group and the group that received the combined treatment. Only background signals were detected for the group AdG12 (+Dox) + shGFP, proving PCR specificity for the detection of siRdRp2 and siRdRp4. Abbreviations: AAV, adeno-associated virus; CVB3, coxsackievirus B3; ELISA, enzyme-linked immunosorbent assay; IgG, immunoglobulin G; PCR, polymerase chain reaction; rRNA, ribosomal RNA; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; SEM, standard error of the mean; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
CVB3 is a major causative agent for acute and chronic myocarditis against which no specific antiviral treatment exists to date. In the present study, we report that combined treatment of CVB3-induced myocarditis with vector-delivered, virusneutralizing sCAR-Fc and shRNAs directed against the CVB3 genome are not antagonistic toward one another and exert additive antiviral effects compared to individual usage. Treatment with both components resulted in significantly reduced viral load in the heart. Consequently, the portion of heart tissue damaged (inflammation and necrosis) was strongly reduced in double-treatment animals compared to the single-treatment ones. Importantly, treatment with both antiviral components significantly improved hemodynamic parameters in CVB3-infected animals compared to mice that only received 1 active component. We have previously shown that sCAR-Fc efficiently inhibits CVB3 infection when given prior to the infection [7]. In the present study, generation of sCAR-Fc was induced after infection of the mice, that is, in a therapeutic manner. However,
Figure 4 continued. water. A positive control group (Sham + CVB3) was sham-operated and CVB3-infected. The negative control animals (Mock) were sham-operated and mock-infected with PBS. Analysis was carried out 7 days postinfection. (n = 6 for each group). B, Left: Myocarditis score of CVB3infected and sCAR-Fc- and/ or shRdRp-treated mice. Complete cardiac integrity is graded with a myocarditis score of 0, whereas the highest myocarditis score of 4 indicates massive inflammation in more than 20% of cardiac tissue. Right: Heart sections of mice were stained with hematoxylin and eosin. Arrows indicate areas of inflammation. Pictures shown are representative for each group. C, Expression levels of inflammatory cytokines in cardiac tissue of the different groups. The mRNA expression levels were analyzed 7 days after infection. Data were normalized to the expression level of 18 S rRNA and are presented as 2−ΔCt. Values shown are mean ± SEM for each group. *P < .05; **P < .01. (n = 6 for each group). D, Viral load in the heart. Left: Viral plaque titer of heart lysates as determined by plaque assay. The viral load in the heart is shown in PFU per mg heart tissue. Plaque titer of each heart sample was determined in 2 independent experiments each performed in duplicate. Right: Quantification of genomic CVB3 RNA in the heart tissue by quantitative RTPCR. Data were normalized to the expression of 18 S rRNA and are shown as 2−ΔCt. Values shown are mean ± SEM for each group. *P < .05; **P < .01. (n = 6 for each group). Abbreviations: AAV, adeno-associated virus; CVB3, coxsackievirus B3; IFNγ, interferon γ; IL-6, interleukin 6; MOI, multiplicity of infection; mRNA, messenger RNA; PBS, phosphate-buffered saline; PFU, plaque-forming unit; rRNA, ribosomal RNA; RT-PCR, reverse transcription polymerase chain reaction; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; SD, standard deviation; SEM, standard error of the mean; shRdRp, short hairpin RNA against RNA dependent RNA polymerase; TNFα, tumor necrosis factor α. RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
619
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
left ventricular (LV) function were therefore conducted by tip catheter at day 7 postinfection (Figure 6 and Supplementary Tables 1 and 2A). Because the heart rate was comparable between all groups, the data obtained from hemodynamic measurements can be compared with respect to the heart-protective effects of our antiviral treatment with sCAR-Fc and shRdRp. Infection with CVB3 resulted in impaired systolic and diastolic LV function, indicating a significant global cardiac dysfunction, as illustrated by a reduced cardiac output and impaired contractility and relaxation. A significant improvement of these parameters was achieved with either of the single treatments, sCAR-Fc or shRdRp. Most importantly, the combined application of both components again led to a further significant improvement of cardiac output and contractility parameters. These data illustrate the potential of the double-treatment with sCAR-Fc and shRdRp to excite additive antiviral effects in vivo even with regard to the most important parameters of heart function.
the application of sCAR-Fc involves some limitations. Viral particles only get trapped by sCAR-Fc extracellularly before they enter their target cells. This limitation is especially disadvantageous when treatment is initiated during an ongoing infection. Vector-delivered shRdRp led to a partial improvement of hemodynamic parameters during CVB3-caused myocarditis [14]. The underlying mechanism of RNAi is limited to the intracellular level, whereas the initial virus uptake is not inhibited. In addition, the therapeutic AAV vector will not transduce all cardiomyocytes, and CVB3 can replicate in cells that are not transduced. Due to the fast replication cycle of CVB3, the high number of virus progeny, and the inability to transduce all cardiomyocytes with AAV2.9-shRdRp, complete cardiac integrity was not preserved by the RNAi approach. 620
•
JID 2015:211 (15 February)
•
Stein et al
In the present study, we therefore chose to combine the extracellular virus trap strategy based on sCAR-Fc with the intracellular RNAi concept and wanted to investigate possible additive effects of the double-treatment. Combination therapies targeting viruses at different stages of their lifecycle are a common strategy to improve the efficiency of the antiviral treatment [20, 21]. Our approach transferred a previous cell culture study employing synthesized anti-CVB3 siRNAs [15] to the preclinical in vivo stage in the current study using a cardiotropic AAV vector for expression of shRdRp in addition to the adenovirus vector AdG12 for the expression of sCAR-Fc. In line with previous results, the shRNAs alone reduced the virus titer in the myocardium by approximately 70%, resulting in partial improvement of heart parameters [14]. Similarly, mice treated
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
Figure 6. Effect of sCAR-Fc and shRdRp treatment on CVB3-induced cardiac dysfunction. Seven days after CVB3 infection, mice were anesthetized (0.8–1.2 g/kg urethane; 0.05 mg/kg buprenorphine intraperitoneally), intubated, and artificially ventilated with a rodent ventilator type 7025. Heart parameters, such as heart rate, LV contractility (dP/dtmax), LV relaxation (dP/dtmin), and cardiac output (CO) as indexes of LV function were recorded via a microconductance catheter (1.2F) system in closed chest animals. Values shown are mean ± SEM for each group. *P < .05; **P < .01. (n = 6 for each group). For a complete list of the recorded cardiac parameters see Supplementary Table 1. Abbreviations: CVB3, coxsackievirus B3; LV, left ventricular; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; SEM, standard error of the mean; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
of either of the individual components. In a mouse CVB3 myocarditis model, we demonstrated that the treatment reduced virus load, tissue damage, and induction of inflammatory cytokines. As a consequence, cardiac function, which is known to be severely disturbed during acute CVB3 infection [27], was significantly improved as shown by hemodynamic measurements. No obvious adverse effects of the antiviral components were observed. Targeting the virus at 2 critical points of its replication cycle by application of both sCAR-Fc and shRdRp has thus the potential to develop into a novel treatment option for CVB3induced myocarditis. Furthermore, to the best of our knowledge, it represents the first described combination approach of an RNAi strategy and a proteinaceous antiviral component and may thus represent a general paradigm to treat severe viral infections. For example, the antiviral agents used here may also be combined with other components such as the inhibitor of CVB3 protease, which was developed more recently [28]. Supplementary Data Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.
Notes Acknowledgments. The authors thank Xiaomin Wang for vector production of AdG12 and Erik J. Wade for critical reading of the manuscript and helpful comments. Financial support. This work was supported by the Deutsche Forschungsgemeinschaft through grants [Ku 1436/6-1] to J. K. and [Fe 785/ 3-1, Fe785/2-1, FE 785/2-2] to H. F. and the Sonderforschungsbereich (SFB) TR19 to C. T., H. F., and W. P. H. Z. is grateful for the support from Otto-Kuhn-Stiftung im Stifterverband für die Deutsche Wissenschaft. Potential conflicts of interest. All authors: No reported conflicts All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References 1. Kandolf R, Klingel K, Zell R, et al. Molecular mechanisms in the pathogenesis of enteroviral heart disease: Acute and persistent infections. Clin Immunol Immunopathol 1993; 68:153–8. 2. Kuhl U, Pauschinger M, Schwimmbeck PL, et al. Interferon-beta treatment eliminates cardiotropic viruses and improves left ventricular function in patients with myocardial persistence of viral genomes and left ventricular dysfunction. Circulation 2003; 107:2793–8. 3. Fox JL. Antivirals become a broader enterprise. Nat Biotechnol 2007; 25:1395–402. 4. Lim BK, Choi JH, Nam JH, et al. Virus receptor trap neutralizes coxsackievirus in experimental murine viral myocarditis. Cardiovasc Res 2006; 71:517–26. 5. Yanagawa B, Spiller OB, Proctor DG, et al. Soluble recombinant coxsackievirus and adenovirus receptor abrogates coxsackievirus b3-mediated pancreatitis and myocarditis in mice. J Infect Dis 2004; 189:1431–9. 6. Dorner A, Grunert HP, Lindig V, et al. Treatment of coxsackievirusB3-infected BALB/c mice with the soluble coxsackie adenovirus
RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
621
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
with sCAR-Fc alone showed a reduction in viral heart titer of approximately 70% and partial restoration of heart function as in previous studies [7]. When combining both approaches, we did not observe any antagonistic effects. Importantly, we now demonstrate the extraordinary potential of both active components to exert additive antiviral effects against CVB3. The viral heart titer of double-treated mice was reduced by approximately 95% compared to untreated mice. Likewise, the reduction of inflammatory cytokines was strongest for the double-treatment group: 4 out of 6 mice of this group had no signs of inflammation or tissue damage in the heart, whereas this was not observed for any animal from a single-treatment group. Most importantly, the protective effects in terms of heart parameters, such as cardiac output and LV contractility and relaxation, was significantly stronger for the double-treated animals compared to mice that only received 1 component, proving the extraordinary potential of our combined approach. We chose to deliver both sCAR-Fc and shRdRp with viral vectors. Although sCAR-Fc was expressed from the adenoviral vector AdG12, an AAV vector of serotype 9 (AAV2.9-shRdRp) was used for in vivo shRNA delivery. Gene therapy approaches are strongly dependent on the safety of the viral vectors used. During the time course of our in vivo study, no obvious adverse effects were observed in the animals treated with either of the viral vectors. As the presence of AdG12 in the body is limited to several weeks, no long-term adverse effects are expected for its use [22]. Furthermore, the doxycycline-inducible system for the expression of sCAR-Fc, which enables a very tight regulation of expression, for example, in case of undesirable adverse effects, provides an additional safety feature of our application. For AAV vectors, long-term expression of the transgene for at least 1 year in the absence of adverse effects was reported [23]. Even though some groups reported cytotoxicity when expressing high amounts of shRNAs from AAV vectors [24], no such phenomena were observed in our study. Therefore, the vectors chosen for the delivery of the antiviral components were safe for our approach. However, the limited half-life of AdG12 delivered sCAR-Fc may become a limitation for its clinical use. To transfer our preclinical study to the clinical stage, sCAR-Fc might be applied as a recombinantly produced protein or from a different vector type (eg, AAV vectors) that enables stable long-term expression without evoking adverse effects. Previous studies have shown that treatment of mice with an AAV vector induces neutralizing antibodies, which make a second treatment with the same serotype inefficient [25]. This problem can be solved by using a different serotype to circumvent the immune response [26]. In conclusion, our proof-of-concept study demonstrates that the combination of sCAR-Fc as an extracellular virus trap and RNAi as an intracellular mechanism to degrade CVB3 genomes has an extraordinarily high antiviral effectiveness. The efficiency of the combination approach is significantly better than that
7.
8.
9.
10. 11.
12.
14.
15.
16.
622
•
JID 2015:211 (15 February)
•
Stein et al
17. Klopfleisch R, Gruber AD. Derlin-1 and stanniocalcin-1 are differentially regulated in metastasizing canine mammary adenocarcinomas. J Comp Pathol 2009; 141:113–20. 18. Szalay G, Sauter M, Hald J, Weinzierl A, Kandolf R, Klingel K. Sustained nitric oxide synthesis contributes to immunopathology in ongoing myocarditis attributable to interleukin-10 disorders. Am J Pathol 2006; 169:2085–93. 19. Leslie K, Blay R, Haisch C, Lodge A, Weller A, Huber S. Clinical and experimental aspects of viral myocarditis. Clin Microbiol Rev 1989; 2:191–203. 20. Li M, Li H, Rossi JJ. RNAi in combination with a ribozyme and TAR decoy for treatment of HIV infection in hematopoietic cell gene therapy. Ann N Y Acad Sci 2006; 1082:172–9. 21. Boutimah F, Eekels JJ, Liu YP, Berkhout B. Antiviral strategies combining antiretroviral drugs with RNAi-mediated attack on HIV-1 and cellular co-factors. Antiviral Res 2013; 98:121–9. 22. Wei K, Kuhnert F, Kuo CJ. Recombinant adenovirus as a methodology for exploration of physiologic functions of growth factor pathways. J Mol Med (Berl) 2008; 86:161–9. 23. Bish LT, Morine K, Sleeper MM, et al. Adeno-associated virus (AAV) serotype 9 provides global cardiac gene transfer superior to AAV1, AAV6, AAV7, and AAV8 in the mouse and rat. Hum Gene Ther 2008; 19:1359–68. 24. Grimm D, Streetz KL, Jopling CL, et al. Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways. Nature 2006; 441:537–41. 25. Moskalenko M, Chen L, van Roey M, et al. Epitope mapping of human anti-adeno-associated virus type 2 neutralizing antibodies: Implications for gene therapy and virus structure. J Virol 2000; 74:1761–6. 26. Li H, Lin SW, Giles-Davis W, et al. A preclinical animal model to assess the effect of pre-existing immunity on AAV-mediated gene transfer. Mol Ther 2009; 17:1215–24. 27. Martino TA, Liu P, Sole MJ. Viral infection and the pathogenesis of dilated cardiomyopathy. Circ Res 1994; 74:182–8. 28. Yun SH, Lee WG, Kim YC, et al. Antiviral activity of coxsackievirus B3 3C protease inhibitor in experimental murine myocarditis. J Infect Dis 2012; 205:491–7.
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
13.
receptor CAR4/7 aggravates cardiac injury. J Mol Med (Berl) 2006; 84: 842–51. Pinkert S, Westermann D, Wang X, et al. Prevention of cardiac dysfunction in acute coxsackievirus B3 cardiomyopathy by inducible expression of a soluble coxsackievirus-adenovirus receptor. Circulation 2009; 120:2358–66. Zhang K, Yu H, Xie W, et al. Expression of coxsackievirus and adenovirus receptor (CAR)-Fc fusion protein in Pichia pastoris and characterization of its anti-coxsackievirus activity. J Biotechnol 2013; 164:461–8. Goodfellow IG, Evans DJ, Blom AM, et al. Inhibition of coxsackie B virus infection by soluble forms of its receptors: Binding affinities, altered particle formation, and competition with cellular receptors. J Virol 2005; 79:12016–24. Haasnoot J, Westerhout EM, Berkhout B. RNA interference against viruses: Strike and counterstrike. Nat Biotechnol 2007; 25:1435–43. Kim YJ, Ahn J, Jeung SY, et al. Recombinant lentivirus-delivered short hairpin RNAs targeted to conserved coxsackievirus sequences protect against viral myocarditis and improve survival rate in an animal model. Virus Genes 2008; 36:141–6. Merl S, Michaelis C, Jaschke B, Vorpahl M, Seidl S, Wessely R. Targeting 2A protease by RNA interference attenuates coxsackieviral cytopathogenicity and promotes survival in highly susceptible mice. Circulation 2005; 111:1583–92. Yuan J, Cheung PK, Zhang HM, Chau D, Yang D. Inhibition of coxsackievirus B3 replication by small interfering RNAs requires perfect sequence match in the central region of the viral positive strand. J Virol 2005; 79:2151–9. Fechner H, Sipo I, Westermann D, et al. Cardiac-targeted RNA interference mediated by an AAV9 vector improves cardiac function in coxsackievirus B3 cardiomyopathy. J Mol Med 2008; 86:987–97. Werk D, Pinkert S, Heim A, et al. Combination of soluble coxsackievirusadenovirus receptor and anti-coxsackievirus siRNAs exerts synergistic antiviral activity against coxsackievirus B3. Antiviral Res 2009; 83:298–306. Becher PM, Lindner D, Miteva K, et al. Role of heart rate reduction in the prevention of experimental heart failure: Comparison between Ifchannel blockade and beta-receptor blockade. Hypertension 2012; 59:949–57.
Combination of RNA Interference and Virus Receptor Trap Exerts Additive Antiviral Activity in Coxsackievirus B3-induced Myocarditis in Mice Elisabeth A. Stein,1 Sandra Pinkert,1 Peter Moritz Becher,2 Anja Geisler,1 Heinz Zeichhardt,3 Robert Klopfleisch,4 Wolfgang Poller,5 Carsten Tschöpe,5 Dirk Lassner,6 Henry Fechner,1,a and Jens Kurreck1,a Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
1
Department of Applied Biochemistry, Technische Universität Berlin, Institute of Biotechnology, 2Department of General and Interventional Cardiology, University Heart Center Hamburg Eppendorf, 3Campus Benjamin Franklin, Institute for Virology, Charité-Universitätsmedizin Berlin, 4Institute of Veterinary Pathology, Freie Universität Berlin, 5Campus Benjamin Franklin, Department of Cardiology and Pneumology, Charité-Universitätsmedizin Berlin, and 6IKDT Institut Kardiale Diagnostik und Therapie GmbH, Berlin, Germany
Background. Coxsackievirus B3 (CVB3) is a major heart pathogen against which no therapy exists to date. The potential of a combination treatment consisting of a proteinaceous virus receptor trap and an RNA interferencebased component to prevent CVB3-induced myocarditis was investigated. Methods and Results. A soluble variant of the extracellular domain of the coxsackievirus-adenovirus receptor (sCAR-Fc) was expressed from an adenoviral vector and 2 short hairpin RNAs (shRdRp2.4) directed against CVB3 were delivered by an adeno-associated virus (AAV) vector. Cell culture experiments revealed additive antiviral activity of the combined application. In a CVB3-induced mouse myocarditis model, both components applied individually significantly reduced inflammation and viral load in the heart. The combination exerted an additive antiviral effect and reduced heart pathology. Hemodynamic measurement revealed that infection with CVB3 resulted in impaired heart function, as illustrated by a drastically reduced cardiac output and impaired contractility and relaxation. Treatment with either sCAR-Fc or shRdRp2.4 significantly improved these parameters. Importantly, the combination of both components led to a further significant improvement of heart function. Conclusions. Combination of sCAR-Fc and shRdRp2.4 exerted additive effects and was significantly more effective than either of the single treatments in inhibiting CVB3-induced myocarditis and preventing cardiac dysfunction. Keywords.
viruses; myocarditis; RNA interference; gene therapy; coxsackievirus.
Coxsackievirus B3 (CVB3) is a plus-stranded RNA virus belonging to the picornavirus family, commonly identified as the infectious agent associated with acute and chronic myocarditis [1]. Strong evidence suggests that dilated cardiomyopathy can develop upon chronic
Received 17 June 2014; accepted 27 August 2014; electronically published 5 September 2014. Presented in part: European Society of Gene and Cell Therapy, Madrid, Spain, 25–28 October 2014. a H. F. and J. K. have contributed equally to this work and thus share last authorship. Correspondence: Jens Kurreck, PhD, Technische Universität Berlin, Institute of Biotechnology, Department of Applied Biochemistry, TIB 4/3-2, Gustav-MeyerAllee 25, 13355 Berlin, Germany ([email protected]) The Journal of Infectious Diseases® 2015;211:613–22 © The Author 2014. Published by Oxford University Press on behalf of the Infectious Diseases Society of America. All rights reserved. For Permissions, please e-mail: [email protected]. DOI: 10.1093/infdis/jiu504
persistence of the infection. This disease may finally progress to terminal heart failure and frequently requires heart transplantation. Although treatment with interferon β was shown to eliminate cardiotropic viruses and improve heart function in patients with myocardial persistence of viral genomes [2], no virus-specific treatment for CVB3 infections exists to date. In recent years, the focus in the development of antiviral drugs has shifted from pharmacologically active small molecule drugs to biological therapeutics [3]. In the case of CVB3, soluble variants of the coxsackievirusadenovirus receptor (sCAR) as a decoy were found to inhibit CVB3 infection in vitro and in vivo [4–8]. Dimeric sCAR fused to the immunoglobulin Fc-region (sCAR-Fc) was shown to neutralize the virus in the absence of undesirable side effects [5, 9]. We recently demonstrated that the in vivo expression of sCAR-Fc RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
613
MATERIAL AND METHODS Coxsackievirus B3
The cardio-virulent, genetically characterized Nancy strain of CVB3 was used for all in vitro and in vivo experiments. Virus propagation and titration on HeLa cells were described previously [14]. Vectors
For the expression of sCAR-Fc, the doxycycline (Dox)-inducible adenoviral vector AdG12 was used [7]. The same vector without Dox was used as control. The expression of shRdRp2 and shRdRp4 (shRNAs directed against the CVB3 RNA dependent RNA polymerase) was accomplished via a self-complementary AAV vector of serotype 2 (AAV2-shRdRp2.4) or 9 (AAV2.9shRdRp2.4) [14]. Vector maps are shown in Figure 1. As a control, the same vector type for the expression of shRNA directed against GFP (shGFP), was used (AAV2-shGFP / AAV2.9shGFP) [14]. Cell Cultures, Plaque and XTT Cell Viability Assays, RNA Isolation and cDNA Synthesis, Quantitative RT-PCR, and IgG ELISA
HeLa cells were cultured in Dulbecco modified Eagle medium (Gibco BRL, Karlsruhe, Germany) supplemented with 5% FCS, 1% MEM nonessential amino acids, 2% HEPES buffer, 0.1% Gentamycin, and 1% penicillin/streptomycin (all from 614
•
JID 2015:211 (15 February)
•
Stein et al
Figure 1. Schematic representation of the vectors used for the expression of sCAR-Fc, shRdRp2, and shRdRp4. A, Structure of the sCAR-Fcexpressing adenoviral vector, AdG12. Two expression cassettes for the constitutive expression of the reverse tetracycline transactivator (rtTA) and the doxycycline-dependent expression of sCAR-Fc, respectively, were inserted into the E1 region between nucleotide positions 453 and 3333 of an E1- and E3-deleted adenovirus 5 backbone. The same vector, without addition of Dox, was used as a control. CMV IE prom indicates the immediate-early CMV promoter; rtTA, reverse tetracycline-controlled transactivator rtTA-M2; tight1, doxycycline-dependent response promoter; sCAR-Fc, fusion protein of the soluble extracellular domain of human CAR and the human IgG Fc region; SV40 and bGH pA, polyadenylation signal of SV40 and bovine growth hormone, respectively; 5′ITR, left inverted terminal repeat of adenovirus type 5; 3′ITR, right inverted terminal repeat of adenovirus 5. B, Structure of the AAV vector used for the U6 promoter (U6 Pro) driven expression of shRdRp2 and shRdRp4 (AAV-shRdRp). Two expression cassettes were inserted in a head to head orientation. The inverted terminal repeats (ITRs) are derived from AAV2. The 3′ITR carries a deletion of the terminal resolution site (trs) and the D-sequence (positions 118–141), resulting in the packaging of self-complementary AAV vector genomes. Capsid proteins of serotype 2 were used for all in vitro experiments, whereas serotype 9 was used for in vivo studies. As a control an AAV vector expressing a shRNA against the nonrelated target GFP (AAVshGFP) was used. Abbreviations: AAV, adeno-associated virus; CAR, coxsackievirus-adenovirus receptor; CMV IE, cytomegalovirus immediateearly; GFP, green fluorescent protein; IgG, immunoglobulin G; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
GIBCO). Plaque assays were carried out as described elsewhere [7]. Cell viability was determined using Roche’s Cell Proliferation Kit II (XTT). RNA was isolated using Life Technologie’s TRIZOL reagent. Thermo Scientific’s RevertAid H Minus First Strand cDNA Synthesis Kit (Waltham, Massachusetts) was used for cDNA synthesis. Expression of inflammatory cytokines was analyzed utilizing the TaqMan Gene expression master mix and specific FAM-tagged TaqMan gene expression assays both from Life Technologies: IL6-Mm00446190_m1, IFNγ-Mm00801778_m1, TNFα-Mm0043258_m1. Detection of CVB3 specific RNA was carried out using Life Technologie’s TaqMan Gene expression master mix and a CVB3 specific primer/probe set. Human immunoglobulin G (IgG) enzymelinked immuosorbent assay (ELISA; Bethyl Laboratories Inc.) was used for detection of sCAR-Fc. In Vivo Mouse Myocarditis Model
Each group consisted of 6 mice. AAV2.9-shRdRp or AAV2.9shGFP vector was injected into the right jugular vein of
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
reduced CVB3 infection, inflammation, and dysfunction of the heart in mice [7]. However, although sCAR-Fc was highly efficient to abrogate CVB3-induced myocarditis when applied prior to infection, the inhibitory effect after onset of the CVB3 infection was limited. Another antiviral approach that gained importance in recent years is the inhibition of viruses by means of RNA interference (RNAi) [10]. We and others have shown that CVB3 can efficiently be inhibited by RNAi in vitro and in vivo [11–14]. However, complete elimination of the virus was not achieved. As both approaches were insufficient when applied individually, we reasoned that the combination of sCAR-Fc as an extracellular trap of CVB3 and RNAi as a mechanism that degrades viral RNA inside cells might exert additive antiviral effects. In a proof-of-concept study we demonstrated that the concurrent application of chemically synthesized siRNAs and sCAR-Fc expressed from an adenoviral vector (AdV) acted synergistically in a cell culture model [15]. In the present study, we translated this strategy to a preclinical in vivo setting of CVB3-induced myocarditis. Both antiviral components were delivered by viral vectors and the efficiency of the double treatment was studied in a CVB3-induced mouse myocarditis model. Our approach demonstrates that the combination treatment results in additive protective effects by inhibiting virus replication, preventing inflammation of heart tissue and improving cardiac function.
6-week-old male BALB/c mice at the dose of 1 × 1012 vector genomes (vg)/ mouse. Eight days later, mice were injected with 1 × 1010 particles of AdG12 into the left jugular vein. Another 2 days later, mice were infected with 5 × 104 plaque-
forming units (PFU) of CVB3 intraperitoneally. Dox (200 µg/ mL) for the expression of sCAR-Fc was given orally by drinking water starting 1 day after CVB3 infection. After 1 week, animals were anesthetized (0.8–1.2 g/kg urethane; 0.05 mg/kg
RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
615
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
Figure 2. Effect of sCAR-Fc and shRdRp on CVB3 replication in vitro. A, Plaque assay. HeLa cells were transduced with AdG12 (MOI 5) and the respective AAV2 vector at 2 × 104 vg/cell. Where indicated, Dox was added to the media at a final concentration of 1 µg/mL. 48 hours later, cell supernatants were preincubated with CVB3 (MOI 0.1) at 4°C for 30 minutes and added to the transduced cells for 30 minutes at 37°C. Subsequently, cells were overlaid with agarcontaining media, cultured at 37°C and 5% CO2 and stained with crystal violet 48 hours later. B, Two-step Plaque Assay. HeLa cells were transduced, cultured, and infected as in A. After infection, virus-containing media was replaced by fresh media. Virus replication was measured by viral plaque assay after another 24 hours of culture. The data presented in this figure as mean ± SD were obtained from three independent experiments, each performed in duplicate. **P < .01. C, XTT cell viability assay. HeLa cells were transduced with the indicated vectors and infected with CVB3 (MOI 0.1) as in A. Following infection, cells were cultured at 37°C and 5% CO2. Cell viability was determined at the indicated time points using the Roche Cell Proliferation Kit II (XTT) according to the manufacturer’s instructions. Data obtained in 2 independent experiments, each performed in quadruplicate, are shown as mean ± SD. Abbreviations: AAV, adeno-associated virus; CVB3, coxsackievirus B3; MOI, multiplicity of infection; PFU, plaque-forming unit; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; SD, standard deviation; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
were transduced and infected with CVB3 as described before. After infection, cells were cultured for another 24 hours and subsequently lysed. The number of infectious viral particles was then determined by plaque assay. Both sCAR-Fc and shRdRp reduced CVB3 replication by 1 to 2 log10 steps. The reduction of the CVB3 titer was significantly higher (approximately 4 log 10 steps) when both components were applied simultaneously (Figure 2B, Supplementary Table 3). The additive protective effect of shRdRp and sCAR-Fc was further underlined by the increase in cell viability as measured by XTT assays (Figure 2C). CVB3-infected cells that received no antiviral treatment were completely lysed 48 hours after infection. Treatment of cells with either of the 2 components resulted in a 50% reduction of cell viability compared to uninfected cells. The combination approach again exerted an additive antiviral effect and cell viability remained virtually at the same level as determined for the uninfected cells for up to 2 days.
Statistical Analysis
To prove the specificity of our approach, dose dependency assays were performed. The multiplicity of infection (MOI) of AdG12 was kept constant at 5 as higher doses completely prevented cell lysis. The dose of AAV2-shRdRp and its control vector, AAV2-shGFP, ranged from 103 to 2 × 104 vg/cell. As expected, the antiviral effect of AdG12 in the presence of Dox was independent of the doses of the added control vector AAV2-shGFP (Figure 3). Unspecific effects of the AAV vector expressing a shRNA can therefore be excluded for the concentration range under investigation. In contrast, the antiviral activity of AAV2-shRdRp increased with higher vector doses. This observation was made for the single RNAi-treatment (ie, sCAR-Fc expression was not induced by Dox), as well as for the combination treatment for which Dox was added to induce sCAR-Fc expression. Importantly, as found above for the singledose experiments, the additive antiviral effect was again clearly detectable for all AAV vector doses under investigation.
Graph Pad Prism 6.0 software was used for statistical analysis. In vivo data were statistically analyzed applying unpaired, 2-tailed Mann–Whitney U testing. For the in vitro data analysis the unpaired student t test was applied. Differences were considered significant at P < .05 (with * referring to .01 < P < .05 and ** referring to P < .01). RESULTS Additive Anti-CVB3 Effect Caused by sCAR-Fc and shRdRp In Vitro
In a first attempt to test for possible additive antiviral effects of sCAR-Fc and shRdRp2.4, the ability of both components to inhibit the generation of new virus progeny was investigated in cell culture by plaque assays (Figure 2A). HeLa cells were transduced with AdG12 (MOI 5) and AAV2-shRdRp or AAV2shGFP (2 × 104 vg/cell) and cultured in the presence or absence of Dox (1 µg/mL). In sum, 48 hours after transduction, cells were infected with CVB3 (MOI 0.1) and overlaid with agar containing media. Another 48 hours later, cells were stained with crystal violet and evaluated for number and size of plaques. Cells treated with noninduced AdG12 or AAV2-shGFP and infected with CVB3 were completely lysed comparable to the mock-treated, CVB3-infected cells. Both sCAR-Fc and shRdRp alone were capable of inhibiting CVB3 replication as approximately 50% of the surface remained covered by viable cells. The lowest number of virus plaques was observed for cells treated with both components simultaneously, illustrating the additive antiviral effect. To quantify the extent of the observed additive effect, 2-step plaque assays were conducted (Figure 2B). To this end, cells 616
•
JID 2015:211 (15 February)
•
Stein et al
The Additive Antiviral Effect Observed is Directly Dependent on the AAV Vector Dose Used In Vitro
Combined Application of sCAR-Fc and shRdRp Significantly Reduces Myocardial Inflammation and Virus Load in the Heart Compared to the Single Component Approach in a CVB3induced Mouse Myocarditis Model
After having observed a strong additive effect of the double treatment in vitro, the combination approach was tested in a mouse myocarditis model in vivo. In our previous study, induction of sCAR-Fc inhibited CVB3 virtually to completion when applied prior to the infection [7]. We therefore chose to induce the expression of sCAR-Fc postinfection to be able to investigate possible additive effects of the double treatment. Antiviral shRNAs were previously found to lead to a partial inhibition of CVB3 when applied prior to infection [14] and were employed in the same manner in the present study.
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
buprenorphine intraperitoneally), intubated, and artificially ventilated with a rodent ventilator type 7025 (Ugo Basile). Hemodynamic parameters were recorded via a microconductance catheter (1.2F) system (Scisense Inc.) in closed chest animals as described elsewhere [16]. Afterward, animals were killed, blood samples were taken, and organs harvested. For histopathological analysis, tissue samples were fixed in 4% neutral buffered formalin for 24 hours and embedded in paraffin wax. Sections (2 µm) were stained with hematoxylin and eosin (HE) according to standard protocols [17]. Evaluation of myocardial integrity and determination of myocarditis score were performed as described elsewhere in a double-blind way [18]. For molecular analysis and plaque assays, organ samples were frozen in liquid nitrogen and stored at −80°C. Plaque assays for determination of CVB3 titer in the heart were carried out as described elsewhere [14]. The investigation was performed in accordance with the principles of laboratory animal care and the German law of animal protection. Approval was obtained by the local ethics review board.
Animals were divided into 5 groups (Figure 4A): Shamoperated and mock-infected animals served as negative control, whereas sham-operated, CVB3-infected mice were used as positive control for the CVB3 myocarditis model. To analyze both, the single components as well as combined treatment, groups received either sCAR-Fc (AdG12 (+Dox) + AAV2.9-shGFP) or the antiviral shRNAs (AdG12 (−Dox) + AAV2.9-shRdRp) or both (AdG12 (+Dox) + AAV2.9-shRdRp). AAV2.9-shRdRp or AAV2.9-shGFP was intravenously injected into mice at the dose of 1 × 1012 vector copies per mouse. Eight days later, 1 × 1010 particles of AdG12 were applied via the same delivery route. Another 2 days later, mice were infected intraperitoneally with 5 × 104 PFU of CVB3. Mice were killed 8 days after the infection. In the animal groups, in which expression of sCAR-Fc was induced by oral administration of Dox, the level of sCAR-Fc was analyzed and found to be comparable for both groups (Figure 5A). The expression of comparable levels of active antiCVB3 shRNAs was confirmed by quantitative stemloop reverse transcription polymerase chain reaction (RT-PCR; Figure 5B). To investigate the inflammation level, the cardiac tissue of all animals was fixed and stained with HE 7 days postinfection, and the level of inflammation was graded (Figure 4B left; Supplementary Table 2C). For the negative control group, staining revealed no signs of inflammation or necrosis in any of the 6 animals. In contrast, all CVB3-infected, untreated mice were
Heart Function is Significantly Improved in the sCAR-Fc and shRdRp Treated Group Compared to Either of the Single Component Groups
Cardiac contractility is the most important clinical parameter that characterizes the function of the heart. Hemodynamic measurements for the determination of systolic and diastolic
RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
617
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
Figure 3. Dose-response relationship of the additive anti-CVB3 effect. HeLa cells were transduced with AdG12 (MOI 5) and AAV2-shRdRp or AAV2-shGFP at the indicated vg/cell and cultured in the presence or absence of Dox (1 µg/mL) as indicated. After 48 hours of culture, cell supernatants were preincubated with CVB3 (MOI 0.1) at 4°C for 30 minutes. Afterward, transduced cells were infected with these supernatants for 30 minutes. After infection, virus-containing media was replaced by fresh media. Virus replication was measured by viral plaque assay after another 24 hours of culture. Data shown were obtained from 2 independent experiments, each performed in duplicate and are presented as mean ± SD. Abbreviations: AAV, adeno-associated virus; CVB3, coxsackievirus B3; MOI, multiplicity of infection; PFU, plaque-forming unit; SD, standard deviation; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
classified with the highest myocarditis score of 4, which is characterized by the presence of mononuclear cell infiltration and myocyte necrosis in >20% of the heart tissue [18]. Single treatment with either sCAR-Fc or shRdRp2.4 reduced the myocarditis score to 3 or 2, in 3 out of 6 mice in each group. Most importantly, animals treated with both antiviral components showed substantial less inflammation in the heart compared to either of the single element approaches. In contrast to any other group, cardiac integrity was completely maintained in 4 out of 6 animals. Exemplary HE-stained heart sections are shown for each group in the right part of Figure 4B. Proinflammatory cytokines trigger the development of inflammation, which is a major event in the development of virus-induced myocarditis [19]. We investigated the levels of inflammatory cytokines in the heart tissue by RT quantitative PCR (qPCR; Figure 4C, Supplementary Table 2B). Following CVB3-infection, the proinflammatory cytokines interleukin 6 (IL-6), tumor necrosis factor α (TNF-α), and interferon γ (IFN-γ) were significantly up-regulated 39-fold, 6-fold, and 134-fold, respectively. Treatment of the animals with either sCAR-Fc or shRdRp led to reduced levels of all 3 cytokines. The decrease of the cytokine levels was even more pronounced in the double-treatment mice, supporting the data obtained from myocardial inflammation scoring. These data were further underlined by the viral load in cardiac tissue determined by plaque assay. For the sham-operated, CVB3-infected group, an average of 4.7 × 104 PFU/mg heart tissue was determined (Figure 4D left, Supplementary Table 2C). This titer was reduced by approximately 70% for the sCAR-Fc (1.35 × 104 PFU/mg heart tissue) and shRdRp (1.47 × 104 PFU/ mg heart tissue) treated mice, respectively. The reduction in viral load was significantly higher (approximately 95%) when both antiviral components were used simultaneously (2.54 × 10 3 PFU/mg heart tissue). CVB3-specific RT qPCR also confirmed these results (Figure 4D right). The highest level of genomic RNA was detected in cardiac tissue of shamoperated, CVB3-infected animals. The level was approximately 50% lower for the sCAR-Fc or shRdRp treated groups. For the combination group, the level of CVB3 genomes was further decreased to approximately 25% compared to the untreated group. Taken together, differences in viral load correlated well with pathohistological and inflammatory differences among the groups. Hence, the additive protective effect of sCAR-Fc and shRdRp compared to single treatment was further underlined by these results.
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
Figure 4. Influence of sCAR-Fc and shRdRp on heart inflammation and virus load in murine CVB3-induced myocarditis. A, Timeline, application scheme and group distribution of the in vivo approach. Ten days prior to infection, 1 × 1012 vector genomes of the respective AAV2.9 vector were injected intravenously via the jugular vein. Eight days later, 1 × 1010 particles of AdG12 were injected via the same delivery route. Two days thereafter, mice were infected intraperitoneally with 5 × 104 PFU of CVB3 and sCAR-Fc expression was induced as indicated 1 day later by addition of Dox (200 µg/mL) to the drinking
618
•
JID 2015:211 (15 February)
•
Stein et al
DISCUSSION
Figure 5. Serum level of sCAR Fc and expression of shRdRp2 and shRdRp4 in the heart 7 days postinfection. A, 7 days after infection, blood samples of all mice were taken, and serum was prepared. Serum level of sCAR Fc was measured using a human IgG specific sandwich ELISA kit. Signals obtained from sham-operated and CVB3-infected animals were subtracted as background. Values shown are mean ± SEM. B, Analysis of shRdRp2 and shRdRp4 expression in heart tissue of mice transduced with the AAV2.9 shRdRp vector in the myocarditis model. The processed siRNAs were detected via quantitative stemloop RT PCR 7 days after infection. Expression data were normalized to the expression of 18 S rRNA and the siRNA level in the group AdG12 (Dox) + shRdRp. Levels of the siRNAs were similar between this group and the group that received the combined treatment. Only background signals were detected for the group AdG12 (+Dox) + shGFP, proving PCR specificity for the detection of siRdRp2 and siRdRp4. Abbreviations: AAV, adeno-associated virus; CVB3, coxsackievirus B3; ELISA, enzyme-linked immunosorbent assay; IgG, immunoglobulin G; PCR, polymerase chain reaction; rRNA, ribosomal RNA; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; SEM, standard error of the mean; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
CVB3 is a major causative agent for acute and chronic myocarditis against which no specific antiviral treatment exists to date. In the present study, we report that combined treatment of CVB3-induced myocarditis with vector-delivered, virusneutralizing sCAR-Fc and shRNAs directed against the CVB3 genome are not antagonistic toward one another and exert additive antiviral effects compared to individual usage. Treatment with both components resulted in significantly reduced viral load in the heart. Consequently, the portion of heart tissue damaged (inflammation and necrosis) was strongly reduced in double-treatment animals compared to the single-treatment ones. Importantly, treatment with both antiviral components significantly improved hemodynamic parameters in CVB3-infected animals compared to mice that only received 1 active component. We have previously shown that sCAR-Fc efficiently inhibits CVB3 infection when given prior to the infection [7]. In the present study, generation of sCAR-Fc was induced after infection of the mice, that is, in a therapeutic manner. However,
Figure 4 continued. water. A positive control group (Sham + CVB3) was sham-operated and CVB3-infected. The negative control animals (Mock) were sham-operated and mock-infected with PBS. Analysis was carried out 7 days postinfection. (n = 6 for each group). B, Left: Myocarditis score of CVB3infected and sCAR-Fc- and/ or shRdRp-treated mice. Complete cardiac integrity is graded with a myocarditis score of 0, whereas the highest myocarditis score of 4 indicates massive inflammation in more than 20% of cardiac tissue. Right: Heart sections of mice were stained with hematoxylin and eosin. Arrows indicate areas of inflammation. Pictures shown are representative for each group. C, Expression levels of inflammatory cytokines in cardiac tissue of the different groups. The mRNA expression levels were analyzed 7 days after infection. Data were normalized to the expression level of 18 S rRNA and are presented as 2−ΔCt. Values shown are mean ± SEM for each group. *P < .05; **P < .01. (n = 6 for each group). D, Viral load in the heart. Left: Viral plaque titer of heart lysates as determined by plaque assay. The viral load in the heart is shown in PFU per mg heart tissue. Plaque titer of each heart sample was determined in 2 independent experiments each performed in duplicate. Right: Quantification of genomic CVB3 RNA in the heart tissue by quantitative RTPCR. Data were normalized to the expression of 18 S rRNA and are shown as 2−ΔCt. Values shown are mean ± SEM for each group. *P < .05; **P < .01. (n = 6 for each group). Abbreviations: AAV, adeno-associated virus; CVB3, coxsackievirus B3; IFNγ, interferon γ; IL-6, interleukin 6; MOI, multiplicity of infection; mRNA, messenger RNA; PBS, phosphate-buffered saline; PFU, plaque-forming unit; rRNA, ribosomal RNA; RT-PCR, reverse transcription polymerase chain reaction; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; SD, standard deviation; SEM, standard error of the mean; shRdRp, short hairpin RNA against RNA dependent RNA polymerase; TNFα, tumor necrosis factor α. RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
619
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
left ventricular (LV) function were therefore conducted by tip catheter at day 7 postinfection (Figure 6 and Supplementary Tables 1 and 2A). Because the heart rate was comparable between all groups, the data obtained from hemodynamic measurements can be compared with respect to the heart-protective effects of our antiviral treatment with sCAR-Fc and shRdRp. Infection with CVB3 resulted in impaired systolic and diastolic LV function, indicating a significant global cardiac dysfunction, as illustrated by a reduced cardiac output and impaired contractility and relaxation. A significant improvement of these parameters was achieved with either of the single treatments, sCAR-Fc or shRdRp. Most importantly, the combined application of both components again led to a further significant improvement of cardiac output and contractility parameters. These data illustrate the potential of the double-treatment with sCAR-Fc and shRdRp to excite additive antiviral effects in vivo even with regard to the most important parameters of heart function.
the application of sCAR-Fc involves some limitations. Viral particles only get trapped by sCAR-Fc extracellularly before they enter their target cells. This limitation is especially disadvantageous when treatment is initiated during an ongoing infection. Vector-delivered shRdRp led to a partial improvement of hemodynamic parameters during CVB3-caused myocarditis [14]. The underlying mechanism of RNAi is limited to the intracellular level, whereas the initial virus uptake is not inhibited. In addition, the therapeutic AAV vector will not transduce all cardiomyocytes, and CVB3 can replicate in cells that are not transduced. Due to the fast replication cycle of CVB3, the high number of virus progeny, and the inability to transduce all cardiomyocytes with AAV2.9-shRdRp, complete cardiac integrity was not preserved by the RNAi approach. 620
•
JID 2015:211 (15 February)
•
Stein et al
In the present study, we therefore chose to combine the extracellular virus trap strategy based on sCAR-Fc with the intracellular RNAi concept and wanted to investigate possible additive effects of the double-treatment. Combination therapies targeting viruses at different stages of their lifecycle are a common strategy to improve the efficiency of the antiviral treatment [20, 21]. Our approach transferred a previous cell culture study employing synthesized anti-CVB3 siRNAs [15] to the preclinical in vivo stage in the current study using a cardiotropic AAV vector for expression of shRdRp in addition to the adenovirus vector AdG12 for the expression of sCAR-Fc. In line with previous results, the shRNAs alone reduced the virus titer in the myocardium by approximately 70%, resulting in partial improvement of heart parameters [14]. Similarly, mice treated
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
Figure 6. Effect of sCAR-Fc and shRdRp treatment on CVB3-induced cardiac dysfunction. Seven days after CVB3 infection, mice were anesthetized (0.8–1.2 g/kg urethane; 0.05 mg/kg buprenorphine intraperitoneally), intubated, and artificially ventilated with a rodent ventilator type 7025. Heart parameters, such as heart rate, LV contractility (dP/dtmax), LV relaxation (dP/dtmin), and cardiac output (CO) as indexes of LV function were recorded via a microconductance catheter (1.2F) system in closed chest animals. Values shown are mean ± SEM for each group. *P < .05; **P < .01. (n = 6 for each group). For a complete list of the recorded cardiac parameters see Supplementary Table 1. Abbreviations: CVB3, coxsackievirus B3; LV, left ventricular; sCAR-Fc, soluble variant of coxsackievirus-adenovirus receptor; SEM, standard error of the mean; shRdRp, short hairpin RNA against RNA dependent RNA polymerase.
of either of the individual components. In a mouse CVB3 myocarditis model, we demonstrated that the treatment reduced virus load, tissue damage, and induction of inflammatory cytokines. As a consequence, cardiac function, which is known to be severely disturbed during acute CVB3 infection [27], was significantly improved as shown by hemodynamic measurements. No obvious adverse effects of the antiviral components were observed. Targeting the virus at 2 critical points of its replication cycle by application of both sCAR-Fc and shRdRp has thus the potential to develop into a novel treatment option for CVB3induced myocarditis. Furthermore, to the best of our knowledge, it represents the first described combination approach of an RNAi strategy and a proteinaceous antiviral component and may thus represent a general paradigm to treat severe viral infections. For example, the antiviral agents used here may also be combined with other components such as the inhibitor of CVB3 protease, which was developed more recently [28]. Supplementary Data Supplementary materials are available at The Journal of Infectious Diseases online (http://jid.oxfordjournals.org). Supplementary materials consist of data provided by the author that are published to benefit the reader. The posted materials are not copyedited. The contents of all supplementary data are the sole responsibility of the authors. Questions or messages regarding errors should be addressed to the author.
Notes Acknowledgments. The authors thank Xiaomin Wang for vector production of AdG12 and Erik J. Wade for critical reading of the manuscript and helpful comments. Financial support. This work was supported by the Deutsche Forschungsgemeinschaft through grants [Ku 1436/6-1] to J. K. and [Fe 785/ 3-1, Fe785/2-1, FE 785/2-2] to H. F. and the Sonderforschungsbereich (SFB) TR19 to C. T., H. F., and W. P. H. Z. is grateful for the support from Otto-Kuhn-Stiftung im Stifterverband für die Deutsche Wissenschaft. Potential conflicts of interest. All authors: No reported conflicts All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
References 1. Kandolf R, Klingel K, Zell R, et al. Molecular mechanisms in the pathogenesis of enteroviral heart disease: Acute and persistent infections. Clin Immunol Immunopathol 1993; 68:153–8. 2. Kuhl U, Pauschinger M, Schwimmbeck PL, et al. Interferon-beta treatment eliminates cardiotropic viruses and improves left ventricular function in patients with myocardial persistence of viral genomes and left ventricular dysfunction. Circulation 2003; 107:2793–8. 3. Fox JL. Antivirals become a broader enterprise. Nat Biotechnol 2007; 25:1395–402. 4. Lim BK, Choi JH, Nam JH, et al. Virus receptor trap neutralizes coxsackievirus in experimental murine viral myocarditis. Cardiovasc Res 2006; 71:517–26. 5. Yanagawa B, Spiller OB, Proctor DG, et al. Soluble recombinant coxsackievirus and adenovirus receptor abrogates coxsackievirus b3-mediated pancreatitis and myocarditis in mice. J Infect Dis 2004; 189:1431–9. 6. Dorner A, Grunert HP, Lindig V, et al. Treatment of coxsackievirusB3-infected BALB/c mice with the soluble coxsackie adenovirus
RNAi-sCAR Treatment of CVB3 Myocarditis
•
JID 2015:211 (15 February)
•
621
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
with sCAR-Fc alone showed a reduction in viral heart titer of approximately 70% and partial restoration of heart function as in previous studies [7]. When combining both approaches, we did not observe any antagonistic effects. Importantly, we now demonstrate the extraordinary potential of both active components to exert additive antiviral effects against CVB3. The viral heart titer of double-treated mice was reduced by approximately 95% compared to untreated mice. Likewise, the reduction of inflammatory cytokines was strongest for the double-treatment group: 4 out of 6 mice of this group had no signs of inflammation or tissue damage in the heart, whereas this was not observed for any animal from a single-treatment group. Most importantly, the protective effects in terms of heart parameters, such as cardiac output and LV contractility and relaxation, was significantly stronger for the double-treated animals compared to mice that only received 1 component, proving the extraordinary potential of our combined approach. We chose to deliver both sCAR-Fc and shRdRp with viral vectors. Although sCAR-Fc was expressed from the adenoviral vector AdG12, an AAV vector of serotype 9 (AAV2.9-shRdRp) was used for in vivo shRNA delivery. Gene therapy approaches are strongly dependent on the safety of the viral vectors used. During the time course of our in vivo study, no obvious adverse effects were observed in the animals treated with either of the viral vectors. As the presence of AdG12 in the body is limited to several weeks, no long-term adverse effects are expected for its use [22]. Furthermore, the doxycycline-inducible system for the expression of sCAR-Fc, which enables a very tight regulation of expression, for example, in case of undesirable adverse effects, provides an additional safety feature of our application. For AAV vectors, long-term expression of the transgene for at least 1 year in the absence of adverse effects was reported [23]. Even though some groups reported cytotoxicity when expressing high amounts of shRNAs from AAV vectors [24], no such phenomena were observed in our study. Therefore, the vectors chosen for the delivery of the antiviral components were safe for our approach. However, the limited half-life of AdG12 delivered sCAR-Fc may become a limitation for its clinical use. To transfer our preclinical study to the clinical stage, sCAR-Fc might be applied as a recombinantly produced protein or from a different vector type (eg, AAV vectors) that enables stable long-term expression without evoking adverse effects. Previous studies have shown that treatment of mice with an AAV vector induces neutralizing antibodies, which make a second treatment with the same serotype inefficient [25]. This problem can be solved by using a different serotype to circumvent the immune response [26]. In conclusion, our proof-of-concept study demonstrates that the combination of sCAR-Fc as an extracellular virus trap and RNAi as an intracellular mechanism to degrade CVB3 genomes has an extraordinarily high antiviral effectiveness. The efficiency of the combination approach is significantly better than that
7.
8.
9.
10. 11.
12.
14.
15.
16.
622
•
JID 2015:211 (15 February)
•
Stein et al
17. Klopfleisch R, Gruber AD. Derlin-1 and stanniocalcin-1 are differentially regulated in metastasizing canine mammary adenocarcinomas. J Comp Pathol 2009; 141:113–20. 18. Szalay G, Sauter M, Hald J, Weinzierl A, Kandolf R, Klingel K. Sustained nitric oxide synthesis contributes to immunopathology in ongoing myocarditis attributable to interleukin-10 disorders. Am J Pathol 2006; 169:2085–93. 19. Leslie K, Blay R, Haisch C, Lodge A, Weller A, Huber S. Clinical and experimental aspects of viral myocarditis. Clin Microbiol Rev 1989; 2:191–203. 20. Li M, Li H, Rossi JJ. RNAi in combination with a ribozyme and TAR decoy for treatment of HIV infection in hematopoietic cell gene therapy. Ann N Y Acad Sci 2006; 1082:172–9. 21. Boutimah F, Eekels JJ, Liu YP, Berkhout B. Antiviral strategies combining antiretroviral drugs with RNAi-mediated attack on HIV-1 and cellular co-factors. Antiviral Res 2013; 98:121–9. 22. Wei K, Kuhnert F, Kuo CJ. Recombinant adenovirus as a methodology for exploration of physiologic functions of growth factor pathways. J Mol Med (Berl) 2008; 86:161–9. 23. Bish LT, Morine K, Sleeper MM, et al. Adeno-associated virus (AAV) serotype 9 provides global cardiac gene transfer superior to AAV1, AAV6, AAV7, and AAV8 in the mouse and rat. Hum Gene Ther 2008; 19:1359–68. 24. Grimm D, Streetz KL, Jopling CL, et al. Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways. Nature 2006; 441:537–41. 25. Moskalenko M, Chen L, van Roey M, et al. Epitope mapping of human anti-adeno-associated virus type 2 neutralizing antibodies: Implications for gene therapy and virus structure. J Virol 2000; 74:1761–6. 26. Li H, Lin SW, Giles-Davis W, et al. A preclinical animal model to assess the effect of pre-existing immunity on AAV-mediated gene transfer. Mol Ther 2009; 17:1215–24. 27. Martino TA, Liu P, Sole MJ. Viral infection and the pathogenesis of dilated cardiomyopathy. Circ Res 1994; 74:182–8. 28. Yun SH, Lee WG, Kim YC, et al. Antiviral activity of coxsackievirus B3 3C protease inhibitor in experimental murine myocarditis. J Infect Dis 2012; 205:491–7.
Downloaded from http://jid.oxfordjournals.org/ at University of Birmingham on April 21, 2015
13.
receptor CAR4/7 aggravates cardiac injury. J Mol Med (Berl) 2006; 84: 842–51. Pinkert S, Westermann D, Wang X, et al. Prevention of cardiac dysfunction in acute coxsackievirus B3 cardiomyopathy by inducible expression of a soluble coxsackievirus-adenovirus receptor. Circulation 2009; 120:2358–66. Zhang K, Yu H, Xie W, et al. Expression of coxsackievirus and adenovirus receptor (CAR)-Fc fusion protein in Pichia pastoris and characterization of its anti-coxsackievirus activity. J Biotechnol 2013; 164:461–8. Goodfellow IG, Evans DJ, Blom AM, et al. Inhibition of coxsackie B virus infection by soluble forms of its receptors: Binding affinities, altered particle formation, and competition with cellular receptors. J Virol 2005; 79:12016–24. Haasnoot J, Westerhout EM, Berkhout B. RNA interference against viruses: Strike and counterstrike. Nat Biotechnol 2007; 25:1435–43. Kim YJ, Ahn J, Jeung SY, et al. Recombinant lentivirus-delivered short hairpin RNAs targeted to conserved coxsackievirus sequences protect against viral myocarditis and improve survival rate in an animal model. Virus Genes 2008; 36:141–6. Merl S, Michaelis C, Jaschke B, Vorpahl M, Seidl S, Wessely R. Targeting 2A protease by RNA interference attenuates coxsackieviral cytopathogenicity and promotes survival in highly susceptible mice. Circulation 2005; 111:1583–92. Yuan J, Cheung PK, Zhang HM, Chau D, Yang D. Inhibition of coxsackievirus B3 replication by small interfering RNAs requires perfect sequence match in the central region of the viral positive strand. J Virol 2005; 79:2151–9. Fechner H, Sipo I, Westermann D, et al. Cardiac-targeted RNA interference mediated by an AAV9 vector improves cardiac function in coxsackievirus B3 cardiomyopathy. J Mol Med 2008; 86:987–97. Werk D, Pinkert S, Heim A, et al. Combination of soluble coxsackievirusadenovirus receptor and anti-coxsackievirus siRNAs exerts synergistic antiviral activity against coxsackievirus B3. Antiviral Res 2009; 83:298–306. Becher PM, Lindner D, Miteva K, et al. Role of heart rate reduction in the prevention of experimental heart failure: Comparison between Ifchannel blockade and beta-receptor blockade. Hypertension 2012; 59:949–57.
