* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Virulence correlates with fitness in vivo for two M group genotypes of
Survey
Document related concepts
Oesophagostomum wikipedia , lookup
Cross-species transmission wikipedia , lookup
Herpes simplex wikipedia , lookup
Hospital-acquired infection wikipedia , lookup
2015–16 Zika virus epidemic wikipedia , lookup
Neonatal infection wikipedia , lookup
Ebola virus disease wikipedia , lookup
Middle East respiratory syndrome wikipedia , lookup
West Nile fever wikipedia , lookup
Human cytomegalovirus wikipedia , lookup
Hepatitis C wikipedia , lookup
Influenza A virus wikipedia , lookup
Orthohantavirus wikipedia , lookup
Marburg virus disease wikipedia , lookup
Lymphocytic choriomeningitis wikipedia , lookup
Henipavirus wikipedia , lookup
Transcript
ARTICLE IN PRESS YVIRO-05727; No. of pages: 8; 4C: Virology xxx (2010) xxx–xxx Contents lists available at ScienceDirect Virology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y v i r o Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV) Andrew R. Wargo a,c,⁎, Kyle A. Garver a,b, Gael Kurath a,c a b c Western Fisheries Research Center, 6505 NE 65th Street, Seattle, WA 98115-5016, USA Pacific Biological Station, 3190 Hammond Bay Road, Nanaimo, B.C., Canada V9T 6N7 University of Washington, Seattle, WA, USA a r t i c l e i n f o Article history: Received 23 February 2010 Returned to author for revision 17 March 2010 Accepted 22 April 2010 Available online xxxx Keywords: Virulence Fitness Evolution Ecology In vivo Vertebrate Virus a b s t r a c t The nature of the association between viral fitness and virulence remains elusive in vertebrate virus systems, partly due to a lack of in vivo experiments using statistically sufficient numbers of replicate hosts. We examined the relationship between virulence and fitness in Infectious hematopoietic necrosis virus (IHNV), in vivo, in intact living rainbow trout. Trout were infected with a high or low virulence genotype of M genogroup IHNV, or a mixture of the two genotypes, so as to calculate relative fitness and the effect of a competition environment on fitness. Fitness was measured as total viral load in the host at time of peak viral density, quantified by genotype-specific quantitative RT-PCR (qRT-PCR). The more virulent IHNV genotype reached higher densities in both single and mixed infections. There was no effect of competition on the performance of either genotype. Our results suggest a positive link between IHNV genotype fitness and virulence. © 2010 Elsevier Inc. All rights reserved. Introduction Viral fitness, here defined as the relative ability of viruses to produce infectious progeny in a given environment, plays a critical role in viral evolution (Domingo and Holland, 1997; Domingo et al., 1997; Holland et al., 1991). The overall selection of viral strains within and between hosts is likely to depend heavily on their relative fitness (Novella et al., 1995; Quinones-Mateu et al., 2000; Troyer et al., 2005). It is also theorized that an association between fitness and viral traits, such as virulence, might exist (Alizon et al., 2009; Ebert, 1998; Troyer et al., 2008). Understanding the nature of these associations could provide valuable insights into how the traits themselves evolve (Brown et al., 2006; Bull, 1994; Frank, 1996; Levin, 1996; Read, 1994). Few in vivo studies have been conducted in vertebrate systems examining the relationship between viral fitness and in particular, the virulence trait, and therefore the nature of this association remains elusive. Here we developed and carried out a methodology to quantify viral fitness in vivo, in a vertebrate virus system, and shed light on the association between fitness and virulence. We defined virulence as mortality caused to the host. The bulk of research examining the relationship between virulence ⁎ Corresponding author. Western Fisheries Research Center, 6505 NE 65th Street, Seattle, WA 98115-5016, USA. E-mail address: [email protected] (A.R. Wargo). and viral fitness has relied on bacteriophage, plant host systems, or vertebrate viruses in cell culture. Although crucial for establishing the foundation of virulence evolution theory, the findings from these studies often differed from each other. In bacteriophage a positive association between virulence and viral density was found under certain modes of transmission (Messenger et al., 1999). Whereas, in plant viruses, research suggested that virulence was positively correlated with the infectivity but not the density of barley stripe mosaic virus (Stewart et al., 2005), and work on cucumber mosaic virus revealed genotypes with intermediate levels of virulence had highest transmissibility (Escriu et al., 2003). Furthermore, in vertebrate viruses, in vitro studies on FMDV, poliovirus, and SIV indicated that in some cases more virulent genotypes had the fitness advantage and in others they did not, depending on the viral genotype or growth conditions examined (Chumakov et al., 1991; Herrera et al., 2007; Voronin et al., 2005). Ultimately, since the host immune response is likely to play an important role in virulence evolution (Alizon et al., 2009; Frank and Schmid-Hempel, 2007), the relevance of in vitro studies to intact vertebrate systems is unknown. To more closely reflect the vertebrate host, an ex vivo system was developed for HIV using human peripheral blood mononuclear cells. This system, in conjunction with epidemiological studies, revealed that viral fitness and transmission potential are important factors for the success and spread of different HIV genogroups (Arien et al., 2005) and may be positively correlated with 0042-6822/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.virol.2010.04.023 Please cite this article as: Wargo, A.R., et al., Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV), Virology (2010), doi:10.1016/j.virol.2010.04.023 ARTICLE IN PRESS 2 A.R. Wargo et al. / Virology xxx (2010) xxx–xxx disease progression in the field (Frasher et al., 2007; Quinones-Mateu et al., 2000; Sagar et al., 2003; Troyer et al., 2005). Few truly in vivo empirical studies examining the association between vertebrate viral virulence and fitness, in intact hosts, are available; likely due to the difficulty of conducting research in such systems. Of the studies conducted, many were not specifically aimed at testing the virulence–fitness association and, similar to in vitro work, produced incongruous results. An examination of foot-andmouth disease in swine and influenza virus in chickens both suggested more virulent genotypes had a replication advantage (Carrillo et al., 1998; Suzuki et al., 2009). Alternatively, investigations of SIV in primates indicated that virulence was not associated with viral replication (Pandrea et al., 2008) and work on hepatitis B in ducks alluded that less virulent strains had the fitness advantage (Lenhoff et al., 1998). A limitation to these in vivo studies was the small number of infected hosts examined; typically less than 10 animals. Natural viral infections are likely to show high levels of variation in viral dynamics, and indeed this phenomenon was observed in the above mentioned, as well as other in vivo studies (Burke et al., 2006; Munster et al., 2009; Troyer et al., 2008). As such, conclusions about population level processes are likely to be more accurate when greater host numbers are analyzed. Recently, an in vivo fitness assay was developed for Infectious hematopoietic necrosis virus (IHNV) in intact living rainbow trout (Oncorhynchus mykiss) hosts. Using this assay it was found that two genetically distinct IHNV isolates (referred to as genotypes from here forward) with equal virulence had equal fitness (Troyer et al., 2008). To more fully elucidate the nature of the virulence relationship in IHNV, we expanded this previous study by comparing the relative fitness of two genotypes which differ in virulence, while also examining the fitness parameter in more detail. IHNV is a single-stranded, negative-sense RNA rhabdovirus, endemic in salmonid fishes throughout western North America (Bootland and Leong, 1999; Kurath et al., 2003). The virus typically causes acute infection, with an often rapid onset of disease and mortality (Bootland and Leong, 1999). Hundreds of isolates of the virus have been obtained from the field and characterized, showing wide ranging levels of virulence, sometimes reaching close to 100% mortality of the host fish population (Garver et al., 2006; LaPatra et al., 1994). Field studies indicate that multiple genotypes of IHNV can cocirculate and that newly emerged genotypes can displace resident ones, suggesting possible fitness differences between genotypes (Anderson, 2000; Troyer and Kurath, 2003; Troyer et al., 2000). Phylogenic analyses of field isolates have revealed three major genogroups of IHNV in North America, designated U, M, and L (Kurath et al., 2003). Among them the M genogroup is hypothesized to have arisen in rainbow trout (Troyer and Kurath, 2003; Troyer et al., 2000). We examined the fitness of a high and low virulence IHNV genotype, both from the M genogroup, when in the fish host alone in a single infection environment, and in the presence of the other genotype in a mixed infection competition environment. This experimental design enabled us to determine relative fitness and the impact competition had on a genotype's fitness (de Roode et al., 2005). The comparison of single versus mixed infections was not done in our previous study (Troyer et al., 2008) and to our knowledge, has not been explored when examining the virulence–fitness association in vertebrate virus systems. To improve the quantification of viral fitness, we developed genotype-specific real-time quantitative PCR (qPCR) assays, making it possible to measure the absolute viral load of each genotype in both single and mixed infections, in individual fish. To fully capture the high levels of anticipated fish-to-fish variation, 23–28 fish per treatment group were utilized, providing robust fitness estimates. Our experiments confirmed the virulence difference between the two M group IHNV genotypes (Garver et al., 2006; Troyer et al., 2000) and revealed that the more virulent genotype reached higher den- sities in the within-host virus populations in both single and mixed infections, suggesting it had a fitness advantage. Both genotypes performed equally well in single infections as they did in mixed infections, implying that there was no direct competition for a limited resource in the fish host. Our results clearly demonstrated a positive association between virulence and viral fitness as defined here. Results Virulence To demonstrate the relative virulence of viral genotypes HV and LV, we compared the mortality induced in triplicate batch challenges of 20 fish infected with genotype HV, to those infected with genotype LV. The replicate tanks all showed the same characteristic IHNV acute infection dynamic (Fig. 1). Mortality began between days 5 and 7 for HV-infected fish and 7 and 9 for LV-infected, continued through day 17, and quickly tapered off, with the last LV-infected fish dying on day 24 and the last HV-infected on day 26. Over the 28 day monitoring period, fish infected with genotype HV had an average median survival of 11 ± 1 days, which was significantly shorter than the 23 ± 1 day median survival of fish infected with genotype LV (Fig. 1; Logrank = 40.35, d.f. = 1, P b 0.001). Fish infected with genotype HV also experienced a significantly higher cumulative percentage of mortality (CPM, 85 ± 5%) than the fish infected with genotype LV (30.5 ± 3%), over the monitoring period (Fig. 1; X21 N 25, P b 0.0001). Thus, these results are consistent with previous observations that HV is the more virulent of the two IHNV genotypes, as defined by infection-induced mortality (Garver et al., 2006). When examining only the fish that died, there was no difference in the mean day to death between HVinfected fish (12 ± 1 days) compared to LV-infected fish (13 ± 1 days), suggesting that the kinetics of disease progression was approximately the same for both virus genotypes. Viral fitness To determine the relative fitness of genotypes HV and LV, the total viral load of each genotype was compared in fish infected with either HV alone, LV alone, or a mixture of the two, in three independent experiments. This made it possible to determine the relative fitness of each genotype alone versus in a competition environment and examine the direct effect of competition on the two genotypes. Across all three fitness experiments there was a substantial degree of Fig. 1. Experimental batch immersion challenge defining virulence difference between IHNV genotypes HV (high virulence, black line) and LV (low virulence, gray line) in triplicate groups of 20 juvenile rainbow trout. Lines are daily mean cumulative percent mortality (± 1 S.E.M) over 30 days of infection. Mock exposed fish showed no mortality (not shown). Please cite this article as: Wargo, A.R., et al., Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV), Virology (2010), doi:10.1016/j.virol.2010.04.023 ARTICLE IN PRESS A.R. Wargo et al. / Virology xxx (2010) xxx–xxx 3 fish-to-fish variation in viral load within treatment groups, such that viral load ranged well over 4 orders of magnitude between the fish (Fig. 2, S2, S3). Despite this level of variation, significant differences were observed between the treatment groups. When comparing mean viral loads for all fish in an experiment, genotype HV produced more viral copies than LV in both single and mixed infections, across all experiments (Fig. 3, genotype effect: F1,233 = 60, P b 0.001). On average the viral production of genotype HV was 1–2 orders of magnitude higher than that for genotype LV (Fig. 3). Ultimately, viral genotype HV had greater fitness than LV to such a degree that overall, approximately 76% of the total viral population produced in the mixed infection groups was HV (Fig. 4A), thus, HV maintained its fitness advantage in both single and mixed infections. Despite the general dominance of HV, the less virulent genotype was not completely Fig. 3. Mean viral loads (± 1 S.E.M) of HV (black bars) or LV (gray bars) determined by qPCR in virus-infected treatment groups from three independent viral fitness experiments (A–C). Fig. 2. Viral loads of genotypes HV (black bars) and LV (gray bars) determined by qRTPCR in individual fish from experiment 1, three days after exposure to A) HV alone, B) LV alone, C) a 1:1 mixture of HV:LV. Analogous data from experiments 2 and 3 can be found in the Supplemental materials. excluded and there were 14 out of 81 individual hosts where LV was the majority genotype in fish receiving a mixed infection (Fig. 4B). When examining the impact of competition on genotype performance, there was no significant difference in the viral copy output of either genotype in single versus mixed infections, despite a suggestive trend of a small decrease during competition for HV in experiments 2 and 3 (Fig. 3, competition effect: F1,232 = 2.3, P = 0.13). Furthermore, the total virus produced in mixed infections (HV + LV) was not significantly different than that produced in single infections of HV. The three replicate experiments produced very similar results, although total viral production decreased with subsequent experiments such that the most virus was produced in experiment 1 and the least in experiment 3 (Fig. 3, experiment effect: F2,233 = 18, P b 0.001). The total drop in viral load between experiment 1 and 3 was less than Please cite this article as: Wargo, A.R., et al., Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV), Virology (2010), doi:10.1016/j.virol.2010.04.023 ARTICLE IN PRESS 4 A.R. Wargo et al. / Virology xxx (2010) xxx–xxx Fig. 4. IHNV genotype dominance in fish exposed to a mixed infection. (A)The mean percent (± 1 S.E.M) of total virus produced that was HV across all fish receiving a mixed infection for the three experiments. (B)The number of individual fish where HV (black bars) or LV (gray bars) was dominant in a mixed infection for the three experiments. an order of magnitude (Fig. 3). In summary, our results consistently suggest that HV had generally higher fitness than LV. Discussion The association between virulence and viral fitness is a heavily debated topic (Alizon et al., 2009; Brown et al., 2006; Bull, 1994; Ebert, 1998; Levin, 1996; Read, 1994). In vertebrate virus systems, the ambiguity in this subject is partly due to a lack of convincing data, particularly from in vivo experiments, with statistically sufficient numbers of replicate hosts. Here we have shown a positive relationship between virulence and the fitness of the salmonid virus IHNV in its natural host, rainbow trout, using in vivo experiments. The more virulent IHNV genotype (HV) consistently produced greater numbers of infectious progeny than the less virulent genotype (LV) in both single and mixed infections in trout. Typically, experiments examining viral fitness have focused on coinfection to estimate relative fitness differences between genotypes (reviewed in Troyer et al., 2008). We added a further aspect by comparing the performance of viral genotypes alone as well as in coinfection. This made it possible to directly assess how the performance of each genotype was influenced by the competition environment and what impact this had on overall fitness (de Roode et al., 2005). Interestingly, both genotypes performed equally well in co-infection as they did alone. These findings suggest that the higher viral production of HV compared to LV is largely due to differences in viral replication rather than exploitation competition governed by limited resources or other similar mechanisms (Brudeseth et al., 2002; Read and Taylor, 2001). This is despite intentionally observing viral dynamics on day 3 at the peak of within-host replication, when resources available to the virus are expected to be most strained (Troyer et al., 2008). Given these results, one would expect that the total virus produced in a mixed infection would be slightly greater than that of HV single infections, due to the added presence of LV. However, the total virus in mixed infections (HV + LV) was not found to significantly differ from that produced in single infections of HV. This was likely due to the very small contribution of genotype LV to the total virus in mixed infections, combined with the slight, although non-significant, drop in viral production for genotype HV in mixed infections in experiments 2 and 3. Thus, there also appeared to be no facilitation response by either genotype to the presence of the other, such as increased replication, as postulated in other systems (Hodgson et al., 2004) and predicted by some theoretical models (van Baalen and Sabelis, 1995). We cannot rule out the possibility that a more pronounced competition dynamic may appear for IHNV later in the infection, after the host immune response develops, if immune mediated apparent competition occurs in the IHNV-trout system, such that the immune response to one genotype has a negative effect on the other (LaPatra et al., 1994; Read and Taylor, 2001). Fish are known to develop both innate and adaptive immunity to IHNV, however, previous studies indicate that viral load dynamics are determined very early in the infection process, suggesting adaptive immunity may play a minor role in shaping the observed virulence differences between IHNV genotypes (Lorenzen and LaPatra, 1999; Peñaranda et al., 2009; Purcell et al., 2009c). It is plausible that the competition dynamic in mixed IHNV genotype infections of trout ultimately depends on whether or not the genotypes co-infect the same host cell. A recent study of interference competition in vitro, in foot-and-mouth disease virus (FMDV), revealed that a less virulent mutant had the advantage in high-MOI conditions, where the probability of cellular co-infection was high; whereas the high virulence mutant had the advantage under low MOI conditions, where the probability of cellular co-infection was low (Ojosnegros et al., 2010). If cellular co-infection does not occur for IHNV, this could explain why genotype LV is out competed by HV, given the findings of Ojosnegros et al. It may be that the IHNV viral load of in vivo infections never reached a level where cells became limited, and thus low MOI conditions were maintained such that competition for cells and the likelihood of cellular co-infection were rare. In general, IHNV is a relatively low titer virus, reaching titers of only 102–107 infectious virus particles per gram of tissue during in vivo infections, but the MOI during in vivo infections is not known. Some evidence of a block to cellular co-infection in IHNV was observed in a study that found cells bound with one species of rhabdovirus had reduced affinity for another species (de las Heras et Please cite this article as: Wargo, A.R., et al., Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV), Virology (2010), doi:10.1016/j.virol.2010.04.023 ARTICLE IN PRESS A.R. Wargo et al. / Virology xxx (2010) xxx–xxx al., 2008), however, co-infection with different genotypes of the same virus species was not examined. A further reduction in the probability of IHNV cellular co-infection could occur if genotypes compartmentalize in different tissues as they spread throughout the host. The course of IHNV infection begins with viral entry via gills, esophagus, or skin, followed by systemic infection, with most virus replication focused in the hematopoietic tissue of the kidney and spleen (Drolet et al., 1994; Harmache et al., 2006; Helmick et al., 1995). Evidence for compartmentalization during this process is lacking. Ultimately, the absence or presence of cellular co-infection is beyond the scope of the work presented here and does not detract from our observation that HV consistently produced more viral copies than LV in both single and mixed infections. It should be noted that in this study our goal was to use genotypespecific qPCR to quantify the dominant consensus component of the HV and LV genotypes in single and mixed infection populations. As a negative-sense RNA virus, IHNV populations exist as quasi-species mixtures of variants, and it is possible that some mutations in the qPCR primer or probe sequence sites could have occurred during these experiments, resulting in a small number of genomes that were not detected. However, the region chosen for quantification showed no sequence variation in 3 clones for both genotypes HV and LV (data not shown), and previous studies indicate that the mutant spectra within natural quasi-species of IHNV is low for an RNA virus (Emmenegger et al., 2003). In addition, given that they are singlestranded, negative-sense RNA viruses, recombination is extremely rare or non-existent for rhabdoviruses. Thus we are confident that our methods are valid for quantifying the two genotypes. Given our results, an obvious question remains, why does HV have higher fitness than LV? The specific factors that provide the more virulent strain with higher viral fitness in the IHNV system remain under investigation. The molecular basis of virulence is unknown for IHNV, but preliminary results suggest that it involves more than one viral gene (unpublished data). Previous studies comparing the course of infection for two IHNV genotypes from different phylogenetic genogroups, indicate that factors such as host entry ability, replication rate, transcription, and innate immune evasion, could all potentially regulate fitness differences (Park et al., in press; Peñaranda et al., 2009; Purcell et al., 2009a). In these previous studies the two IHNV genotypes were from genogroups M and U, that differ in host specificity, presumably due to evolutionary adaptation to rainbow trout and sockeye salmon respectively (Kurath et al., 2003). A key difference between this previous work, and the present study, is that the two IHNV genotypes used here are both in the M genogroup, and thus have the same rainbow trout host specificity (Kurath et al., 2003). Furthermore, the other studies did not examine co-infection and were primarily interested in the kinetics of viral infection over time. Interestingly, in regards to the present study, in 17% of cases the less virulent strain was dominant in the mixed infections. This suggests that part of the general mechanism involved in regulating fitness may involve a significant advantage for the first virus genotype to establish an infection. In the future, mapping the location of fitness and virulence determinants on the viral genome could provide further insight into why genotypes differ in fitness. The broader question is how the fitness differences observed here will influence the evolution of viral virulence at the host population level. Our findings are consistent, although not conclusive, with the much debated trade-off hypothesis: the cost of shortened infection duration due to higher virulence comes with the benefit of increased viral fitness (Alizon et al., 2009). To fully understand the evolution of virulence on a host population scale, and accurately assess the tradeoff hypothesis, requires data on transmission of individual genotypes, and ultimately a measurement of lifetime fitness. The present study does not attempt to provide such information but is rather aimed at setting the foundation for understanding the nature of the virulence– viral fitness association. The logical progression from this work would 5 be an examination of IHNV genotype transmission and fitness over the entire course of infection. How our conclusions transcend across different host-pathogen taxa is also unknown. Our results agreed with some studies where a correlation between fitness and virulence was found (Arien et al., 2005; Carrillo et al., 1998; Frasher et al., 2007; Messenger et al., 1999; Sagar et al., 2003; Stewart et al., 2005; Suzuki et al., 2009; Troyer et al., 2005; Voronin et al., 2005) and differed from some where such a correlation was not found (Escriu et al., 2003; Herrera et al., 2007; Lenhoff et al., 1998; Pandrea et al., 2008). We suspect that to some degree these phenomena may be taxon dependent. Here we attempted to mimic field conditions by conducting all infections in vivo, in a naturally co-evolved host-pathogen association. Our work represents one of the few studies to examine the association of virulence with replication in a vertebrate virus system, using high numbers of intact living hosts. Rainbow trout mount well characterized innate, humoral, and cellular immune responses to IHNV infection (Lorenzen and LaPatra, 1999; Purcell et al., 2009b), an important factor for shaping the evolution of virulence across all vertebrates (Alizon et al., 2009; Frank and Schmid-Hempel, 2007). It is interesting to note that this study revealed a high level of fish-to-fish variation (Fig. 2, S2, S3), and indeed even a small degree of variability in the mean virus produced was seen between experiments, possibly due to differences between the fish lots used. The observation of up to four orders of magnitude difference in viral load between individual fish in these experiments is similar to the variation in virus titres found in the field (Mulcahy et al., 1982), thus indicating that our experiments mimic natural field infections. Furthermore, this variability exemplifies how powerful host genetic diversity as well as stochastic processes may be for shaping pathogen fitness, and reaffirms the importance of using a large, naturally diverse host population when empirically exploring virulence evolution theory. Given the large sample sizes in this study, this work offers a solid foundation to begin developing population level estimates of virulence evolution for further validation against field data and experimentation. Materials and methods Virus and host We used two genetically distinct IHNV isolates (genotypes) labeled HV (high virulence) and LV(low virulence), previously referred to as 220-90 and WRAC/039-82 respectively, which were originally obtained from farm cultured rainbow trout in the field and subsequently characterized for virulence (Garver et al., 2006). Over the entire viral G gene there is 3.6% (58/1621 nucleotides) sequence divergence between LV and HV. Both genotypes belong to the M genogroup of IHNV and thus have evolved with host specificity for rainbow trout (Kurath et al., 2003). The virus was propagated on EPC cells (Fijan et al., 1983) to create viral stocks stored at − 80 °C, then titered by replicate plaque assay (Batts and Winton, 1989; Troyer et al., 2008) as HV: 2.71 ± 0.067 × 107 pfu/ml and LV: 2.6 ± 0.69 × 107 pfu/ml (mean ± 1 S.E.M). The viral load of the virus stocks was determined by qRT-PCR on 8 independent replicates as HV: 0.906 ± 0.0832 × 109 and LV: 1.00 ± 0.221 × 109 viral copies/ml (mean ± 1 S.E. M). The ratio of viral load (qPCR) to pfu (plaque assay) was therefore HV: 33.6 ± 3.86 and LV: 43.8 ± 20.1 (mean ± 1 S.E.M). The hosts for all experiments were 1–3 g research grade juvenile rainbow trout, Oncorhynchus mykiss (Clear Springs Food, Inc.), maintained in flowthrough, pathogen-free, sand-filtered and UV-irradiated fresh water at 15 °C as described elsewhere (Troyer et al., 2008). All trout came from the same source, however the fish for experiment 1 were obtained at a different time of the year and therefore were a different lot. All animal procedures were approved under the University of Washington IACUC protocol 3042-11. Please cite this article as: Wargo, A.R., et al., Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV), Virology (2010), doi:10.1016/j.virol.2010.04.023 ARTICLE IN PRESS 6 A.R. Wargo et al. / Virology xxx (2010) xxx–xxx Virulence challenges The relative virulence of IHNV genotypes HV and LV was determined using standard viral challenges of fish in batch. Triplicate groups of 20 juvenile (2.5 g) rainbow trout were infected by immersion in water containing virus at a concentration of 2 × 105 pfu virus/ ml H2O of each genotype, or mock infected as a control. The challenge dose was chosen to be consistent with numerous previous batch virulence challenge studies of IHNV in juvenile salmonids (Garver et al., 2006). After a 1 hour exposure to virus in static water the fish were maintained as separate groups, in water flow-through conditions, and daily mortality was monitored for a period of 28 days as outlined elsewhere (Garver et al., 2006). In vivo viral competition To determine if virulence correlates with viral fitness in IHNV we quantified the within-host viral load of genotypes HV and LV in single and mixed infections, as a proxy for genotype fitness. Juvenile rainbow trout were infected by batch immersion in water with 1 × 104 pfu virus/ml H2O of each genotype alone, a mixed infection of 1 × 104 pfu virus/ml H2O of each genotype at a 1:1 ratio (2 × 104 pfu/ ml total virus), or mock infected (no virus), in group sizes of 23–28 fish per viral treatment as outlined in Table 1. We used twice as much virus in the mixed infection group so we could compare the performance of a viral genotype alone verses in competition at the same input dose. Previous studies indicated that this doubling of virus has no consequence for the viral dynamics (Peñaranda and Kurath, unpublished). The fish were exposed to virus for 12 h in static water conditions and then washed in flowing water for one hour and separated into individual beakers to prevent cross-infection (Troyer et al., 2008). The viral dosage for fitness challenges was chosen as the minimum dose at which all fish become infected with either viral genotype when held in isolation as previously described (Troyer et al., 2008). To assess genotype fitness in the fitness challenges, fish were euthanized and the viral load of each viral genotype in individual fish was quantified using genotype-specific quantitative reverse transcription PCR (qRT-PCR), after a 3 day period of in vivo growth and competition in static water. Day three was chosen as the point of fish harvest because it is known from viral growth kinetics studies to be when viral load reaches peak levels (Peñaranda et al., 2009; Troyer et al., 2008). RNA extraction and cDNA synthesis Viral RNA was extracted from whole fish as previously outlined (Troyer et al., 2008) with the following modifications. Denaturing solution (D.S. — 4 M guanidinium thiocynate, 25 mM sodium citrate pH7, 0.1 M 2-mercaptoethanol, 0.5% N-lauroylsarcosine) was added to each fish in individual whirl-packs™ at a ratio of 4 ml D.S. per gram Table 1 Treatment groups for the 3 experiments of the study. Treatment group (genotype, dose) LV, 1 × 104 pfu/ml HV, 1 × 104 pfu/ml LV:HV mixed, 2 × 104 pfu/ml Mock control (no virus)b Number of fish analyzeda Exp. 1 Exp. 2 Exp. 3 27 28 26 6 27 23 27 4 27 26 28 4 Genotype-specific qRT-PCR analysis LV assay HV assay LV and HV assays LV and HV assays a Excludes fish that died during the experiment. Mortalities were: experiment 1:1 (mix); experiment 2:0; experiment 3:1(HV). b Three of the fourteen mock samples were positive for HV by qPCR and two for LV, at virus concentrations an order of magnitude or greater below fish exposed to virus. This level of possible contamination, which likely occurred during sample processing, did not significantly affect the results. of fish (1/5 dilution). The fish were homogenized in a Seward Stomacher® 80 (Biomaster) at high speed for 2 × 60 s, or until completely liquefied, at room temperature. A 1 ml aliquot of the fish homogenate was then transferred to a 15 ml tube on ice for subsequent phenol-chloroform extraction as outlined elsewhere (Troyer et al., 2008) (Supplemental materials). To assess RNA extraction and concentration, spectrophotometry readings were taken of each sample. The RNA extraction method has proven to be highly repeatable with no significant difference in viral load quantified by qPCR of multiple aliquots of a sample extracted independently (data not shown). Extracted RNA was subsequently converted to cDNA by combining 0.125 µg oligo-dT, 0.125 µg random hexamers (Promega), 2 µl–3 µl RNA sample (∼1–2 µg total RNA), and enough H2O to bring the final volume to 13 µl. The mix was incubated at 70 °C for 5 min, held on ice and the volume brought to 20 µl with: 1 mM dNTPs, 4 µl 5× M-MLV reaction buffer, 100 U MMLV RT, 10 U RNAsin, and H2O (Promega). The sample was then incubated at 42 °C for 1 h, 70 °C for 15 min, and held at 4 °C until storage at −80 °C. Quantitative PCR The genotype-specific qPCR assays utilized genotype differentiating forward primers (HV: 5′-CCC GAT GCC AAT GGT ACA CT-3′; LV: 5′-CCG ATG CCA ATG GTA CAC C-3′), MGB TaqMan probes (HV: 5′MGB-VIC-CCC CAA AGA GTG TTC TGA-NFQ-3′; LV: 5′-MBG-6′FAMCAA CAC AAA AGA GTG TTC TTA-NFQ-3′; Applied Biosystems), and reverse primers (HV: 5′-GTG CAT TTT CCT CCA ATA AAG TCC-3′, LV: 5′-TTG GTA CAT TTT CCT CCA ATA AAA TCT-3′). Reliable quantification was obtained for HV and LV in artificial mixtures spanning ratios of 10,000:1–1:1–1:10,000 HV:LV (HV: Log (observed quantity) = Log (expected quantity) ⁎ (1.005 ± 0.01) − 0.03 ± 0.06, F1,29 = 8400, P b 0.001, R2 = 0.99; LV: Log(observed quantity) = Log(expected quantity) ⁎ (1.002 ± 0.01) + 0.01 ± 0.05, F1,32 = 9700, P b 0.001, R2 = 0.99) with no detection of non-target at 1 × 107 copies and target minimum detection limits of 10 copies/reaction (104 viral copies/g fish). All qPCR reactions were at a total volume of 12 µl, containing 900 nM each primer, 200 nM probe, 6 µl 2× TaqMan Universal PCR Master Mix (Applied Biosystems), 0.328 µl H2O, and 5 µl of a 1/10 dilution of sample cDNA, and run on a 7900HT ABI Prism machine. Cycling conditions were 50 °C for 2 min, 95 °C for 10 min, then 40 cycles of 95 °C for 15 s and 60 °C for 1 min. Given the robust nature of the qPCR methodology to detect all viral RNA, it quantified total viral genetic material, including both genome copies and mRNA, and thus indicates “viral load” rather than numbers of virus particles. A consistent relationship between viable virus quantities found in plaque assays and qPCR viral load has previously been observed for IHNV in rainbow trout (Purcell et al., 2006) and qPCR viral load has regularly been used to quantify viral replication (Peñaranda et al., 2009; Purcell et al., 2009c). Transcript standard development RNA transcripts of the full viral glycoprotein (G) gene (target of qPCR) were developed for HV and LV and used as qRT-PCR standards at known concentrations. The development of the LV-G plasmid (pWracG) has previously been described (Corbeil et al., 2000), pHV-G was developed as outlined here. As a template for the pHV-G plasmid, HV viral RNA was extracted from a 400 µl volume of virus in cell free media with 500 µl of Tri® Reagent (Sigma-Aldrich) plus 100 µl EPC cells as a carrier according to the manufacturer's protocol (supplemental materials). The viral RNA was converted to cDNA using SuperScript II™ reverse transcriptase (Invitrogen) as suggested by the manufacturer, with 1.6 µg RNA and 2 pmol of a G gene specific primer (5′-GGA CTA GTA TGG ACA CCA CGA TCA CCA CTC CG-3′). The cDNA was then PCR amplified to produce the full 1.6 kb G gene product using Taq DNA polymerase (Promega) in a 100 µl reaction with: 10 µl Please cite this article as: Wargo, A.R., et al., Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV), Virology (2010), doi:10.1016/j.virol.2010.04.023 ARTICLE IN PRESS A.R. Wargo et al. / Virology xxx (2010) xxx–xxx 10× PCR buffer, 0.2 mM DNTPs, 2.5 mM MgCl2, 25 U Taq, 7.5 mg RNase A (Promega), 5 µl cDNA, 1 pmol primers (forward: 5′-GGA CTA GTA TGG ACA CCA CGA TCA CCA CTC CG-3′; reverse: 5′-GGC CCG GGT TAG GAC CGG TTT GCC ACG TGA T-3′). Cycling conditions were 94 °C 2 min, 30 cycles: 94 °C 30 s, 50 °C 30 s, 72 °C 2 min; followed by 72 °C 10 min, 4 °C hold. The HV PCR product was gel purified using the QIAquick® gel extraction kit according to the manufacturer's protocol (Qiagen) and then inserted into pSTBlue-1 vector using the AccepTor® vector kit as outlined by the manufacturer (Novagen). Appropriate colonies were sequenced to ensure full G gene insertion and a stock of plasmid pHVG was prepared using the EndoFree® Plasmid Maxi kit (Qiagen). To obtain negative-sense transcript RNA, 20 µg of plasmids pHV-G and pLV-G were digested with HindIII and BamHI respectively and RNA transcripts were synthesized with T7 RNA polymerase (Promega). Residual DNA was digested with RQ1 DNase and DNA contamination was subsequently found to be less than 0.1%. The concentrations and gene copy numbers of transcript standards rHV-G and rLV-G (length 1626 bp and 1745 bp respectively) were determined by spectrophotometry as outlined elsewhere (Purcell et al., 2006). We also used a general non-genotype-specific IHNV qPCR reaction to further calibrate the standards against each other (Purcell et al., 2006). Every set of cDNA reactions included an RNA transcript standard (rHV or rLV) which was utilized for qPCR in a 10 fold cDNA dilution series with G gene copies ranging from 5 × 108 to 1 × 101 (standard curves: HV: R2 = 0.997 ± 0.001, slope = −3.48 ± 0.04, intercept = 45.36 ± 0.31; LV: R2 = 0.998 ± 0.001, slope= -3.49±0.03, intercept =44.49 ± 0.13). Statistics Virulence data was tested with a Cox-test (SPSS), revealing that replicate tank was not a significant predictor, which was thus dropped from all further analyses (P N 0.10). Kaplan–Meier and log-rank tests (SPSS) were conducted to determine differences in survivorship. Data divided or pooled by tank showed the same result. Mean day to death and percent mortality were analyzed using Chi-square tests in excel. Analysis of genotype viral production was carried out in SPSS using general linear models (GLM) starting with the maximum model (viral load = viral genotype (HV or LV) + competition (alone versus mixed) + experiment (I, II, III) + viral genotype ⁎ competition + viral genotype⁎ experiment+ competition ⁎ experiment+ viral genotype⁎ competition ⁎ experiment) and non-significant terms were dropped until the minimal model was reached (supplemental materials). Tukey's posthoc test was utilized for multiple comparisons within groups. To avoid violating independence assumptions, for statistical analysis we randomly assigned half the mixed infection fish to the HV group and the other half to the LV group. The same results were obtained when analyzing all fish (Tables S1–S6, Fig. S1) and therefore these data are shown in the figures. All data were log transformed to meet the test assumptions and dead fish were excluded from the GLM viral production analysis because exact time of death could not be determined and may influence viral replication (Table 1). Acknowledgments We thank B. Batts and J. Ranson for technical assistance and J. Winton and B. Kerr for valuable discussions, as well as two anonymous reviewers for thoughtful comments. We are grateful to S.E. LaPatra of Clear Springs Foods, Inc., for providing research grade rainbow trout. This work was supported by the USGS Western Fisheries Research Center, and funding for A. Wargo was provided by the National Institute of Health R.L. Kirschestein NRSA University of Washington institutional service award 5 T32 AI007509 and National Science Foundation Ecology of Infectious Diseases grant 0812603. The funders had no role in study design, data collection and analysis, 7 decision to publish, or preparation of the manuscript. Mention of trade names does not imply U.S. Government endorsement. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.virol.2010.04.023. References Alizon, S., Hurford, A., Mideo, N., Van Baalen, M., 2009. Virulence evolution and the trade-off hypothesis: history, current state of affairs and the future. J. Evol. Biol. 22 (2), 245–259. Anderson, E.D., 2000. Molecular epidemiology reveals emergence of a virulent infectious hematopoietic necrosis (IHN) virus strain in wild salmon and its transmission to hatchery fish. J. Aquat. Anim. Health 12 (2), 85–99. Arien, K.K., Abraha, A., Quinones-Mateu, M.E., Kestens, L., Vanham, G., Arts, E.J., 2005. The replicative fitness of primary human immunodeficiency virus type 1 (HIV-1) group M, HIV-1 group O, and HIV-2 isolates. J. Virol. 79 (14), 8979–8990. Batts, W.N., Winton, J.R., 1989. Enhanced detection of infectious hematopoietic necrosis virus and other fish viruses by pretreatment of cell monolayers with polyethylene glycol. J. Aquat. Anim. Health 1 (4), 284–290. Bootland, L.M., Leong, J.C., 1999. Infectious hematopoietic necrosis virus. In: Woo, P.T.K., Bruno, D.W. (Eds.), Fish diseases and disorders: CAB International, Vol. 3, pp. 57–112. Brown, N.F., Wickham, M.E., Coombes, B.K., Finlay, B.B., 2006. Crossing the line: Selection and evolution of virulence traits. PLoS Pathog. 2 (5), e42. Brudeseth, B.E., Castric, J., Evensen, O., 2002. Studies on pathogenesis following single and double infection with viral hemorrhagic septicemia virus and infectious hematopoietic necrosis virus in rainbow trout (Oncorhynchus mykiss). Vet. Pathol. 39 (2), 180–189. Bull, J.J., 1994. Perspective: Virulence. Evolution 48 (5), 1423–1437. Burke, B., Derby, N.R., Kraft, Z., Saunders, C.J., Dai, C., Llewellyn, N., Zharkikh, I., Vojtech, L., Zhu, T., Srivastava, I.K., Barnett, S.W., Stamatatos, L., 2006. Viral evolution in macaques coinfected with CCR5- and CXCR4-tropic SHIVs in the presence or absence of vaccine-elicited anti-CCR5 SHIV neutralizing antibodies. Virology 355 (2), 138–151. Carrillo, C., Borca, M., Moore, D.M., Morgan, D.O., Sobrino, F., 1998. In vivo analysis of the stability and fitness of variants recovered from foot-and-mouth disease virus quasispecies. J. Gen. Virol. 79 (7), 1699–1706. Chumakov, K.M., Powers, L.B., Noonan, K.E., Roninson, I.B., Levenbook, I.S., 1991. Correlation between amount of virus with altered nucleotide sequence and the monkey test for acceptability of oral poliovirus vaccine. Proc. Natl. Acad. Sci. U. S. A. 88 (1), 199–203. Corbeil, S., LaPatra, S.E., Anderson, E.D., Kurath, G., 2000. Nanogram quantities of a DNA vaccine protect rainbow trout fry against heterologous strains of infectious hematopoietic necrosis virus. Vaccine 18 (25), 2817–2824. de las Heras, A.I., Saint-Jean, S.R., Perez-Prieto, S.I., 2008. Salmonid fish viruses and cell interactions at early steps of the infective cycle. J. Fish Dis. 31 (7), 535–546. de Roode, J.C., Pansini, R., Cheesman, S.J., Helinski, M.E.H., Huijben, S., Wargo, A.R., Bell, A.S., Chan, B.H.K., Walliker, D., Read, A.F., 2005. Virulence and competitive ability in genetically diverse malaria infections. Proc. Natl. Acad. Sci. U. S. A. 102 (21), 7624–7628. Domingo, E., Holland, J.J., 1997. RNA virus mutations and fitness for survival. Ann. Rev. Microbiol. 51, 151–178. Domingo, E., Menéndez-Arias, L., Holland, J.J., 1997. RNA virus fitness. Rev. Med. Virol. 7 (2), 87–96. Drolet, B.S., Rohovec, J.S., Leong, J.C., 1994. The route of entry and progression of Infectious Hematopoietic Necrosis Virus in Oncoryhynchus-mykiss (Walbaum)- A sequential immunohistochemical study. J. Fish Dis. 17 (4), 337–347. Ebert, D., 1998. Evolution - Experimental evolution of parasites. Science 282 (5393), 1432–1435. Emmenegger, E.J., Troyer, R.M., Kurath, G., 2003. Characterization of the mutant spectra of a fish RNA virus within individual hosts during natural infections. Virus Res. 96, 15–25. Escriu, F., Fraile, A., Garcia-Arenal, F., 2003. The evolution of virulence in a plant virus. Evolution 57 (4), 755–765. Fijan, N., Sulimanovic, E., Bearzotti, M., Muzinic, D., Zwillenberg, L.O., Chilmonczyk, S., Vautherot, J.F., de Kinkelin, P., 1983. Some properties of the epithelioma papulosum cyprini (EPC) cell line from carp Cyprinus carpi. Annales de l'Institut Pasteur Virologie 134 (2), 207–220. Frank, S.A., 1996. Models of parasite virulence. Q. Rev. Biol. 71 (1), 37–78. Frank, S.A., Schmid-Hempel, P., 2007. Mechanisms of pathogensis and the evolution of parasite virulence. J. Evol. Biol. 21 (2), 396–404. Frasher, C., Hollingsworth, T.D., Chapman, R., de Wolf, F., Hanage, W.P., 2007. Variation in HIV-1 set-point viral load: Epidemiological analysis and an evolutionary hypothesis. Proc. Natl. Acad. Sci. U. S. A. 104 (44), 17441–17446. Garver, K., Batts, W., Kurath, G., 2006. Virulence comparisons of infectious hematopoietic necrosis virus U and M genogroups in sockeye salmon and rainbow trout. J. Aquat. Anim. Health 18 (4), 232–243. Harmache, A., LeBerre, M., Droineau, S., Giovannini, M., Bremont, M., 2006. Bioluminescence imaging of live infected salmonids reveals that the fin bases are the major portal of entry for Novirhabdovirus. J. Virol. 80 (7), 3655–3659. Please cite this article as: Wargo, A.R., et al., Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV), Virology (2010), doi:10.1016/j.virol.2010.04.023 ARTICLE IN PRESS 8 A.R. Wargo et al. / Virology xxx (2010) xxx–xxx Helmick, C.M., Bailey, J.F., LaPatra, S., Ristow, S., 1995. The esophagus/cardiac stomach region: Site of attachment and internalization of infectious hematopoietic necrosis virus in challenged juvenile rainbow trout Oncorhynchus mykiss and coho salmon O-kisutch. Dis. Aquat. Org. 23 (3), 189–199. Herrera, M., Garcia-Arriaza, J., Pariente, N., Escarmis, C., Domingo, E., 2007. Molecular basis for a lack of correlation between viral fitness and cell killing capacity. PLoS Pathog. 3 (4), 498–507. Hodgson, D.J., Hitchman, R.B., Vanbergen, A.J., Hails, R.S., Possee, R.D., Cory, J.S., 2004. Host ecology determines the relative fitness of virus genotypes in mixed-genotype nucleopolyhedrovirus infections. J. Evol. Biol. 17 (5), 1018–1025. Holland, J.J., Delatorre, J.C., Clarke, D.K., Duarte, E., 1991. Quantitation of relative fitness and great adaptability of clonal populations of RNA viruses. J. Virol. 65 (6), 2960–2967. Kurath, G., Garver, K.A., Troyer, R.M., Emmenegger, E.J., Einer-Jensen, K., Anderson, E.D., 2003. Phylogeography of infectious haematopoietic necrosis virus in North America. J. Gen. Virol. 84 (4), 803–814. LaPatra, S.E., Lauda, K.A., Jones, G.R., 1994. Antigenic variants of infectious hematopoietic necrosis virus and implications for vaccine development. Dis. Aquat. Org. 20 (2), 119–126. Lenhoff, R.J., Luscombe, C.A., Summers, J., 1998. Competition in vivo between a cytopathic variant and a wild-type duck hepatitis B virus. Virology 251 (1), 85–95. Levin, B.R., 1996. The evolution and maintenance of virulence in microparasites. Emerging Infect. Dis. 2 (2), 93–102. Lorenzen, N., LaPatra, S.E., 1999. Immunity to rhabdoviruses in rainbow trout: the antibody response. Fish Shellfish Immunol. 9 (4), 345–360. Messenger, S.L., Molineux, I.J., Bull, J.J., 1999. Virulence evolution in a virus obeys a trade-off. Proc. R. Soc. London Ser. B. 266 (1417), 397–404. Mulcahy, D., Burke, J., Pascho, R., Jenes, C.K., 1982. Pathogenesis of infectious hematopoietic necrosis virus in adult sokeye salmon (Oncorhynchus nerka). Can. J. Fish. Aquat. Sci. 39 (8), 1144–1149. Munster, V.J., de Wit, E., van den Brand, J.M.A., Herfst, S., Schrauwen, E.J.A., Bestebroer, T.M., van de Vijver, D., Boucher, C.A., Koopmans, M., Rimmelzwaan, G.F., Kuiken, T., Osterhaus, A., Fouchier, R.A.M., 2009. Pathogenesis and transmission of swineorigin 2009 A(H1N1) influenza virus in ferrets. Science 325 (5939), 481–483. Novella, I.S., Duarte, E.A., Elena, S.F., Moya, A., Domingo, E., Holland, J.J., 1995. Exponential increases of RNA virus fitness during large population transmissions. Proc. Natl. Acad. Sci. U. S. A. 92 (13), 5841–5844. Ojosnegros, S., Beerenwinkel, N., Antal, T., Nowak, M.A., Escarmis, C., Domingo, E., 2010. Competition-colonization dynamics in an RNA virus. Proc. Natl. Acad. Sci. U. S. A. 107 (5), 2108–2112. Pandrea, L., Sodora, D.L., Silvestri, G., Apetrei, C., 2008. Into the wild: simian immunodeficiency virus (SIV) infection in natural hosts. Trends Immunol. 29 (9), 419–428. Park, J.W., Moon, C.H., Wargo, A.R., Purcell, M.K., and Kurath, G., in press. Differential growth of U and M type Infectious hematopoietic necrosis virus in a rainbow trout derived cell line, RTG-2. J. Fish Dis. doi:10.1111/j.1365-2761.2010.01153.x. Peñaranda, M.M.D., Purcell, M.K., Kurath, G., 2009. Differential virulence mechanisms of infectious hematopoietic necrosis virus (IHNV) in rainbow trout (Oncorhynchus mykiss) include host entry and virus replication kinetics. J. Gen. Virol. 90 (9), 2172–2182. Purcell, M.K., Hart, S.A., Kurath, G., Winton, J.R., 2006. Strand-specific, real-time RT-PCR assays for quantification of genomic and positive-sense RNAs of the fish rhabdovirus, infectious hematopoietic necrosis virus. J. Virol. Methods 132 (1–2), 18–24. Purcell, M.K., Garver, K.A., Conway, C., Elliott, D.G., Kurath, G., 2009a. Infectious haematopoietic necrosis virus genogroup-specific virulence mechanisms in sockeye salmon, Oncorhynchus nerka (Walbaum), from Redfish Lake, Idaho. J. Fish Dis. 32 (7), 619–631. Purcell, M.K., Laing, K., Woodson, J., Thorgaard, G., Hansen, J., 2009b. Characterization of the interferon genes in homozygous rainbow trout reveals two novel genes, alternate splicing and differential regulation of duplicated genes. Fish Shellfish Immunol. 26 (2), 293–304. Purcell, M.K., Lapatra, S.E., Woodson, J.C., Kurath, G., Winton, J.R., 2009c. Early viral replication and induced or constitutive immunity in rainbow trout families with differential resistance to Infectious hematopoietic necrosis virus (IHNV). Fish Shellfish Immunol. 28 (1), 98–105. Quinones-Mateu, M.E., Ball, S.C., Marozsan, A.J., Torre, V.S., Albright, J.L., Vanham, G., van der Groen, G., Colebunders, R.L., Arts, E.J., 2000. A dual infection/competition assay shows a correlation between ex vivo human immunodeficiency virus type 1 fitness and disease progression. J. Virol. 74 (19), 9222–9233. Read, A.F., 1994. The evolution of virulence. Trends Microbiol. 2 (3), 73–76. Read, A.F., Taylor, L.H., 2001. The ecology of genetically diverse infections. Science 292 (5519), 1099–1102. Sagar, M., Lavreys, L., Baetan, J.M., Richardson, B.A., Mandaliya, K., Chohan, B.H., Kreiss, J. K., Overbaugh, J., 2003. Infection with multiple human immunodeficiency virus type 1 variant is associated with faster disease progression. J. Virol. 77 (23), 12921–12926. Stewart, A., Logsdon, J.J., Kelly, S., 2005. An empirical study of the evolution of virulence under both horizontal and vertical transmission. Evolution 59 (4), 730–739. Suzuki, K., Okada, H., Itoh, T., Tada, T., Mase, M., Nakamura, K., Kubo, M., Tsukamoto, K., 2009. Association of increased pathogenicity of Asian H5N1 highly pathogenic avian influenza viruses in chickens with highly efficient viral replication accompanied by early destruction of innate immune responses. J. Virol. 83 (15), 7475–7486. Troyer, R.M., Kurath, G., 2003. Molecular epidemiology of infectious hematopoietic necrosis virus reveals complex virus traffic and evolution within southern Idaho aquaculture. Dis. Aquat. Org. 55 (3), 175–185. Troyer, R.M., LaPatra, S.E., Kurath, G., 2000. Genetic analyses reveal unusually high diversity of infectious haematopoietic necrosis virus in rainbow trout aquaculture. J. Gen. Virol. 81 (12), 2823–2832. Troyer, R.M., Collins, K.R., Abraha, A., Fraundorf, E., Moore, D.M., Krizan, R.W., Toossi, Z., Colebunders, R.L., Jensen, M.A., Mullins, J.I., Vanham, G., Arts, E.J., 2005. Changes in human immunodeficiency virus type 1 fitness and genetic diversity during disease progression. J. Virol. 79 (14), 9006–9018. Troyer, R.M., Garver, K.A., Ranson, J.C., Wargo, A.R., Kurath, G., 2008. In vivo virus growth competition assays demonstrate equal fitness of fish rhabdovirus strains that co-circulate in aquaculture. Virus Res. 137 (2), 179–188. van Baalen, M., Sabelis, M.W., 1995. The dynamics of multiple infection and the evolution of virulence. Am. Nat. 146 (6), 881–910. Voronin, Y., Overbaugh, J., Emerman, M., 2005. Simian immunodeficiency virus variants that differ in pathogenicity differ in fitness under rapid cell turnover conditions. J. Virol. 79 (24), 15091–15098. Please cite this article as: Wargo, A.R., et al., Virulence correlates with fitness in vivo for two M group genotypes of Infectious hematopoietic necrosis virus (IHNV), Virology (2010), doi:10.1016/j.virol.2010.04.023