Genes and Immunity (2015) 16, 261–267 © 2015 Macmillan Publishers Limited All rights reserved 1466-4879/15 www.nature.com/gene
ORIGINAL ARTICLE
Mapping of a quantitative trait locus controlling susceptibility to Coxsackievirus B3-induced viral hepatitis SA Wiltshire, J Marton, GA Leiva-Torres and SM Vidal The pathogenesis of coxsackieviral infection is a multifactorial process involving host genetics, viral genetics and the environment in which they interact. We have used a mouse model of Coxsackievirus B3 infection to characterize the contribution of host genetics to infection survival and to viral hepatitis. Twenty-five AcB/BcA recombinant congenic mouse strains were screened. One, BcA86, was found to be particularly susceptible to early mortality; 100% of BcA86 mice died by day 6 compared with 0% of B6 mice (P = 0.0012). This increased mortality was accompanied by an increased hepatic necrosis as measured by serum alanine aminotransferase (ALT) levels (19547 ± 10556 vs 769 ± 109, P = 0.0055). This occurred despite a predominantly resistant (C57BL/6) genetic background. Linkage analysis in a cohort (n = 210) of (BcA86x C56Bl/10)F2 animals revealed a new locus on chromosome 13 (peak linkage 101.2 Mbp, lod 4.50 and P = 0.003). This locus controlled serum ALT levels as early as 48 h following the infection, and led to an elevated expression of type I interferon. Another locus on chromosome 17 (peak linkage 57.2 Mbp) was significantly linked to heart viral titer (lod 3.4 and P = 0.046). These results provide new evidence for the presence of genetic loci contributing to the susceptibility of mice to viral hepatitis. Genes and Immunity (2015) 16, 261–267; doi:10.1038/gene.2015.5; published online 19 March 2015
INTRODUCTION Coxsackievirus B3 (CVB3) is frequently studied in the context of myocarditis and its long-term sequala dilated cardiomyopathy. However, coxsackievirus infection is systemic: early accounts detail that CVB3 is capable of inducing myositis, meningitis, pancreatitis, hepatitis as well as myocarditis.1 A recent study based on seropositivity in diabetics estimates that up to 75% of people are seropositive to any given strain of coxsackievirus, 66% for CVB3 alone.2 In contrast, severe symptoms are comparatively rare. In newborns, severe cases of viral hepatitis are periodically reported as a result of group B Coxsackieviruses, including CVB3.3–5 Clinically observed heterogeneity is recapitulated in the mouse model, where disease progression and presentation depend on both host and viral factors.6–9 The central role of the host immune response in determining the outcome of CVB3 infection has been clearly demonstrated in a number of reverse genetic (knockout) studies. The type I Interferon (IFN) axis seems to be central in controlling viral replication within the liver, but may be redundant in controlling viral replication in the heart. IFN receptor (Ifnar1) knockout mice succumbed early to a lethal infection, with greatly increased viral replication in the liver, and 10-fold higher liver damage as measured by serum alanine aminotransferase (ALT) compared with the wild type. However no significant changes in viral replication or inflammation were observed in Ifnar1− / −hearts,10 observations that are mirrored in IFNβ knockout mice.11 Type I IFN signalling therefore seems to have a pivotal role in the liver with no apparent effect in the heart. Conversely, abrogation of all JAK/STAT signalling by the transgenic overexpression of the suppressor of cytokine signalling 1 (Socs1), greatly enhanced the myocarditis with no effect in the liver.12 Deficiency for toll-like receptor 3 (Tlr3), a viral pattern recognition receptor sensitive to lumenal double stranded RNA, also increases
susceptibility to myocarditis, but not hepatitis.13 Absence of the cytoplasmic dsRNA sensor melanoma differentiation-associated protein 5 (Ifih1 or MDA5) leads to a greatly increased mortality, accompanied by a severe hepatic necrosis14 and greatly elevated ALT levels. Absence of the MDA-5 transducing mitochondrial antiviral signalling protein (Mavs) also leads to an early mortality (days 2–3) accompanied by a severe hepatic necrosis. Further complicating matters, the genetic background has both characterized and unknown effects that contribute to pathogenesis and tissue tropism. The signal lymphocytic activation molecule 2 haplotype controls heart vs liver tropism in the C57BL/6 (B6) background.15 Moreover, while in the B6 genetic background MDA5 is essential for the survival of CVB3 infection and protection from an hepatic necrosis, MDA5 deficiency in the 129SvJ background has no apparent effect.16 Forward genetic analysis by using mouse models constitutes a powerful tool to study the natural variation and background variation in complex traits and disease. It is known that the A/J and B6 strains are respectively susceptible and resistant to CVB3 infection.17 We have previously made use of a forward genetic approach to scan the mouse genome for loci linked to susceptibility to CVB3-induced myocarditis between the A/J and B6 strains.18,19 To identify additional loci controlling susceptibility to CVB3 infection, the recombinant congenic (RC) AcB/BcA panel was screened. RC strains are generated by two sequential backcrosses of one inbred donor strain (12.5% of the genome) to a background recipient strain (87.5% of the genome) and have been developed as a tool for complex trait mapping. Following a screen of the AcB/BcA panel, one strain (BcA86) with a dramatically discordant phenotype was singled out for further analysis. Viral replication and common serum biochemical markers were tested to characterize pathogenesis in BcA86 mice. Serum
Department of Human Genetics, McGill University, Montreal, QC, Canada. Correspondence: Dr SM Vidal, Department of Human Genetics, McGill University, McGill Life Sciences, Complex, Room 367, 3649 Promenade Sir William Osler, Montreal H3G 0B1, QC, Canada. E-mail: [email protected] Received 7 November 2014; revised 23 December 2014; accepted 5 January 2015; published online 19 March 2015
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
262 ALT in the BcA86 mice was found to be elevated compared with other strains, and was pursued by using an informative F2-cross approach. Linkage analysis revealed a single significant locus on the distal chromosome 13 that controls ALT levels. RESULTS Mice of the BcA86 RC strain succumb to lethal CVB3 infection The course of infection with CVB3 is highly variable and depends on both host and viral background.8,9 It is known that the A/J and B6 strains are respectively susceptible and resistant to CVB3 infection,17 though little is known about the genetic determinants underlying this trait. It was therefore of interest to challenge the RC strain panel of AcB/BcA mice with CVB3 infection. The screen was undertaken by using an infectious dose which had previously been used to study myocarditic potential of CVB3 in A/J and B10. A-H2a mice.19 It was observed early into screening the RC strain panel that the majority of AcB (A/J background, B6 donor) and BcA (B6 background, A/J donor) mice did not survive the infection to the previously established experimental end point on day 8. Therefore survival of infection was used as a phenotype. In total, 109 male RC mice were screened, with between 2 and 8 per strain (median 4) (Figure 1a). Consistent with previous reports of the involvement of H2 in susceptibility to CVB3 infection,8,17,20 the two H2a-carrying BcA strains (BcA69 and BcA74) were relatively resistant, and survived the infection. Most other strains were comparatively susceptible to lethal infection. In particular, the BcA86 strain succumbed to early mortality between days 2 and 5 post infection (n = 6, P = 0.0012 compared with B6) (Figure 1b). Further phenotyping was conducted in a panel of A/J, BcA86, B6 and the closely related C57BL/10 (B10). B10 mice, which are resistant to CVB3, were included knowing that the genetic architecture underlying the phenotype of BcA86 might include not only A/J loci but also de novo mutations on the B6 background. Two de novo mutations have been mapped so far within the AcB/ BcA panel on the background strain.21,22 Reports of early mortality within CVB3 infection have implicated metabolic failure secondary to hepatic necrosis,14,23,24 so viral titer in the heart and liver, as well as serum biomarkers for heart and liver damage were assessed at the early time point of 48 h post infection (Figure 2). It was found that viral titer in BcA86 was slightly higher in the heart than in B6 (one-way analysis of variance with Tukeys multiple comparison test, P = 0.03), while being comparable in the spleen and liver. No significant differences in the heart damage biomarker creatine kinase were observed at this time point. BcA86 mice showed a marked increase in serum levels of the liver damage biomarker ALT as compared with A/J, B6 and B10 mice (one-way analysis of variance with Tukeys multiple comparison test P = 0.0005, 0.0055 and 0.0007). The BcA86 phenotype of elevated
ALT relative to B10 mice was reproduced in independent experiments (data not shown). Segregation analysis of serum alanine aminotransferase levels, and viral titer reveals complex inheritance In order to identify the genetic determinants conferring susceptibility to BcA86 mice, we undertook a [B10x BcA86]F2 cross (N = 210) [Figure 3]. Segregation analysis indicated that serum ALT levels, liver and heart viral titer were all highly skewed distributions (+2.45, +2.90, +6.61, respectively, Po1e-9) [Figures 3a,c and e] that deviated very significantly from the normal distribution (P = 3.35 × 10 − 16, 3.706 × 10 − 19 and 4.995 × 10 − 18). Given that quantitative trait loci mapping assumes normal phenotypic distributions, these phenotypes were normalized by using the Box–Cox method (Figures 3b,d and f).25 The distributions of the transformed data resembled a normal distribution more closely than those of the raw data (P = 0.03424, 0.0002131 and 0.04858). Viral titer in the liver was significantly correlated with serum ALT (R2 = 0.44), and also with heart viral titer but to a lesser degree (R2 = 0.30). Heart and liver viral titer were also correlated to each other (R2 = 0.36). These results are consistent with common and distinct genetic determinants of susceptibility within the [BcA86x B10]F2 mice. Linkage analysis reveals two loci controlling susceptibility to CVB3 infection in [BcA86x B10]F2mice In order to perform linkage analysis, a panel of genetic markers was created to distinguish between BcA86 and B10 backgrounds (described further in methods). The full B6/B10 panel has been successfully used and described elsewhere.26,27 Linkage to serum ALT identified a single significant peak on the distal chromosome 13 (lod = 4.50, P = 0.003, 9.4% variance explained, peak marker rs3702296) (Figure 4 and Figure 5a). No significant linkage was found to viral titer in the liver, however the maximum lod score (2.81) and lowest P-value (0.14) were also obtained at marker rs3702296. Linkage to viral titer in the heart revealed one peak on chromosome 17 (lod = 3.4, P = 0.046) (Figure 4, Figure 5d). The 1-lod support interval on chromosome 13 extends between 86 Mb and the end of the chromosome 120 Mb, with a peak of linkage at 102 Mb. On chromosome 17, the 1-lod support interval extends between 49.7 and 69.7 Mb. An A/J haplotype block beginning at 47.7 Mb and extending to 58.6 Mb, is present within the chromosome 17 interval. Allele effect plots indicated that in both cases, the BcA86 genotype conferred susceptibility at the two loci (Figures 5b,c,e and f). Mice inheriting homozygous BcA86 alleles on chromosome 13 had a median serum ALT level of 8820IU/L and mice homozygous for B10 had a median serum ALT level of 1986IU/L. These same mice homozygous for chromosome 13 had average liver viral titers of 4.68 log[pfu/mg] and 4.01 log[pfu/mg],
Figure 1. Survival of CVB3 infection in AcB/BcA RC strains of mice. (a) Strain distribution plot of mean survival time for RC strains of mice. Between 2 and 8 male mice per strain (median 4) were tested for a total of 109 mice. Both H2a carrying BcA strains (BcA69 and BcA74) survived the infection. Bars represent means ± s.d. (b) The BcA86 strain succumbed to early mortality between days 2 and 5 post infection (n = 6, P = 0.0012 compared with B6). Genes and Immunity (2015) 261 – 267
© 2015 Macmillan Publishers Limited
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
263
Figure 2. Quantification of viral titers and serum metabolites in BcA86 mice compared with genetic background and donor strains. Viral replication in the heart (a), liver (b) and spleen (c) were measured in A/J, BcA86, B6, B10 mice (n ⩾ 4) 2 days following the infection. Only viral replication in the heart was statistically significantly different between the four strains of mice (one-way analysis of variance with Tukeys multiple comparison test, P o0.05). (d, e) Levels of serum creatine kinase and ALT in A/J, BcA86, B6 and B10 mice (n ⩾ 5) 2 days following the infection. Serum ALT levels were significantly greater in BcA86 mice (one-way analysis of variance with Tukeys multiple comparison test P = 0.005). Bars represent mean ± s.d.
Figure 3. Distributions of viral replication and ALT within F2 mice. Histograms depicting the distribution of serum ALT (a), liver viral titer (c) and heart viral titer (e) in [BcA86xC67BL/10]F2 (n = 210) mice. Strong skewness was detected in the three phenotypic distributions, and so phenotypes were normalized by Box–Cox (b, d, f) for further analysis. © 2015 Macmillan Publishers Limited
Genes and Immunity (2015) 261 – 267
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
264
Figure 4. Genome-wide linkage analysis in [BcA86xB10]F2 mice reveals two significant loci. Genome-wide linkage scans for day 2 serum ALT (solid line), liver viral titer (dotted line), and heart viral titer (dashed line). Linkage analysis was performed on normalized phenotypes, and genome-wide significance was determined by 10 000 permutations (with alpha = 0.05 shown as a dashed horizontal line). A single locus on distal chromosome 13 was significantly linked to ALT levels (lod = 4.50, P = 0.003). An additional locus on chromosome 17 was linked to viral replication in the heart (lod 3.4, P = 0.046).
Figure 5. Linkage of chromosomes 13 and 17 to CVB3 viral replication and serum ALT levels. (a) Lod score plot of serum ALT (solid line), viral replication in the liver (dotted line) and in the heart (dashed line) calculated by using interval mapping. Peak linkage to serum ALT on distal chromosome 13 occurs at rs3702296 with lod 4.50 and P = 0.003. (b) Phenotype by genotype plot of Box–Cox normalized serum ALT levels in 210 F2 mice at rs3702296. 86/86 Represents animals homozygous for BcA86 alleles. B10/B10 represents animals homozygous for B10 alleles. 86/B10 represents animals heterozygous for the two allele types. F2 mice homozygous for the BcA86 allele at this locus showed increased susceptibility to CVB3-induced viral hepatitis. (c) Phenotype by genotype plot of raw serum ALT levels at rs3702296. (d) Lod score plot for serum ALT (solid line), liver viral titer (dotted line) and heart viral titer (dashed line) on chromosome 17 calculated by using interval mapping. Peak linkage to heart viral titer occurs at D17Mit88 with lod 3.4 and P = 0.046. (e) Phenotype by genotype plot of Box–Cox normalized heart viral titer in 102 F2 mice at D17Mit88. F2 mice homozygous for the BcA86 allele at this locus showed increased viral replication in the heart. (f) Phenotype by genotype plot of raw heart viral titer in log10 pfu/mg units at D17Mit88.
respectively. Mice inheriting homozygous BcA86 genotype on chromosome 17 had an average heart viral titer of 4.32 log[pfu/mg] and mice homozygous for B10 had an average heart viral titre of 3.28 log[pfu/mg]. Differential control of type I Interferon in BcA86 and B10 mice and in F2 mice Given the known role of the Type I IFN circuit in hepatic control of CVB3,10–13 quantitative PCR with reverse transcription was Genes and Immunity (2015) 261 – 267
performed to ascertain differences in gene expression between B10, and BcA86 for coxsackieviral mRNA, and select cytokines (Figure 6). In addition to the parental strains, [BcA86xB10]F2 mice homozygous at the peak marker (rs3702296) for each genotype were included in the analysis. Unlike viral titer by plaque forming assay (Figure 2b), differential viral mRNA was observed between BcA86 and B10 by using the more sensitive technique of PCR reverse transcription (P = 0.008, Mann–Whitney). Differential viral mRNA levels and titer levels (not shown) were not recapitulated in the F2 mice, consistent with no significant linkage between © 2015 Macmillan Publishers Limited
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
Figure 6. Differential expression of type I Interferon based on genotype at the distal chromosome 13. Quantitative PCR with reverse transcription comparing CVB3 mRNA levels (a) and IFNβ mRNA (b) levels in B10 and BcA86 mice and in [BcA86xB10]F2 mice (n ⩾ 5) at 48 h post infection. F2-chr13@102(B10) signifies mice that possess homozygous B10 alleles at rs3702296. F2-chr13@102(BcA86) signifies mice that possess homozygous BcA86 alleles at rs3702296. Significantly higher viral mRNA was detected in BcA86 mice (P = 0.008, Mann–Whitney), but CVB3 mRNA was equivalent between the selected [BcA86xC67BL/10]F2 mice (P = 0.6, one-tail Mann–Whitney). Significantly higher Ifnb1 RNA was detected in BcA86 mice (P = 0.010, T-test), and also in F2-chr13@102(BcA86) mice as compared with F2-chr13@102(B10) mice (P = 0.0263, one-tail T-test). Bars represent means ± s.d.
rs3702296 and control of viral titer (Figure 6a). Type I IFN measured by mRNA levels of Ifnb was elevated in BcA86 mice as compared with B10 (P = 0.01) as well as in F2 mice carrying the BcA86 genotype (P = 0.026 one-tailed T-test) [Figure 6b]. Other cytokines (Il6, Il1b, Tnfa) were similarly tested, but none achieved statistical significance within the F2 mice (not shown). These results identify the presence of a locus controlling susceptibility to CVB3, which results in elevated serum ALT, and increased hepatic expression of interferon ß following an infection. DISCUSSION Most individuals appear to become infected with CVB3 or closely related species at some point in their lives,2,28,29 and systemic infection can be severe.3–5 The contrast raised between comparatively rare reports of severe symptoms with a common infection suggests a non-uniform response to the virus. There remains a need to account for the divergent clinical presentation between otherwise normal and immunocompetent individuals. Mouse models represent a simplified system where aspects of pathogenesis can be modulated and interrogated one at a time. Natural variation in response to Coxsackievirus in the mouse model is diverse, and depends on host genetic factors,8,18–20,30 viral genetic factors31–33 and also environmental factors such as age,34 sex35,36 or diet.37,38 To study host genetic factors which determine ability to control infection within relatively resistant B6 and relatively susceptible A/J mice, we have made use of a RC strain panel.39 In a typical RC strain study,21,40 strains are compared against their respective genetic background (for example, BcA vs B6 and AcB vs A/J) to identify discordant strains. In this manner, donor segments conferring susceptibility (or resistance) can then be mapped either directly or in a subsequent test cross. In the present study, it was determined that most BcA strains were more susceptible to CVB3 infection than B6 mice. This suggests that the resistance of B6 mice is either relatively easily overcome by a polygenic A/J donated susceptibility, or that susceptibility-conferring mutations had occurred during or before the creation of this panel. Three out of seventeen strains challenged completely survived the CVB3 infection, of these there were two that carried the A/J allele of the © 2015 Macmillan Publishers Limited
major histocompatibility complex (H2a). Previous reports have indicated a role for H2 not oly in controlling CVB3 infection but also that non-H2 loci seem to control the majority of phenotypic variance between strains.8,20,30 This is in agreement with our findings, and so the specific role of H2a in enhancing or maintaining B6 resistance was not pursued further. Of the BcA strains, the most discordant studied was strain BcA86, which succumbed to infection rapidly (2–4 days), and presented greatly elevated serum ALT (a marker of necrotic hepatocyte death) 48 h post infection. An informative F2 mapping cross between BcA86 and B10 revealed one locus on chromosome 13 that controlled serum ALT. The locus increased median ALT levels by 6834 IU/L: mice homozygous for BcA86 at this locus (median 8820IU/L) compared with mice homozygous for B10 (median 1986IU/L). Further phenotyping of F2 mice with a homozygous BcA86 genotype at this locus revealed approximately double the levels of Ifnb1 expression as compared with mice homozygous for B10. Lack of significant linkage of viral replication and chromosome 13, with no observed difference in CVB3 viral RNA levels in these same F2 mice suggests a novel phenotype unlike MDA-5, IFNAR1 or IFN-ß knockouts,10,11,14 where high viral replication in the liver is accompanied by hepatic necrosis and early mortality. The one-lod interval was calculated using interval mapping as being between 85 Mb and the end of the chromosome (120 Mb). This region contains over 199 known protein coding genes, several of which would make interesting candidates for further study. The RAS P21 protein activator (GTPase activating protein) 1 (Rasa1) gene located towards the proximal end of the interval (85.2 Mb) was recently identified to be strongly antiviral in the context of CVB3 infection of polarized human brain microvascular endothelial cells.41 Coagulation factor II (thrombin) receptor (F2r or PAR1) located at 95.6 Mb has been shown to control CVB3 infection in the heart and liver at late (but not early) time points, and may be involved in cooperative signaling with TLR3.42 3-hydroxy-3-methylglutaryl-coenzyme A reductase (Hmgcr 96.6 Mb), the enzymatic target of statin class of pharmaceuticals, performs a rate limiting step in cholesterol biosynthesis. The link between CVB3 replication and cellular cholesterol availability was recently established in a study that demonstrated CVB3 causes cholesterol to be trafficked from the plasma membrane and the specialized organelles for viral production.43 Phosphatidylinositol 3-kinase, regulatory subunit, polypeptide 1 (p85 alpha) (Pi3kr1) located at 101 Mb is a regulatory subunit of PI3K. It has been shown in HeLa cells that PI3K signaling is beneficial for CVB3 replication, presumably by providing a survival signal (possibly via NFkB) to counter apoptotic signals.44,45 A gene cluster of NLR family, apoptosis inhibitory proteins (Naip1-Naip7, 100 Mb) have been reported to bind bacterial components and lead to cellular pyroptosis.46,47 Distally, the IL-6 receptor signal transducer (Il6st, or gp130) is located at 112 Mb. Gp130 signalling does not appear to be directly antiviral, but provides a strong cytoprotective effect in the context of CVB3 infection. Furthermore, cardiac specific ablation of gp130 signaling by SOCS3 overexpression leads to greatly increase cardiac necrosis, and lethality between 4 and 6 days after infection.48 Mutations in one or more of these genes or others with no known role in CVB3 infection are possible, and further studies will include exome sequencing and candidate gene analysis. MATERIALS AND METHODS Mice, cells and virus Inbred B6, B10, and A/J mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). The set of 36 AcB/BcA RC strains, derived from A/J and B6 parents, were generated according to a previously described breeding scheme and genotyping protocol.39 (BcA86 X B10) and (B10 X BcA86) F1 crosses were generated; F2 crosses were then performed by Genes and Immunity (2015) 261 – 267
265
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
266 standard (F1brother X F1sister) mating. The mice were maintained in the McGill University animal facility in compliance with the Canada Council on Animal Care as approved by the McGill University Animal Care Committee. CVB3 used for all experiments was CVB3-CG strain as previously.49 HeLa cells (ATCC: CCL-2) were grown and maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% FBS, 100 μg/ml Penicillin/ Streptomycin (1 × media). Homogenized organs were serially diluted 10-fold in non-supplemented DMEM and the dilutions were used to infect HeLa cells in triplicate for 60 min. After initial infection by homogenate, HeLa cells are covered in a layer of media with 0.25% agarose and incubated for three days. Cells were fixed in formaldehyde and stained with crystal violet to count plaques as previously.18
Mouse infection and phenotype determination For each experiment, a vial of viral stock was thawed and diluted in cold PBS at 5 × 104 PFU/ml. Inbred, F1 and F2 mice were inoculated intraperitoneally with 400 PFU of CVB3-CG per gram of body weight at 7–9 weeks of age. Animals were weighed daily and observed for changes in fur characteristics, level of activity and appearance of respiratory distress. During the RC strain survival screen, moribund mice were humanely euthanized at an acceptable clinical end point. A total of 70 BcA mice and 30 AcB mice, and 210 [BcA86xC67BL/10]F2 mice were phenotyped. For subsequent experiments, mice were euthanized at day 2 post infection, and exsanguinated by cardiac puncture to serum. Heart and liver tissues were removed aseptically, snap frozen and stored at − 80 °C in two separate aliquots. One portion was subsequently homogenized and submitted to three freeze-thaw cycles to release infectious virus for determination of viral titre by plaque assay. Total RNA was extracted from the other portion by TRIzol (Ambion, Burlington, ON, Canada), and reverse transcribed by MMLV (New England Biolabs, Ipswich, MA, USA) by using random hexamers according to the manufacturers’ instructions. Serum levels of creatine kinase and ALT were measured by the McGill Comparative Medicine Animal Resources Centre's Diagnostic and Research Support Service by using a colorimetric method. Quantitative PCR primers were designed to span exon junctions with the help of primer3plus.50 The primer sequences are as follows: CVB3-H3 forward: ATGGCAGAAAACCTCACCAG; CVB3-H3 reverse: ATGTTCTGCCCAAACAGTCC; Ifnb1 forward: TGACGGAGAAGATGCAGAAG; Ifnb1 reverse: ACCCAGTGC TGGAGAAATTG.
Genotype analysis The B10 strain is not a Mouse Phenome Project priority strain (http:// phenome.jax.org/), and so existing publically available genotype information is limited. Most single-nucleotides polymorphic between B10 and B6 for use in the generation of a panel capable of distinguishing between B10 and BcA86 were identified by using the JAX Diversity Array (Jackson Laboratory).51 In this panel, 7499 single-nucleotides were polymorphic between B6 and B10, though these tended to be clustered and gaps remained (Supplementary Figure 1). These gaps in the genetic map were filled in by comparing the genotype of B6 against all other C57BL strains available in all datasets contained within the Mouse Phenome Database. When the B6 genotype was divergent, B6 and B10 were genotyped by Sanger sequencing to confirm the private B6 single-nucleotide polypeptide. Finally, the map was completed with microsatellites identified by us as previously,19 and in collaboration with others26 for a total of 106 markers that are informative between BcA86 and B10. All single-nucleotide polymorphism genotyping was performed by Sequenom iPlex Gold technology. Genotyping of the F2 was performed at the McGill University and Genome Quebec Innovation Centre (Montreal, QC, Canada). Microsatellite PCR reactions performed manually by using a 10-μl total reaction volume, 200-μM dNTPs, 1.5- mM MgCl2, 2 pmol of each primer and 0.5 U of Taq polymerase (Invitrogen, Burlinton, ON, USA). Reactions were performed as follows: 96˚ for 2 min; 30 cycles of 94˚ for 45 s, 56˚ for 45 s, 72˚ for 60 s; and a final extension step 72˚ 7 min. PCR products were then separated on 2–4% high-resolution agarose (USB, Cleveland, OH, USA) gels containing Ethidium bromide and visualized under UV light.
Statistical analysis Statistical analyses were conducted with the freely available program R (www.r-project.org) and all packages were installed using the 'install. packages' function in R. Linkage was performed with the package ‘R/qtl’ version 3.02. Survival analysis was performed using the ‘Survival’ package Genes and Immunity (2015) 261 – 267
version 2.37. Skewness was detected using the D’Agostino test from the ‘moments’ package version 0.13. The normality of phenotypic distributions was assessed using the Shapiro Test from the ‘stats’ package version 2.15.3. Serum ALT, liver and heart viral titre F2 phenotypes were normalized using Box–Cox52 method as implemented in the ‘Caret’ package version 5.17. Briefly, for every value, x, in a phenotypic distribution, the Box–Cox method λ transforms x according to the formula x 0λ ¼ x λ- 1, where λ is optimized such 0 that the distribution of x λ best approximates the normal distribution. The determined value of λ was 0.3 for ALT and liver viral titer and 0.2 for heart viral titer. QQPLOTs were generated using the qqnorm function in the ‘stats’ package version 2.15.3. Viral replication is also reported as Log10[Pfu/ mg] in figures for the sake of interpretability. The scanone function of the ‘R/qtl’ package version 1.28-19, was used to perform Haley–Knott regression of the phenotype on genetic markers. Significance values were evaluated with 10 000 permutations. The lod support interval of quantitative trait loci peaks was calculated using a 1.5 lod drop by ‘R/qtl’ on the interval map. P-values shown in the figures are a result of two-tailed t-tests with *Po0.05, **P o0.01 and ***Po0.001 unless otherwise indicated.
CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGEMENTS This work was supported by grants from the Canadian Institutes of Health Research (CIHR) SAW, GAL and JM were supported by FRSQ Scholarships and SMV by the Canada Research Chair program.
REFERENCES 1 Melnick JL, Godman GC. Pathogenesis of coxsackie virus infection; multiplication of virus and evolution of the muscle lesion in mice. J Exp Med 1951; 93: 247–266. 2 Gkrania-Klotsas E, Langenberg C, Tauriainen S, Sharp SJ, Luben R, Forouhi NG et al. The association between prior infection with five serotypes of Coxsackievirus B and incident type 2 diabetes mellitus in the EPIC-Norfolk study. Diabetologia 2012; 55: 967–970. 3 Cheng LL, Ng PC, Chan PK, Wong HL, Cheng FW, Tang JW. Probable intrafamilial transmission of coxsackievirus b3 with vertical transmission, severe early-onset neonatal hepatitis, and prolonged viral RNA shedding. Pediatrics 2006; 118: e929–e933. 4 Tseng F-C, Huang H-C, Chi C-Y, Lin T-L, Liu C-C, Jian J-W et al. Epidemiological survey of enterovirus infections occurring in Taiwan between 2000 and 2005: analysis of sentinel physician surveillance data. J Med Virol 2007; 79: 1850–1860. 5 Wang SM, Liu CC, Yang YJ, Yang HB, Lin CH, Wang JR. Fatal coxsackievirus B infection in early infancy characterized by fulminant hepatitis. J Infect 1998; 37: 270–273. 6 Yajima T, Knowlton KU. Viral myocarditis: from the perspective of the virus. Circulation 2009; 119: 2615–2624. 7 Gauntt C, Huber S. Coxsackievirus experimental heart diseases. Front Biosci 2003; 8: e23–e35. 8 Wolfgram L, Beisel K, Herskowitz A, Rose NR. Variations in the susceptibility to Coxsackievirus B3-induced myocarditis among different strains of mice. J Immunol 1986; 136: 1846–1852. 9 Gauntt C, Gomez P, Duffey P, Grant J, Trent D, Witherspoon S et al. Characterization and myocarditic capabilities of coxsackievirus B3 variants in selected mouse strains. J Virol 1984; 52: 598–605. 10 Wessely R, Klingel K, Knowlton KU, Kandolf R. Cardioselective infection with coxsackievirus B3 requires intact type I interferon signaling: implications for mortality and early viral replication. Circulation 2001; 103: 756–761. 11 Deonarain R, Cerullo D, Fuse K, Liu P, Fish E. Protective role for interferon-beta in coxsackievirus B3 infection. Circulation 2004; 110: 3540–3543. 12 Yasukawa H, Yajima T, Duplain H, Iwatate M, Kido M, Hoshijima M et al. The suppressor of cytokine signaling-1 (SOCS1) is a novel therapeutic target for enterovirus-induced cardiac injury. J Clin Invest 2003; 111: 469–478. 13 Richer MJ, Lavallee DJ, Shanina I, Horwitz MS. Toll-like receptor 3 signaling on macrophages is required for survival following coxsackievirus B4 infection. PLoS One 2009; 4: e4127. 14 Wang JP, Cerny A, Asher DR, Kurt-Jones EA, Bronson RT, MDA5 Finberg RW. and MAVS mediate type I interferon responses to coxsackie B virus. J Virol 2010; 84: 254–260.
© 2015 Macmillan Publishers Limited
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
267 15 Huber SA, Roberts B, Moussawi M, Boyson JE. Slam haplotype 2 promotes NKT but suppresses Vgamma4+ T-cell activation in coxsackievirus B3 infection leading to increased liver damage but reduced myocarditis. Am J Pathol 2013; 182: 401–409. 16 Huhn MH, McCartney SA, Lind K, Svedin E, Colonna M, Flodstrom-Tullberg M. Melanoma differentiation-associated protein-5 (MDA-5) limits early viral replication but is not essential for the induction of type 1 interferons after Coxsackievirus infection. Virology 2010; 401: 42–48. 17 Chow LH, Gauntt CJ, McManus BM. Differential effects of myocarditic variants of Coxsackievirus B3 in inbred mice. A pathologic characterization of heart tissue damage. Lab Invest 1991; 64: 55–64. 18 Wiltshire SA, Leiva-Torres GA, Vidal SM. Quantitative trait locus analysis, pathway analysis, and consomic mapping show genetic variants of Tnni3k, Fpgt, or H28 control susceptibility to viral myocarditis. J Immunol 2011; 186: 6398–6405. 19 Aly M, Wiltshire SA, Chahrour G, Osti J-CL, Vidal SM. Complex genetic control of host susceptibility to coxsackievirus B3-induced myocarditis. Genes Immun 2007; 8: 193–204. 20 Herskowitz A, Wolfgram L, Rose NR, Beisel K. Coxsackievirus B3 murine myocarditis: a pathologic spectrum of myocarditis in genetically defined inbred strains. J Am Coll Cardiol 1987; 9: 1311–1319. 21 Wiltshire SA, Diez E, Miao Q, Dubé M-P, Gagné M, Paquette O et al. Genetic control of high density lipoprotein-cholesterol in AcB/BcA recombinant congenic strains of mice. Physiol Genomics 2012; 44: 843–852. 22 Min-Oo G, Fortin A, Tam M, Nantel A, Stevenson M, Gros P. Pyruvate kinase deficiency in mice protects against malaria. Nat Genet 2003; 35: 357–362. 23 Chow L, Gauntt C, McManus B. Differential effects of myocarditic variants of Coxsackievirus B3 in inbred mice. A pathologic characterization of heart tissue damage. Lab Invest 1991; 64: 55–64. 24 Wessely R, Klingel K, Knowlton K, Kandolf R. Cardioselective infection with coxsackievirus B3 requires intact type I interferon signaling: implications for mortality and early viral replication. Circulation 2001; 103: 756–761. 25 Yang RQ, Yi NJ, Xu SZ. Box-Cox transformation for QTL mapping. Genetica 2006; 128: 133–143. 26 Xia Y, Won S, Du X, Lin P, Ross C, La Vine D et al. Bulk segregation mapping of mutations in closely related strains of mice. Genetics 2010; 186: 1139–1146. 27 Bongfen SE, Rodrigue-Gervais I-G, Berghout J, Torre S, Cingolani P, Wiltshire SA et al. An N-Ethyl-N-Nitrosourea (ENU)-Induced Dominant Negative Mutation in the JAK3 Kinase Protects against Cerebral Malaria. PloS One 2012; 7: e31012. 28 Khetsuriani N, Lamonte-Fowlkes A, Oberst S, Pallansch M. Enterovirus surveillance--United States, 1970-2005. MMWR Surveill Summ 2006; 55: 1–20. 29 CfDCaP. CDC. Nonpolio enterovirus and human parechovirus surveillance— United States, 2006–2008. MMWR Morb Mortal Wkly Rep 2010; 59: 1577–1580. 30 Traystman MD, Chow LH, McManus BM, Herskowitz A, Nesbitt MN, Beisel KW. Susceptibility to Coxsackievirus B3-induced chronic myocarditis maps near the murine Tcr alpha and Myhc alpha loci on chromosome 14. Am J Pathol 1991; 138: 721–726. 31 Knowlton KU, Jeon ES, Berkley N, Wessely R, Huber S. A mutation in the puff region of VP2 attenuates the myocarditic phenotype of an infectious cDNA of the Woodruff variant of coxsackievirus B3. J Virol 1996; 70: 7811–7818. 32 Stadnick E, Dan M, Sadeghi A, Chantler JK. Attenuating mutations in coxsackievirus B3 map to a conformational epitope that comprises the puff region of VP2 and the knob of VP3. J Virol 2004; 78: 13987–14002. 33 Schmidtke M, Merkle I, Klingel K, Hammerschmidt E, Zautner AE, Wutzler P. The viral genetic background determines the outcome of coxsackievirus B3 infection in outbred NMRI mice. J Med Virol 2007; 79: 1334–1342.
34 Gay RT, Belisle S, Beck MA, Meydani SN. An aged host promotes the evolution of avirulent coxsackievirus into a virulent strain. Proc Natl Acad Sci USA 2006; 103: 13825–13830. 35 Frisancho-Kiss S, Davis SE, Nyland JF, Frisancho JA, Cihakova D, Barrett MA et al. Cutting edge: cross-regulation by TLR4 and T cell Ig mucin-3 determines sex differences in inflammatory heart disease. J Immunol 2007; 178: 6710–6714. 36 Frisancho-Kiss S, Nyland JF, Davis SE, Frisancho JA, Barrett MA, Rose NR et al. Sex differences in coxsackievirus B3-induced myocarditis: IL-12Rbeta1 signaling and IFN-gamma increase inflammation in males independent from STAT4. Brain Res 2006; 1126: 139–147. 37 Woodruff JF. The influence of quantitated post-weaning undernutrition on coxsackievirus B3 infection of adult mice. II. Alteration of host defense mechanisms. J Infect Dis 1970; 121: 164–181. 38 Woodruff JF, Woodruff JJ. Modification of severe coxsackievirus B 3 infection in marasmic mice by transfer of immune lymphoid cells. Proc Natl Acad Sci USA 1971; 68: 2108–2111. 39 Fortin A, Diez E, Rochefort D, Laroche L, Malo D, Rouleau G et al. Recombinant congenic strains derived from A/J and C57BL/6 J: a tool for genetic dissection of complex traits. Genomics 2001; 74: 21–35. 40 Boivin GA, Pothlichet J, Skamene E, Brown EG, Loredo-Osti JC, Sladek R et al. Mapping of clinical and expression quantitative trait loci in a sex-dependent effect of host susceptibility to mouse-adapted influenza H3N2/HK/1/68. J Immunol 2012; 188: 3949–3960. 41 Coyne CB, Bozym R, Morosky SA, Hanna SL, Mukherjee A, Tudor M et al. Comparative RNAi screening reveals host factors involved in enterovirus infection of polarized endothelial monolayers. Cell Host Microbe 2011; 9: 70–82. 42 Antoniak S, Owens AP 3rd, Baunacke M, Williams JC, Lee RD, Weithauser A et al. PAR-1 contributes to the innate immune response during viral infection. J Clin Investig 2013; 123: 1310–1322. 43 Ilnytska O, Santiana M, Hsu NY, Du WL, Chen YH, Viktorova EG et al. Enteroviruses harness the cellular endocytic machinery to remodel the host cell cholesterol landscape for effective viral replication. Cell Host Microbe 2013; 14: 281–293. 44 Esfandiarei M, Boroomand S, Suarez A, Si X, Rahmani M, McManus B. Coxsackievirus B3 activates nuclear factor kappa B transcription factor via a phosphatidylinositol-3 kinase/protein kinase B-dependent pathway to improve host cell viability. Cell Microbiol 2007; 9: 2358–2371. 45 Esfandiarei M, Luo H, Yanagawa B, Suarez A, Dabiri D, Zhang J et al. Protein kinase B/Akt regulates coxsackievirus B3 replication through a mechanism which is not caspase dependent. J Virol 2004; 78: 4289–4298. 46 Diez E, Lee SH, Gauthier S, Yaraghi Z, Tremblay M, Vidal S et al. Birc1e is the gene within the Lgn1 locus associated with resistance to Legionella pneumophila. Nat Genet 2003; 33: 55–60. 47 Bortoluci KR, Medzhitov R. Control of infection by pyroptosis and autophagy: role of TLR and NLR. Cell Mol Life Sci 2010; 67: 1643–1651. 48 Suppiah V, Moldovan M, Ahlenstiel G, Berg T, Weltman M, Abate ML et al. IL28B is associated with response to chronic hepatitis C interferon-alpha and ribavirin therapy. Nat Genet 2009; 41: 1100–1104. 49 Aly M, Wiltshire S, Chahrour G, Osti JC, Vidal SM. Complex genetic control of host susceptibility to coxsackievirus B3-induced myocarditis. Genes Immun 2007; 8: 193–204. 50 Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 2000; 132: 365–386. 51 Yang H, Ding Y, Hutchins L, Szatkiewicz J, Bell T, Paigen B et al. A customized and versatile high-density genotyping array for the mouse. Nat Methods 2009; 6: 663–666. 52 Box GCD. An Analysis of Transformations. J R Stat Soc Ser B 1964; 16: 211–252.
Supplementary Information accompanies this paper on Genes and Immunity website (http://www.nature.com/gene)
© 2015 Macmillan Publishers Limited
Genes and Immunity (2015) 261 – 267
Copyright of Genes & Immunity is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
ORIGINAL ARTICLE
Mapping of a quantitative trait locus controlling susceptibility to Coxsackievirus B3-induced viral hepatitis SA Wiltshire, J Marton, GA Leiva-Torres and SM Vidal The pathogenesis of coxsackieviral infection is a multifactorial process involving host genetics, viral genetics and the environment in which they interact. We have used a mouse model of Coxsackievirus B3 infection to characterize the contribution of host genetics to infection survival and to viral hepatitis. Twenty-five AcB/BcA recombinant congenic mouse strains were screened. One, BcA86, was found to be particularly susceptible to early mortality; 100% of BcA86 mice died by day 6 compared with 0% of B6 mice (P = 0.0012). This increased mortality was accompanied by an increased hepatic necrosis as measured by serum alanine aminotransferase (ALT) levels (19547 ± 10556 vs 769 ± 109, P = 0.0055). This occurred despite a predominantly resistant (C57BL/6) genetic background. Linkage analysis in a cohort (n = 210) of (BcA86x C56Bl/10)F2 animals revealed a new locus on chromosome 13 (peak linkage 101.2 Mbp, lod 4.50 and P = 0.003). This locus controlled serum ALT levels as early as 48 h following the infection, and led to an elevated expression of type I interferon. Another locus on chromosome 17 (peak linkage 57.2 Mbp) was significantly linked to heart viral titer (lod 3.4 and P = 0.046). These results provide new evidence for the presence of genetic loci contributing to the susceptibility of mice to viral hepatitis. Genes and Immunity (2015) 16, 261–267; doi:10.1038/gene.2015.5; published online 19 March 2015
INTRODUCTION Coxsackievirus B3 (CVB3) is frequently studied in the context of myocarditis and its long-term sequala dilated cardiomyopathy. However, coxsackievirus infection is systemic: early accounts detail that CVB3 is capable of inducing myositis, meningitis, pancreatitis, hepatitis as well as myocarditis.1 A recent study based on seropositivity in diabetics estimates that up to 75% of people are seropositive to any given strain of coxsackievirus, 66% for CVB3 alone.2 In contrast, severe symptoms are comparatively rare. In newborns, severe cases of viral hepatitis are periodically reported as a result of group B Coxsackieviruses, including CVB3.3–5 Clinically observed heterogeneity is recapitulated in the mouse model, where disease progression and presentation depend on both host and viral factors.6–9 The central role of the host immune response in determining the outcome of CVB3 infection has been clearly demonstrated in a number of reverse genetic (knockout) studies. The type I Interferon (IFN) axis seems to be central in controlling viral replication within the liver, but may be redundant in controlling viral replication in the heart. IFN receptor (Ifnar1) knockout mice succumbed early to a lethal infection, with greatly increased viral replication in the liver, and 10-fold higher liver damage as measured by serum alanine aminotransferase (ALT) compared with the wild type. However no significant changes in viral replication or inflammation were observed in Ifnar1− / −hearts,10 observations that are mirrored in IFNβ knockout mice.11 Type I IFN signalling therefore seems to have a pivotal role in the liver with no apparent effect in the heart. Conversely, abrogation of all JAK/STAT signalling by the transgenic overexpression of the suppressor of cytokine signalling 1 (Socs1), greatly enhanced the myocarditis with no effect in the liver.12 Deficiency for toll-like receptor 3 (Tlr3), a viral pattern recognition receptor sensitive to lumenal double stranded RNA, also increases
susceptibility to myocarditis, but not hepatitis.13 Absence of the cytoplasmic dsRNA sensor melanoma differentiation-associated protein 5 (Ifih1 or MDA5) leads to a greatly increased mortality, accompanied by a severe hepatic necrosis14 and greatly elevated ALT levels. Absence of the MDA-5 transducing mitochondrial antiviral signalling protein (Mavs) also leads to an early mortality (days 2–3) accompanied by a severe hepatic necrosis. Further complicating matters, the genetic background has both characterized and unknown effects that contribute to pathogenesis and tissue tropism. The signal lymphocytic activation molecule 2 haplotype controls heart vs liver tropism in the C57BL/6 (B6) background.15 Moreover, while in the B6 genetic background MDA5 is essential for the survival of CVB3 infection and protection from an hepatic necrosis, MDA5 deficiency in the 129SvJ background has no apparent effect.16 Forward genetic analysis by using mouse models constitutes a powerful tool to study the natural variation and background variation in complex traits and disease. It is known that the A/J and B6 strains are respectively susceptible and resistant to CVB3 infection.17 We have previously made use of a forward genetic approach to scan the mouse genome for loci linked to susceptibility to CVB3-induced myocarditis between the A/J and B6 strains.18,19 To identify additional loci controlling susceptibility to CVB3 infection, the recombinant congenic (RC) AcB/BcA panel was screened. RC strains are generated by two sequential backcrosses of one inbred donor strain (12.5% of the genome) to a background recipient strain (87.5% of the genome) and have been developed as a tool for complex trait mapping. Following a screen of the AcB/BcA panel, one strain (BcA86) with a dramatically discordant phenotype was singled out for further analysis. Viral replication and common serum biochemical markers were tested to characterize pathogenesis in BcA86 mice. Serum
Department of Human Genetics, McGill University, Montreal, QC, Canada. Correspondence: Dr SM Vidal, Department of Human Genetics, McGill University, McGill Life Sciences, Complex, Room 367, 3649 Promenade Sir William Osler, Montreal H3G 0B1, QC, Canada. E-mail: [email protected] Received 7 November 2014; revised 23 December 2014; accepted 5 January 2015; published online 19 March 2015
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
262 ALT in the BcA86 mice was found to be elevated compared with other strains, and was pursued by using an informative F2-cross approach. Linkage analysis revealed a single significant locus on the distal chromosome 13 that controls ALT levels. RESULTS Mice of the BcA86 RC strain succumb to lethal CVB3 infection The course of infection with CVB3 is highly variable and depends on both host and viral background.8,9 It is known that the A/J and B6 strains are respectively susceptible and resistant to CVB3 infection,17 though little is known about the genetic determinants underlying this trait. It was therefore of interest to challenge the RC strain panel of AcB/BcA mice with CVB3 infection. The screen was undertaken by using an infectious dose which had previously been used to study myocarditic potential of CVB3 in A/J and B10. A-H2a mice.19 It was observed early into screening the RC strain panel that the majority of AcB (A/J background, B6 donor) and BcA (B6 background, A/J donor) mice did not survive the infection to the previously established experimental end point on day 8. Therefore survival of infection was used as a phenotype. In total, 109 male RC mice were screened, with between 2 and 8 per strain (median 4) (Figure 1a). Consistent with previous reports of the involvement of H2 in susceptibility to CVB3 infection,8,17,20 the two H2a-carrying BcA strains (BcA69 and BcA74) were relatively resistant, and survived the infection. Most other strains were comparatively susceptible to lethal infection. In particular, the BcA86 strain succumbed to early mortality between days 2 and 5 post infection (n = 6, P = 0.0012 compared with B6) (Figure 1b). Further phenotyping was conducted in a panel of A/J, BcA86, B6 and the closely related C57BL/10 (B10). B10 mice, which are resistant to CVB3, were included knowing that the genetic architecture underlying the phenotype of BcA86 might include not only A/J loci but also de novo mutations on the B6 background. Two de novo mutations have been mapped so far within the AcB/ BcA panel on the background strain.21,22 Reports of early mortality within CVB3 infection have implicated metabolic failure secondary to hepatic necrosis,14,23,24 so viral titer in the heart and liver, as well as serum biomarkers for heart and liver damage were assessed at the early time point of 48 h post infection (Figure 2). It was found that viral titer in BcA86 was slightly higher in the heart than in B6 (one-way analysis of variance with Tukeys multiple comparison test, P = 0.03), while being comparable in the spleen and liver. No significant differences in the heart damage biomarker creatine kinase were observed at this time point. BcA86 mice showed a marked increase in serum levels of the liver damage biomarker ALT as compared with A/J, B6 and B10 mice (one-way analysis of variance with Tukeys multiple comparison test P = 0.0005, 0.0055 and 0.0007). The BcA86 phenotype of elevated
ALT relative to B10 mice was reproduced in independent experiments (data not shown). Segregation analysis of serum alanine aminotransferase levels, and viral titer reveals complex inheritance In order to identify the genetic determinants conferring susceptibility to BcA86 mice, we undertook a [B10x BcA86]F2 cross (N = 210) [Figure 3]. Segregation analysis indicated that serum ALT levels, liver and heart viral titer were all highly skewed distributions (+2.45, +2.90, +6.61, respectively, Po1e-9) [Figures 3a,c and e] that deviated very significantly from the normal distribution (P = 3.35 × 10 − 16, 3.706 × 10 − 19 and 4.995 × 10 − 18). Given that quantitative trait loci mapping assumes normal phenotypic distributions, these phenotypes were normalized by using the Box–Cox method (Figures 3b,d and f).25 The distributions of the transformed data resembled a normal distribution more closely than those of the raw data (P = 0.03424, 0.0002131 and 0.04858). Viral titer in the liver was significantly correlated with serum ALT (R2 = 0.44), and also with heart viral titer but to a lesser degree (R2 = 0.30). Heart and liver viral titer were also correlated to each other (R2 = 0.36). These results are consistent with common and distinct genetic determinants of susceptibility within the [BcA86x B10]F2 mice. Linkage analysis reveals two loci controlling susceptibility to CVB3 infection in [BcA86x B10]F2mice In order to perform linkage analysis, a panel of genetic markers was created to distinguish between BcA86 and B10 backgrounds (described further in methods). The full B6/B10 panel has been successfully used and described elsewhere.26,27 Linkage to serum ALT identified a single significant peak on the distal chromosome 13 (lod = 4.50, P = 0.003, 9.4% variance explained, peak marker rs3702296) (Figure 4 and Figure 5a). No significant linkage was found to viral titer in the liver, however the maximum lod score (2.81) and lowest P-value (0.14) were also obtained at marker rs3702296. Linkage to viral titer in the heart revealed one peak on chromosome 17 (lod = 3.4, P = 0.046) (Figure 4, Figure 5d). The 1-lod support interval on chromosome 13 extends between 86 Mb and the end of the chromosome 120 Mb, with a peak of linkage at 102 Mb. On chromosome 17, the 1-lod support interval extends between 49.7 and 69.7 Mb. An A/J haplotype block beginning at 47.7 Mb and extending to 58.6 Mb, is present within the chromosome 17 interval. Allele effect plots indicated that in both cases, the BcA86 genotype conferred susceptibility at the two loci (Figures 5b,c,e and f). Mice inheriting homozygous BcA86 alleles on chromosome 13 had a median serum ALT level of 8820IU/L and mice homozygous for B10 had a median serum ALT level of 1986IU/L. These same mice homozygous for chromosome 13 had average liver viral titers of 4.68 log[pfu/mg] and 4.01 log[pfu/mg],
Figure 1. Survival of CVB3 infection in AcB/BcA RC strains of mice. (a) Strain distribution plot of mean survival time for RC strains of mice. Between 2 and 8 male mice per strain (median 4) were tested for a total of 109 mice. Both H2a carrying BcA strains (BcA69 and BcA74) survived the infection. Bars represent means ± s.d. (b) The BcA86 strain succumbed to early mortality between days 2 and 5 post infection (n = 6, P = 0.0012 compared with B6). Genes and Immunity (2015) 261 – 267
© 2015 Macmillan Publishers Limited
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
263
Figure 2. Quantification of viral titers and serum metabolites in BcA86 mice compared with genetic background and donor strains. Viral replication in the heart (a), liver (b) and spleen (c) were measured in A/J, BcA86, B6, B10 mice (n ⩾ 4) 2 days following the infection. Only viral replication in the heart was statistically significantly different between the four strains of mice (one-way analysis of variance with Tukeys multiple comparison test, P o0.05). (d, e) Levels of serum creatine kinase and ALT in A/J, BcA86, B6 and B10 mice (n ⩾ 5) 2 days following the infection. Serum ALT levels were significantly greater in BcA86 mice (one-way analysis of variance with Tukeys multiple comparison test P = 0.005). Bars represent mean ± s.d.
Figure 3. Distributions of viral replication and ALT within F2 mice. Histograms depicting the distribution of serum ALT (a), liver viral titer (c) and heart viral titer (e) in [BcA86xC67BL/10]F2 (n = 210) mice. Strong skewness was detected in the three phenotypic distributions, and so phenotypes were normalized by Box–Cox (b, d, f) for further analysis. © 2015 Macmillan Publishers Limited
Genes and Immunity (2015) 261 – 267
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
264
Figure 4. Genome-wide linkage analysis in [BcA86xB10]F2 mice reveals two significant loci. Genome-wide linkage scans for day 2 serum ALT (solid line), liver viral titer (dotted line), and heart viral titer (dashed line). Linkage analysis was performed on normalized phenotypes, and genome-wide significance was determined by 10 000 permutations (with alpha = 0.05 shown as a dashed horizontal line). A single locus on distal chromosome 13 was significantly linked to ALT levels (lod = 4.50, P = 0.003). An additional locus on chromosome 17 was linked to viral replication in the heart (lod 3.4, P = 0.046).
Figure 5. Linkage of chromosomes 13 and 17 to CVB3 viral replication and serum ALT levels. (a) Lod score plot of serum ALT (solid line), viral replication in the liver (dotted line) and in the heart (dashed line) calculated by using interval mapping. Peak linkage to serum ALT on distal chromosome 13 occurs at rs3702296 with lod 4.50 and P = 0.003. (b) Phenotype by genotype plot of Box–Cox normalized serum ALT levels in 210 F2 mice at rs3702296. 86/86 Represents animals homozygous for BcA86 alleles. B10/B10 represents animals homozygous for B10 alleles. 86/B10 represents animals heterozygous for the two allele types. F2 mice homozygous for the BcA86 allele at this locus showed increased susceptibility to CVB3-induced viral hepatitis. (c) Phenotype by genotype plot of raw serum ALT levels at rs3702296. (d) Lod score plot for serum ALT (solid line), liver viral titer (dotted line) and heart viral titer (dashed line) on chromosome 17 calculated by using interval mapping. Peak linkage to heart viral titer occurs at D17Mit88 with lod 3.4 and P = 0.046. (e) Phenotype by genotype plot of Box–Cox normalized heart viral titer in 102 F2 mice at D17Mit88. F2 mice homozygous for the BcA86 allele at this locus showed increased viral replication in the heart. (f) Phenotype by genotype plot of raw heart viral titer in log10 pfu/mg units at D17Mit88.
respectively. Mice inheriting homozygous BcA86 genotype on chromosome 17 had an average heart viral titer of 4.32 log[pfu/mg] and mice homozygous for B10 had an average heart viral titre of 3.28 log[pfu/mg]. Differential control of type I Interferon in BcA86 and B10 mice and in F2 mice Given the known role of the Type I IFN circuit in hepatic control of CVB3,10–13 quantitative PCR with reverse transcription was Genes and Immunity (2015) 261 – 267
performed to ascertain differences in gene expression between B10, and BcA86 for coxsackieviral mRNA, and select cytokines (Figure 6). In addition to the parental strains, [BcA86xB10]F2 mice homozygous at the peak marker (rs3702296) for each genotype were included in the analysis. Unlike viral titer by plaque forming assay (Figure 2b), differential viral mRNA was observed between BcA86 and B10 by using the more sensitive technique of PCR reverse transcription (P = 0.008, Mann–Whitney). Differential viral mRNA levels and titer levels (not shown) were not recapitulated in the F2 mice, consistent with no significant linkage between © 2015 Macmillan Publishers Limited
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
Figure 6. Differential expression of type I Interferon based on genotype at the distal chromosome 13. Quantitative PCR with reverse transcription comparing CVB3 mRNA levels (a) and IFNβ mRNA (b) levels in B10 and BcA86 mice and in [BcA86xB10]F2 mice (n ⩾ 5) at 48 h post infection. F2-chr13@102(B10) signifies mice that possess homozygous B10 alleles at rs3702296. F2-chr13@102(BcA86) signifies mice that possess homozygous BcA86 alleles at rs3702296. Significantly higher viral mRNA was detected in BcA86 mice (P = 0.008, Mann–Whitney), but CVB3 mRNA was equivalent between the selected [BcA86xC67BL/10]F2 mice (P = 0.6, one-tail Mann–Whitney). Significantly higher Ifnb1 RNA was detected in BcA86 mice (P = 0.010, T-test), and also in F2-chr13@102(BcA86) mice as compared with F2-chr13@102(B10) mice (P = 0.0263, one-tail T-test). Bars represent means ± s.d.
rs3702296 and control of viral titer (Figure 6a). Type I IFN measured by mRNA levels of Ifnb was elevated in BcA86 mice as compared with B10 (P = 0.01) as well as in F2 mice carrying the BcA86 genotype (P = 0.026 one-tailed T-test) [Figure 6b]. Other cytokines (Il6, Il1b, Tnfa) were similarly tested, but none achieved statistical significance within the F2 mice (not shown). These results identify the presence of a locus controlling susceptibility to CVB3, which results in elevated serum ALT, and increased hepatic expression of interferon ß following an infection. DISCUSSION Most individuals appear to become infected with CVB3 or closely related species at some point in their lives,2,28,29 and systemic infection can be severe.3–5 The contrast raised between comparatively rare reports of severe symptoms with a common infection suggests a non-uniform response to the virus. There remains a need to account for the divergent clinical presentation between otherwise normal and immunocompetent individuals. Mouse models represent a simplified system where aspects of pathogenesis can be modulated and interrogated one at a time. Natural variation in response to Coxsackievirus in the mouse model is diverse, and depends on host genetic factors,8,18–20,30 viral genetic factors31–33 and also environmental factors such as age,34 sex35,36 or diet.37,38 To study host genetic factors which determine ability to control infection within relatively resistant B6 and relatively susceptible A/J mice, we have made use of a RC strain panel.39 In a typical RC strain study,21,40 strains are compared against their respective genetic background (for example, BcA vs B6 and AcB vs A/J) to identify discordant strains. In this manner, donor segments conferring susceptibility (or resistance) can then be mapped either directly or in a subsequent test cross. In the present study, it was determined that most BcA strains were more susceptible to CVB3 infection than B6 mice. This suggests that the resistance of B6 mice is either relatively easily overcome by a polygenic A/J donated susceptibility, or that susceptibility-conferring mutations had occurred during or before the creation of this panel. Three out of seventeen strains challenged completely survived the CVB3 infection, of these there were two that carried the A/J allele of the © 2015 Macmillan Publishers Limited
major histocompatibility complex (H2a). Previous reports have indicated a role for H2 not oly in controlling CVB3 infection but also that non-H2 loci seem to control the majority of phenotypic variance between strains.8,20,30 This is in agreement with our findings, and so the specific role of H2a in enhancing or maintaining B6 resistance was not pursued further. Of the BcA strains, the most discordant studied was strain BcA86, which succumbed to infection rapidly (2–4 days), and presented greatly elevated serum ALT (a marker of necrotic hepatocyte death) 48 h post infection. An informative F2 mapping cross between BcA86 and B10 revealed one locus on chromosome 13 that controlled serum ALT. The locus increased median ALT levels by 6834 IU/L: mice homozygous for BcA86 at this locus (median 8820IU/L) compared with mice homozygous for B10 (median 1986IU/L). Further phenotyping of F2 mice with a homozygous BcA86 genotype at this locus revealed approximately double the levels of Ifnb1 expression as compared with mice homozygous for B10. Lack of significant linkage of viral replication and chromosome 13, with no observed difference in CVB3 viral RNA levels in these same F2 mice suggests a novel phenotype unlike MDA-5, IFNAR1 or IFN-ß knockouts,10,11,14 where high viral replication in the liver is accompanied by hepatic necrosis and early mortality. The one-lod interval was calculated using interval mapping as being between 85 Mb and the end of the chromosome (120 Mb). This region contains over 199 known protein coding genes, several of which would make interesting candidates for further study. The RAS P21 protein activator (GTPase activating protein) 1 (Rasa1) gene located towards the proximal end of the interval (85.2 Mb) was recently identified to be strongly antiviral in the context of CVB3 infection of polarized human brain microvascular endothelial cells.41 Coagulation factor II (thrombin) receptor (F2r or PAR1) located at 95.6 Mb has been shown to control CVB3 infection in the heart and liver at late (but not early) time points, and may be involved in cooperative signaling with TLR3.42 3-hydroxy-3-methylglutaryl-coenzyme A reductase (Hmgcr 96.6 Mb), the enzymatic target of statin class of pharmaceuticals, performs a rate limiting step in cholesterol biosynthesis. The link between CVB3 replication and cellular cholesterol availability was recently established in a study that demonstrated CVB3 causes cholesterol to be trafficked from the plasma membrane and the specialized organelles for viral production.43 Phosphatidylinositol 3-kinase, regulatory subunit, polypeptide 1 (p85 alpha) (Pi3kr1) located at 101 Mb is a regulatory subunit of PI3K. It has been shown in HeLa cells that PI3K signaling is beneficial for CVB3 replication, presumably by providing a survival signal (possibly via NFkB) to counter apoptotic signals.44,45 A gene cluster of NLR family, apoptosis inhibitory proteins (Naip1-Naip7, 100 Mb) have been reported to bind bacterial components and lead to cellular pyroptosis.46,47 Distally, the IL-6 receptor signal transducer (Il6st, or gp130) is located at 112 Mb. Gp130 signalling does not appear to be directly antiviral, but provides a strong cytoprotective effect in the context of CVB3 infection. Furthermore, cardiac specific ablation of gp130 signaling by SOCS3 overexpression leads to greatly increase cardiac necrosis, and lethality between 4 and 6 days after infection.48 Mutations in one or more of these genes or others with no known role in CVB3 infection are possible, and further studies will include exome sequencing and candidate gene analysis. MATERIALS AND METHODS Mice, cells and virus Inbred B6, B10, and A/J mice were purchased from the Jackson Laboratory (Bar Harbor, ME, USA). The set of 36 AcB/BcA RC strains, derived from A/J and B6 parents, were generated according to a previously described breeding scheme and genotyping protocol.39 (BcA86 X B10) and (B10 X BcA86) F1 crosses were generated; F2 crosses were then performed by Genes and Immunity (2015) 261 – 267
265
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
266 standard (F1brother X F1sister) mating. The mice were maintained in the McGill University animal facility in compliance with the Canada Council on Animal Care as approved by the McGill University Animal Care Committee. CVB3 used for all experiments was CVB3-CG strain as previously.49 HeLa cells (ATCC: CCL-2) were grown and maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% FBS, 100 μg/ml Penicillin/ Streptomycin (1 × media). Homogenized organs were serially diluted 10-fold in non-supplemented DMEM and the dilutions were used to infect HeLa cells in triplicate for 60 min. After initial infection by homogenate, HeLa cells are covered in a layer of media with 0.25% agarose and incubated for three days. Cells were fixed in formaldehyde and stained with crystal violet to count plaques as previously.18
Mouse infection and phenotype determination For each experiment, a vial of viral stock was thawed and diluted in cold PBS at 5 × 104 PFU/ml. Inbred, F1 and F2 mice were inoculated intraperitoneally with 400 PFU of CVB3-CG per gram of body weight at 7–9 weeks of age. Animals were weighed daily and observed for changes in fur characteristics, level of activity and appearance of respiratory distress. During the RC strain survival screen, moribund mice were humanely euthanized at an acceptable clinical end point. A total of 70 BcA mice and 30 AcB mice, and 210 [BcA86xC67BL/10]F2 mice were phenotyped. For subsequent experiments, mice were euthanized at day 2 post infection, and exsanguinated by cardiac puncture to serum. Heart and liver tissues were removed aseptically, snap frozen and stored at − 80 °C in two separate aliquots. One portion was subsequently homogenized and submitted to three freeze-thaw cycles to release infectious virus for determination of viral titre by plaque assay. Total RNA was extracted from the other portion by TRIzol (Ambion, Burlington, ON, Canada), and reverse transcribed by MMLV (New England Biolabs, Ipswich, MA, USA) by using random hexamers according to the manufacturers’ instructions. Serum levels of creatine kinase and ALT were measured by the McGill Comparative Medicine Animal Resources Centre's Diagnostic and Research Support Service by using a colorimetric method. Quantitative PCR primers were designed to span exon junctions with the help of primer3plus.50 The primer sequences are as follows: CVB3-H3 forward: ATGGCAGAAAACCTCACCAG; CVB3-H3 reverse: ATGTTCTGCCCAAACAGTCC; Ifnb1 forward: TGACGGAGAAGATGCAGAAG; Ifnb1 reverse: ACCCAGTGC TGGAGAAATTG.
Genotype analysis The B10 strain is not a Mouse Phenome Project priority strain (http:// phenome.jax.org/), and so existing publically available genotype information is limited. Most single-nucleotides polymorphic between B10 and B6 for use in the generation of a panel capable of distinguishing between B10 and BcA86 were identified by using the JAX Diversity Array (Jackson Laboratory).51 In this panel, 7499 single-nucleotides were polymorphic between B6 and B10, though these tended to be clustered and gaps remained (Supplementary Figure 1). These gaps in the genetic map were filled in by comparing the genotype of B6 against all other C57BL strains available in all datasets contained within the Mouse Phenome Database. When the B6 genotype was divergent, B6 and B10 were genotyped by Sanger sequencing to confirm the private B6 single-nucleotide polypeptide. Finally, the map was completed with microsatellites identified by us as previously,19 and in collaboration with others26 for a total of 106 markers that are informative between BcA86 and B10. All single-nucleotide polymorphism genotyping was performed by Sequenom iPlex Gold technology. Genotyping of the F2 was performed at the McGill University and Genome Quebec Innovation Centre (Montreal, QC, Canada). Microsatellite PCR reactions performed manually by using a 10-μl total reaction volume, 200-μM dNTPs, 1.5- mM MgCl2, 2 pmol of each primer and 0.5 U of Taq polymerase (Invitrogen, Burlinton, ON, USA). Reactions were performed as follows: 96˚ for 2 min; 30 cycles of 94˚ for 45 s, 56˚ for 45 s, 72˚ for 60 s; and a final extension step 72˚ 7 min. PCR products were then separated on 2–4% high-resolution agarose (USB, Cleveland, OH, USA) gels containing Ethidium bromide and visualized under UV light.
Statistical analysis Statistical analyses were conducted with the freely available program R (www.r-project.org) and all packages were installed using the 'install. packages' function in R. Linkage was performed with the package ‘R/qtl’ version 3.02. Survival analysis was performed using the ‘Survival’ package Genes and Immunity (2015) 261 – 267
version 2.37. Skewness was detected using the D’Agostino test from the ‘moments’ package version 0.13. The normality of phenotypic distributions was assessed using the Shapiro Test from the ‘stats’ package version 2.15.3. Serum ALT, liver and heart viral titre F2 phenotypes were normalized using Box–Cox52 method as implemented in the ‘Caret’ package version 5.17. Briefly, for every value, x, in a phenotypic distribution, the Box–Cox method λ transforms x according to the formula x 0λ ¼ x λ- 1, where λ is optimized such 0 that the distribution of x λ best approximates the normal distribution. The determined value of λ was 0.3 for ALT and liver viral titer and 0.2 for heart viral titer. QQPLOTs were generated using the qqnorm function in the ‘stats’ package version 2.15.3. Viral replication is also reported as Log10[Pfu/ mg] in figures for the sake of interpretability. The scanone function of the ‘R/qtl’ package version 1.28-19, was used to perform Haley–Knott regression of the phenotype on genetic markers. Significance values were evaluated with 10 000 permutations. The lod support interval of quantitative trait loci peaks was calculated using a 1.5 lod drop by ‘R/qtl’ on the interval map. P-values shown in the figures are a result of two-tailed t-tests with *Po0.05, **P o0.01 and ***Po0.001 unless otherwise indicated.
CONFLICT OF INTEREST The authors declare no conflict of interest.
ACKNOWLEDGEMENTS This work was supported by grants from the Canadian Institutes of Health Research (CIHR) SAW, GAL and JM were supported by FRSQ Scholarships and SMV by the Canada Research Chair program.
REFERENCES 1 Melnick JL, Godman GC. Pathogenesis of coxsackie virus infection; multiplication of virus and evolution of the muscle lesion in mice. J Exp Med 1951; 93: 247–266. 2 Gkrania-Klotsas E, Langenberg C, Tauriainen S, Sharp SJ, Luben R, Forouhi NG et al. The association between prior infection with five serotypes of Coxsackievirus B and incident type 2 diabetes mellitus in the EPIC-Norfolk study. Diabetologia 2012; 55: 967–970. 3 Cheng LL, Ng PC, Chan PK, Wong HL, Cheng FW, Tang JW. Probable intrafamilial transmission of coxsackievirus b3 with vertical transmission, severe early-onset neonatal hepatitis, and prolonged viral RNA shedding. Pediatrics 2006; 118: e929–e933. 4 Tseng F-C, Huang H-C, Chi C-Y, Lin T-L, Liu C-C, Jian J-W et al. Epidemiological survey of enterovirus infections occurring in Taiwan between 2000 and 2005: analysis of sentinel physician surveillance data. J Med Virol 2007; 79: 1850–1860. 5 Wang SM, Liu CC, Yang YJ, Yang HB, Lin CH, Wang JR. Fatal coxsackievirus B infection in early infancy characterized by fulminant hepatitis. J Infect 1998; 37: 270–273. 6 Yajima T, Knowlton KU. Viral myocarditis: from the perspective of the virus. Circulation 2009; 119: 2615–2624. 7 Gauntt C, Huber S. Coxsackievirus experimental heart diseases. Front Biosci 2003; 8: e23–e35. 8 Wolfgram L, Beisel K, Herskowitz A, Rose NR. Variations in the susceptibility to Coxsackievirus B3-induced myocarditis among different strains of mice. J Immunol 1986; 136: 1846–1852. 9 Gauntt C, Gomez P, Duffey P, Grant J, Trent D, Witherspoon S et al. Characterization and myocarditic capabilities of coxsackievirus B3 variants in selected mouse strains. J Virol 1984; 52: 598–605. 10 Wessely R, Klingel K, Knowlton KU, Kandolf R. Cardioselective infection with coxsackievirus B3 requires intact type I interferon signaling: implications for mortality and early viral replication. Circulation 2001; 103: 756–761. 11 Deonarain R, Cerullo D, Fuse K, Liu P, Fish E. Protective role for interferon-beta in coxsackievirus B3 infection. Circulation 2004; 110: 3540–3543. 12 Yasukawa H, Yajima T, Duplain H, Iwatate M, Kido M, Hoshijima M et al. The suppressor of cytokine signaling-1 (SOCS1) is a novel therapeutic target for enterovirus-induced cardiac injury. J Clin Invest 2003; 111: 469–478. 13 Richer MJ, Lavallee DJ, Shanina I, Horwitz MS. Toll-like receptor 3 signaling on macrophages is required for survival following coxsackievirus B4 infection. PLoS One 2009; 4: e4127. 14 Wang JP, Cerny A, Asher DR, Kurt-Jones EA, Bronson RT, MDA5 Finberg RW. and MAVS mediate type I interferon responses to coxsackie B virus. J Virol 2010; 84: 254–260.
© 2015 Macmillan Publishers Limited
Mapping a locus controlling susceptibility to CVB3 viral hepatitis SA Wiltshire et al
267 15 Huber SA, Roberts B, Moussawi M, Boyson JE. Slam haplotype 2 promotes NKT but suppresses Vgamma4+ T-cell activation in coxsackievirus B3 infection leading to increased liver damage but reduced myocarditis. Am J Pathol 2013; 182: 401–409. 16 Huhn MH, McCartney SA, Lind K, Svedin E, Colonna M, Flodstrom-Tullberg M. Melanoma differentiation-associated protein-5 (MDA-5) limits early viral replication but is not essential for the induction of type 1 interferons after Coxsackievirus infection. Virology 2010; 401: 42–48. 17 Chow LH, Gauntt CJ, McManus BM. Differential effects of myocarditic variants of Coxsackievirus B3 in inbred mice. A pathologic characterization of heart tissue damage. Lab Invest 1991; 64: 55–64. 18 Wiltshire SA, Leiva-Torres GA, Vidal SM. Quantitative trait locus analysis, pathway analysis, and consomic mapping show genetic variants of Tnni3k, Fpgt, or H28 control susceptibility to viral myocarditis. J Immunol 2011; 186: 6398–6405. 19 Aly M, Wiltshire SA, Chahrour G, Osti J-CL, Vidal SM. Complex genetic control of host susceptibility to coxsackievirus B3-induced myocarditis. Genes Immun 2007; 8: 193–204. 20 Herskowitz A, Wolfgram L, Rose NR, Beisel K. Coxsackievirus B3 murine myocarditis: a pathologic spectrum of myocarditis in genetically defined inbred strains. J Am Coll Cardiol 1987; 9: 1311–1319. 21 Wiltshire SA, Diez E, Miao Q, Dubé M-P, Gagné M, Paquette O et al. Genetic control of high density lipoprotein-cholesterol in AcB/BcA recombinant congenic strains of mice. Physiol Genomics 2012; 44: 843–852. 22 Min-Oo G, Fortin A, Tam M, Nantel A, Stevenson M, Gros P. Pyruvate kinase deficiency in mice protects against malaria. Nat Genet 2003; 35: 357–362. 23 Chow L, Gauntt C, McManus B. Differential effects of myocarditic variants of Coxsackievirus B3 in inbred mice. A pathologic characterization of heart tissue damage. Lab Invest 1991; 64: 55–64. 24 Wessely R, Klingel K, Knowlton K, Kandolf R. Cardioselective infection with coxsackievirus B3 requires intact type I interferon signaling: implications for mortality and early viral replication. Circulation 2001; 103: 756–761. 25 Yang RQ, Yi NJ, Xu SZ. Box-Cox transformation for QTL mapping. Genetica 2006; 128: 133–143. 26 Xia Y, Won S, Du X, Lin P, Ross C, La Vine D et al. Bulk segregation mapping of mutations in closely related strains of mice. Genetics 2010; 186: 1139–1146. 27 Bongfen SE, Rodrigue-Gervais I-G, Berghout J, Torre S, Cingolani P, Wiltshire SA et al. An N-Ethyl-N-Nitrosourea (ENU)-Induced Dominant Negative Mutation in the JAK3 Kinase Protects against Cerebral Malaria. PloS One 2012; 7: e31012. 28 Khetsuriani N, Lamonte-Fowlkes A, Oberst S, Pallansch M. Enterovirus surveillance--United States, 1970-2005. MMWR Surveill Summ 2006; 55: 1–20. 29 CfDCaP. CDC. Nonpolio enterovirus and human parechovirus surveillance— United States, 2006–2008. MMWR Morb Mortal Wkly Rep 2010; 59: 1577–1580. 30 Traystman MD, Chow LH, McManus BM, Herskowitz A, Nesbitt MN, Beisel KW. Susceptibility to Coxsackievirus B3-induced chronic myocarditis maps near the murine Tcr alpha and Myhc alpha loci on chromosome 14. Am J Pathol 1991; 138: 721–726. 31 Knowlton KU, Jeon ES, Berkley N, Wessely R, Huber S. A mutation in the puff region of VP2 attenuates the myocarditic phenotype of an infectious cDNA of the Woodruff variant of coxsackievirus B3. J Virol 1996; 70: 7811–7818. 32 Stadnick E, Dan M, Sadeghi A, Chantler JK. Attenuating mutations in coxsackievirus B3 map to a conformational epitope that comprises the puff region of VP2 and the knob of VP3. J Virol 2004; 78: 13987–14002. 33 Schmidtke M, Merkle I, Klingel K, Hammerschmidt E, Zautner AE, Wutzler P. The viral genetic background determines the outcome of coxsackievirus B3 infection in outbred NMRI mice. J Med Virol 2007; 79: 1334–1342.
34 Gay RT, Belisle S, Beck MA, Meydani SN. An aged host promotes the evolution of avirulent coxsackievirus into a virulent strain. Proc Natl Acad Sci USA 2006; 103: 13825–13830. 35 Frisancho-Kiss S, Davis SE, Nyland JF, Frisancho JA, Cihakova D, Barrett MA et al. Cutting edge: cross-regulation by TLR4 and T cell Ig mucin-3 determines sex differences in inflammatory heart disease. J Immunol 2007; 178: 6710–6714. 36 Frisancho-Kiss S, Nyland JF, Davis SE, Frisancho JA, Barrett MA, Rose NR et al. Sex differences in coxsackievirus B3-induced myocarditis: IL-12Rbeta1 signaling and IFN-gamma increase inflammation in males independent from STAT4. Brain Res 2006; 1126: 139–147. 37 Woodruff JF. The influence of quantitated post-weaning undernutrition on coxsackievirus B3 infection of adult mice. II. Alteration of host defense mechanisms. J Infect Dis 1970; 121: 164–181. 38 Woodruff JF, Woodruff JJ. Modification of severe coxsackievirus B 3 infection in marasmic mice by transfer of immune lymphoid cells. Proc Natl Acad Sci USA 1971; 68: 2108–2111. 39 Fortin A, Diez E, Rochefort D, Laroche L, Malo D, Rouleau G et al. Recombinant congenic strains derived from A/J and C57BL/6 J: a tool for genetic dissection of complex traits. Genomics 2001; 74: 21–35. 40 Boivin GA, Pothlichet J, Skamene E, Brown EG, Loredo-Osti JC, Sladek R et al. Mapping of clinical and expression quantitative trait loci in a sex-dependent effect of host susceptibility to mouse-adapted influenza H3N2/HK/1/68. J Immunol 2012; 188: 3949–3960. 41 Coyne CB, Bozym R, Morosky SA, Hanna SL, Mukherjee A, Tudor M et al. Comparative RNAi screening reveals host factors involved in enterovirus infection of polarized endothelial monolayers. Cell Host Microbe 2011; 9: 70–82. 42 Antoniak S, Owens AP 3rd, Baunacke M, Williams JC, Lee RD, Weithauser A et al. PAR-1 contributes to the innate immune response during viral infection. J Clin Investig 2013; 123: 1310–1322. 43 Ilnytska O, Santiana M, Hsu NY, Du WL, Chen YH, Viktorova EG et al. Enteroviruses harness the cellular endocytic machinery to remodel the host cell cholesterol landscape for effective viral replication. Cell Host Microbe 2013; 14: 281–293. 44 Esfandiarei M, Boroomand S, Suarez A, Si X, Rahmani M, McManus B. Coxsackievirus B3 activates nuclear factor kappa B transcription factor via a phosphatidylinositol-3 kinase/protein kinase B-dependent pathway to improve host cell viability. Cell Microbiol 2007; 9: 2358–2371. 45 Esfandiarei M, Luo H, Yanagawa B, Suarez A, Dabiri D, Zhang J et al. Protein kinase B/Akt regulates coxsackievirus B3 replication through a mechanism which is not caspase dependent. J Virol 2004; 78: 4289–4298. 46 Diez E, Lee SH, Gauthier S, Yaraghi Z, Tremblay M, Vidal S et al. Birc1e is the gene within the Lgn1 locus associated with resistance to Legionella pneumophila. Nat Genet 2003; 33: 55–60. 47 Bortoluci KR, Medzhitov R. Control of infection by pyroptosis and autophagy: role of TLR and NLR. Cell Mol Life Sci 2010; 67: 1643–1651. 48 Suppiah V, Moldovan M, Ahlenstiel G, Berg T, Weltman M, Abate ML et al. IL28B is associated with response to chronic hepatitis C interferon-alpha and ribavirin therapy. Nat Genet 2009; 41: 1100–1104. 49 Aly M, Wiltshire S, Chahrour G, Osti JC, Vidal SM. Complex genetic control of host susceptibility to coxsackievirus B3-induced myocarditis. Genes Immun 2007; 8: 193–204. 50 Rozen S, Skaletsky H. Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol 2000; 132: 365–386. 51 Yang H, Ding Y, Hutchins L, Szatkiewicz J, Bell T, Paigen B et al. A customized and versatile high-density genotyping array for the mouse. Nat Methods 2009; 6: 663–666. 52 Box GCD. An Analysis of Transformations. J R Stat Soc Ser B 1964; 16: 211–252.
Supplementary Information accompanies this paper on Genes and Immunity website (http://www.nature.com/gene)
© 2015 Macmillan Publishers Limited
Genes and Immunity (2015) 261 – 267
Copyright of Genes & Immunity is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
