Interferon Induction and/or Production and Its Sup
http://www.100md.com
病菌学杂志 2005年第5期
Department of Molecular and Cell Biology
Department of Pathobiology and Veterinary Science
Center for Excellence in Vaccine Research, University of Connecticut, Storrs, Connecticut
ABSTRACT
Developmentally aged chicken embryo cells which hyperproduce interferon (IFN) when induced were used to quantify IFN production and its suppression by eight strains of type A influenza viruses (AIV). Over 90% of the IFN-inducing or IFN induction-suppressing activity of AIV populations resided in noninfectious particles. The IFN-inducer moiety of AIV appears to preexist in, or be generated by, virions termed IFN-inducing particles (IFP) and was detectable under conditions in which a single molecule of double-stranded RNA introduced into a cell via endocytosis induced IFN, whereas single-stranded RNA did not. Some AIV strains suppressed IFN production, an activity that resided in a noninfectious virion termed an IFN induction-suppressing particle (ISP). The ISP phenotype was dominant over the IFP phenotype. Strains of AIV varied 100-fold in their capacity to induce IFN. AIV genetically compromised in NS1 expression induced about 20 times more IFN than NS1-competent parental strains. UV irradiation further enhanced the IFN-inducing capacity of AIV up to 100-fold, converting ISP into IFP and IFP into more efficient IFP. AIV is known to prevent IFN induction and/or production by expressing NS1 from a small UV target (gene NS). Evidence is presented for an additional downregulator of IFN production, identified as a large UV target postulated to consist of AIV polymerase genes PB1 + PB2 + PA, through the ensuing action of their cap-snatching endonuclease on pre-IFN-mRNA. The products of both the small and large UV targets act in concert to regulate IFN induction and/or production. Knowledge of the IFP/ISP phenotype may be useful in the development of attenuated AIV strains that maximally induce cytokines favorable to the immune response.
INTRODUCTION
Almost 50 years ago Alick Isaacs and Jean Lindenmann exposed fragments of chorioallantoic membranes from chicken eggs to a strain of influenza virus A (AIV) that had been inactivated by heat (26). This treatment rendered the virus noninfectious, yet capable of stimulating cells to produce a secreted factor which, when added to fresh cells for 24 h, interfered with the replication of an active preparation of that same virus. With this experiment, they discovered the active component of the most common type of viral interference and aptly termed it interferon (IFN). Many of the basic properties of IFN and its induction and/or production in chicken cells soon were established in Isaacs' laboratory (7, 8, 27, 38). These seminal studies launched a new era in our understanding of viral interference, representing as they did the first report of the antiviral activity of an inducible, secreted, and isolable cytokine.
The intrinsic sensitivity of influenza virus to the antiviral action of IFN in vitro in both chicken and mammalian cells has been established (3, 22, 23, 67, 69, 71). This provides further impetus to determine the full complement of viral genes and gene products that regulate the capacity of AIV to induce and/or produce IFN, a first step in the activation of the IFN action pathways. The critical role of the IFN system during influenza virus infection was first demonstrated in a classical experiment in which mice infected with influenza virus and injected with anti-IFN serum were observed to produce higher titers of virus and an exacerbated pathology (21). Subsequent studies with knockout mice defective in the IFN system have confirmed this basic observation. Recent studies in which the IFN-inducing capacity varied in otherwise genetically related AIV showed that the strain of AIV capable of producing the most IFN was attenuated in its pathogenicity in mice (13) and in chickens (A. N. Cathuen and D. Suarez, Southeast Poultry Research Laboratory, Athens, Ga.; personal communication).
Although type A influenza viruses (AIV) are intrinsically sensitive to the action of the IFN they induce, the virus continues to expand its propensity to cause pandemics and epornitics of major public health concern. The perennial worldwide appearance of influenza as an infectious disease attests to its potential to decimate both human and avian populations. Consider that up to 47,000 deaths are attributed to influenza in the U.S. each year (10), coupled with the unprecedented deaths of humans in Asia from H5N1 avian-derived influenza virus and the extermination of more than 100 million fowl (39). And now, the emergence of a pathogenic species of the H5N1 avian virus that kills waterfowl—heretofore considered naturally resistant hosts (39)—and felines (15a) alerts us to the increasingly broader host range for lethality acquired by influenza virus. With chickens the largest source of animal protein, the periodic depopulation of millions of chickens as a means of containing AIV-based epornitics adds to the urgency to understand the role of the IFN system in regulating the pathogenicity of this virus—especially since IFN acts independently of viral antigenic variation, can enhance the immune response (reviewed in reference 75), and may be effective against several avian viruses (57, 60, 64).
We have revisited the influenza virus-chicken cell system to address biological features of IFN induction and/or production by AIV in light of new advances in our understanding of the molecular features of this important pathogen (33, 86). This report examines eight strains of AIV for two biological attributes not previously quantified for populations of influenza viruses, namely their content of virions that (i) induce IFN (IFN-inducing particles [IFP]) (46), or (ii) suppress the induction of IFN in cells otherwise programmed to produce it (IFN induction-suppressing particles [ISP]) (52). These two antagonistic phenotypes (44) were measured in populations of AIV that were (i) active, (ii) genetically compromised for NS1 expression, (iii) UV inactivated, or (iv) heat inactivated. The nature of the molecule inducing IFN in the AIV-chicken cell systems is examined, and evidence is presented for a new means by which AIV downregulates IFN induction/production.
(This work was presented in part at the 23rd Meeting of the American Society for Virology, Montreal, Quebec, Canada, 13 July 2004 [ASV Abstr., p. 200, 2004].)
MATERIALS AND METHODS
Cells and media. The preparation and developmental aging of primary chicken embryo cells (CEC) prepared from 9-day-old embryonated eggs (Charles River SPAFAS, Inc., North Franklin, Conn.) has been described (70, 73). Briefly, confluent monolayers of primary chicken embryo cells were established overnight and "developmentally aged" for 7 to 10 days: i.e., incubated at 37.5°C in NCI medium plus 6% calf serum (attachment solution) without a medium change. When appropriately induced and incubated at 40.5°C, these cells will produce levels of IFN many-fold higher than nonaged cells (73). Chick cells aged 4 to 5 days were used for the IFN assay.
Viruses: source, preparation, and assay. Colleagues kindly provided seed stocks of the following AIV: of avian origin; A/TK/ONT/7732/66 (H5N9) (Virginia Hinshaw, University of California, Davis) (83); A/ONT/7732/66(Clone 1B) (H5N9), which was plaque derived from the former stock after one passage on CEC treated with 50 U of recombinant chicken alpha IFN (rChIFN-) per ml for 24 h (71); A/TK/WI/66 (H9N2) (Theodore Girshick, Charles River SPAFAS, Inc.); A/TK/OR/71(H7N3); and A/TK/OR/71(delNS1[1-124]), which contains the DSR-binding region but with deletion of amino acids 125 to 230 and produces NS1 protein lacking the effector region at the C terminus (D. Suarez, Southeast Poultry Research Laboratory) (62, 77). The seed stocks of AIV of human origin were A/WSN/33 (H1N1), A/PR/8/34 (H1N1), and A/PR/8/34/delNS1 (H1N1) (A. García-Sastre and P. Palese, Mount Sinai School of Medicine, New York, N.Y.). The delNS1 strain has a deletion in the NS gene which eliminates all of the NS1 open reading frame except for the first 10 amino acids, which are shared with the viral nuclear export protein NS2 (19); thus, it lacks both the DSR-binding and effector regions. For convenience, this PR/8 strain is termed NS1 to distinguish it from the strain of A/TK/OR delNS1 with the truncated C terminus lacking amino acids 125 to 230 (77).
All stocks of influenza virus were grown in the amniotic/allantoic membranes of 7-to 9-day-old embryonated chicken eggs from specific-pathogen-free flocks from Charles River SPAFAS, Inc.. Each egg was injected with 0.1 ml containing 103 infectious particles, incubated for 48 to 72 h at 34°C with forced air circulation and egg rotation, and held at 4°C for 24 h before harvesting. The amniotic/chorioallantoic fluid was harvested separately from each egg. Those with high hemagglutinating activity were pooled, and aliquots were prepared and stored at –80°C.
Strain A/TK/WI/66 stocks were assayed as 50% egg infective doses (EID50) (in units per milliliter) by Theodore Girshick of Charles River SPAFAS, Inc. Infectivities of A/TK/ONT/7732/66 and WSN/33 stocks were measured by plaque assay in the absence of trypsin. All other strains required washing the monolayers twice with serum-free medium followed by 1.4 to 1.6 μg of trypsin per ml (code T 1426; Sigma-Aldrich, Co., St. Louis, Mo.) in the agarose overlay to achieve consistent plaque formation. Serum was omitted from the overlay during all trypsin-based assays.
IFN induction and assay. Detailed protocols have been described for the induction and assay of chicken IFN in primary CEC (73). UV-irradiated avian reovirus (UV-ARV) was used as a standard IFN inducer (85). It induced 10,000 U/107 aged CEC as determined from over 50 assays. This inducer was used as a standard to assess the intrinsic capacity of each batch of developmentally aged primary CEC to produce IFN. No basal levels of IFN were ever detected in noninduced cells (<1 U/ml.) Aliquots of a partially purified natural type I ChIFN (72) were prepared and used as a standard to assess the sensitivity of each batch of primary CEC to the action of IFN (73) and were usually assayed at 8,000 U/ml. For each batch of primary CEC, the intrinsic capacity to produce IFN and to respond to its action were normalized to these two standards. All IFN inductions started with a 1-h attachment of virus in 300 μl of attachment solution (NCI medium plus 6% calf serum) to a monolayer of 107 primary CEC that had been aged in vitro for 7 to 10 days in 50-mm-diameter petri dishes. Following incubation at 40.5°C, the 3 ml of medium bathing the cells was removed and processed for acid-stable, type I IFN and assayed in a 96-well system as described previously (73).
Assay of IFP and ISP. The assay of IFP was carried out as described earlier by generating and analyzing IFN induction dose (multiplicity)-response (IFN yield) curves (46) under conditions where a single defective-interfering particle which contains 1 molecule of self-complementary [±]RNA covalently linked in the middle induces a full yield of IFN and a related defective-interfering particle with [–] RNA induces little or no IFN (45, 50). This assay measures the IFN-inducing capacity of a virus particle independent of infectivity. The ISP content of influenza virus stocks was determined as previously reported (52). Briefly, a monolayer of CEC was exposed simultaneously to an IFN-inducing virus at a multiplicity high enough to ensure that all cells produce IFN and at increasing multiplicities of the putative ISP. After 24 h at 40.5°C, the medium was assayed for IFN. The fraction of surviving yield of IFN as a multiplicity of ISP was calculated, and the concentration of ISP in the stock was determined as described in Results.
UV irradiation and heat inactivation. UV irradiation was carried out as previously described (49). Vesicular stomatitis virus (VSV) was used as an actinometer. A survival curve of VSV plaque-forming particle (PFP) activity generated as a function of time exposed to UV radiation (254 nm) revealed an exponential loss of infectivity with D37 = 52.5 ergs/mm2, as measured originally with a Laterjet instrument. For influenza virus, D37 = 82.8 ergs/mm2 as measured by us. This compares with a stated value of 72 ergs/mm2 reported previously (1) and a value of 87 ergs/mm2 calculated by us from the data presented (1). Heat inactivation followed the regimen used by Isaacs and Lindenmann (26). However, 50°C was used rather than 56°C because the rates of inactivation were too rapid at the higher temperature to measure accurately for the strains of virus tested. NaHCO3-buffered NCI solution equilibrated with air was used to achieve pH 8.4. The pH 8.5 borate buffer described by Isaacs and Lindenmann (26) was used with comparable results. The relatively low yields of IFN they reported are attributed in part to the use of chorioallantoic membranes from insufficiently mature embryos and Earle's saline as a rinse—which in our hands usually lowered the yield of IFN.
RESULTS
Time course of IFN production in CEC infected with active influenza viruses. Figure 1 shows the time course of IFN production from developmentally aged primary CEC cultures induced by five strains of active influenza virus, including the contrasting responses generated by parental virus and their genetically related NS1-compromised counterparts. Virus multiplicities were sufficiently high to initially infect all cells in the monolayer with particles capable of inducing IFN. All IFN induction experiments reported in this study were carried out in the absence of trypsin, a protease required for five of the eight strains to sustain plaque formation—the two TK/ONT-derived isolates and WSN being exceptions. Since trypsin-dependent virus strains were capable of inducing IFN, it follows that the egg-derived virus used in these studies contained activated fusion protein and was able to attach to and enter cells (35) to induce IFN. However, the fusion protein of any newly released virus was not cleaved by the primary CEC, thus preventing secondary infection. This condition preserves the input multiplicity and allows analysis of IFN induction dose-response curves as previously reported (46).
Figure 1A is representative of most time course curves generated from good inducers of IFN. IFN was detectable in the medium after a lag of about 2 to 4 h. This lag was followed by a linear increase in the accumulation of IFN in the growth medium. Significant differences were observed in the rate at which maximum yield of IFN was obtained. However, peak yields generally were reached by about 15 to 25 h. Figures 1B and C compare the time course of IFN production by two parental viruses and their genetically related strains with compromised expression of NS1 protein. Figure 1B shows the parent A/TK/OR/71 is a weak inducer of IFN, but its genetically-related strain, A/TK/OR/71(delNS1[1-124]), which produces a C-terminus-truncated NS1 protein, is an excellent inducer of IFN. Figure 1C illustrates a comparable pair of strains, the A/PR/8/34 parent and its genetically derived mutant with the severely ablated NS1 gene. For some strains, the maximal accumulation of IFN was followed by a gradual loss in activity, attributed in part to the release of cellular proteases which accompany the apoptosis-based cytopathic effects we observed at higher multiplicities, and has been reported for influenza viruses (25, 54, 79, 82).
IFN induction dose-response curves generated by different strains of active influenza virus. Figure 2 illustrates as a function of the input concentration of active virus, the yield of IFN produced by monolayers of aged CEC 24 h after induction and incubation at 40.5°C. AIV strains with truncated (Fig. 2A) or deleted (Fig. 2B) NS1 genes are compared with their genetically related parents. These results are in keeping with the compromised replication of NS1-deficient AIV observed in IFN-competent cells (4, 13, 19). Dose-response curves are representative of two to three separate experiments. An IFN dose-response curve typical of the other four strains tested is shown and analyzed in depth in Fig. 4. An inducible double-stranded RNase (dsRNase) was coproduced with the IFN (59; data not shown).
Maximal yields of IFN from different strains of active AIV. Figure 3 shows that the maximal yields of IFN produced by aged CEC monolayers induced with eight different active allantoic fluid-derived stocks of AIV covered a 100-fold range, from about 85 to 8,500 U/107 cells. It documents the differences that characterize their intrinsic IFN-inducing capacity as averaged from three or four independent determinations.
The IFP titer of active influenza virus. In the most common form of the IFN induction dose-response curve, the yield of IFN increases as a function of virus multiplicity until it reaches a plateau. Based on a Poisson distribution of virus particles, the plateau is attained when virtually all cells in the monolayer are infected, i.e., induced to produce a quantum yield of IFN characteristic of that strain (46). The plateau is maintained at higher multiplicities unless the virus has a deleterious effect on the cell, in which case a decline in IFN production may be observed. This may be due to the apoptosis that commonly accompanies influenza virus infection (25, 79) or the action of ISP present at higher multiplicities (see below). The number of particles in a virus population capable of inducing IFN, i.e., IFP, can be calculated from dose-response curves as illustrated in Fig. 4. From the Poisson distribution of virus particles attached to cells in a monolayer, the dilution of virus at a 0.63 maximum yield of IFN represents the fraction of the cell population, 0.37, that by chance has not received a virus particle and hence has not been induced. This situation pertains when the multiplicity of IFN-inducing particles (mIFP) = 1. Since the number of cells in the monolayer is known (107) and virus attachment is essentially 100%, the titer of IFP becomes (mIFP = 1)(107 cells)(virus dilution factor = 36) = IFP/ml. In the example illustrated, the stock virus was calculated to contain 3.6 x 108 IFP/ml. The same stock was determined to contain 2.2 x 107 infectious particles (IP)/ml (from an EID50 assay). Thus, the IFP/IP ratio = 16. This indicates that about 95% of the particles of this AIV population are IFP and are noninfectious. There is generally about a 10- to 20-fold excess of noninfectious IFP to IP in influenza virus populations with analyzable dose-response curves. Comparable measurements for virus stocks that were weak inducers of IFN were not possible because a full yield of IFN per cell was not realized: i.e., was variable due to the presence of an antagonistic activity expressed by ISP (see below).
ISP activity of influenza viruses. Two of the strains of AIV, A/TKOR/71 and A/PR8/34, induced low levels of IFN (Fig. 1 to 3) and hence were tested for their capacity to suppress IFN induction because of earlier reports that weak or noninducers of IFN could downregulate IFN production (18, 44, 52, 53). Suppression of IFN induction is a dominant phenotype that can be quantified in terms of ISP. Figure 5 demonstrates that influenza virus also can express an ISP phenotype and illustrates a representative assay using influenza virus A/TK/OR/71 as the ISP. Monolayers of CEC were infected with a multiplicity of UV-A/TK/WI/66 sufficiently high to induce maximal yields of IFN from virtually all cells in the monolayer. The absolute yield induced was 105,000 U/107 cells. This value represents the maximum yield of IFN produced in the absence of any ISP, i.e., 1.0 (upper dashed line). The highest level of IFN induced by A/TK/OR/71, the virus being tested for ISP activity (lower dashed line), was 600 U/107 cells, i.e., 0.006 of the maximum IFN yield induced by the IFP UV-A/TK/WI/66. The test curve (solid line) shows the yield of IFN from CEC monolayers simultaneously infected with the multiplicity of UV-A/TK/WI/66 that by itself induces a maximal yield of IFN, along with increasing multiplicities of A/TK/OR/71, the putative ISP. There is an initial steep decline in the production of IFN at low multiplicities of A/TK/OR/71, which resulted in an 95% suppression of IFN yield. About 5% of the IFN produced was 25-fold more resistant to suppression, as deduced from the final slope of the survival curve for IFN yield.
The titer of A/TK/OR/71 ISP was calculated from the initial slope of the curve which represents the fraction of surviving yield of IFN (52). Based on a Poisson distribution of ISP in the cell monolayer, as in the calculation for IFP described above, the fraction of the IFN yield that is 0.37 of the maximum is presumed to contain, on average, 1 ISP. The reciprocal of the virus dilution that resulted in 0.37 survival of the control yield of IFN was 200. Thus, the ISP titer = (1)(1 x 107)(200) = 2 x 109 ISP/ml. Since this stock of virus contained 4 x 107 PFP/ml, the ratio of ISP to PFP = 50, indicating a large excess of ISP to PFP and that influenza virus particles need not be infectious to be scored as ISP. Comparable results were obtained when the IFN-inducing virus was UV-ARV, demonstrating that the IFN-inducing capacity of both homologous and heterologous virus species could be suppressed by ISP of influenza virus origin. This suggests a common action of ISP on the host cell independent of the nature of the inducing virus (53). A/PR/8/34 acted similarly as an ISP. The actual ratio of ISP to PFP may be lower if the plaquing efficiency of AIV strains that require trypsin to activate the fusion protein is not optimal.
The IFN-inducing capacity of UV-irradiated influenza virus. UV-irradiated influenza virus has been shown to induce IFN (7-9, 38). These observations were extended to generate a profile of the IFN-inducing capacity of eight different AIV strains as a function of UV dose in order to gain insight into the size of the UV target(s) involved in converting a weak inducer or noninducer into a good inducer or a good inducer into a better inducer. In the case of influenza virus, each gene segment has been shown to be independently transcribed, with the effective size of the UV target proportional to the number of nucleotides in the gene. Thus, the longer the length of the sequence, the larger the UV target and the lower the dose required to inactivate it (1, 6, 66, 74).
Virus stocks were irradiated with various UV doses, and their IFN-inducing capacity was determined at a multiplicity of virus comparable to that of the active virus, which induced peak or plateau yields of IFN (cf. Fig. 2). The IFN yield recorded at time zero represents the maximum amount of IFN induced by the unirradiated virus. Figure 6 reveals a qualitatively uniform response of AIV upon exposure to UV radiation irrespective of the intrinsic IFN-inducing capacity of the active virus. The viruses displayed three phases as a function of increasing UV dose with respect to IFN-inducing capacity: (i) an initial rapid increase, (ii) a peak yield, and (iii) a marked decline. Data from the parent and corresponding strains with the truncated (Fig. 6C) and deleted (Fig. 6D) NS1 genes are plotted on the same scale for ease of comparison. The increase and then decline in the IFN-inducing capacity observed for all strains of AIV following UV radiation confirm and extend the original observations (8, 9, 38).
The histogram in Fig. 7 shows the maximum yields of IFN produced by the eight strains of AIV as active virus and following UV irradiation. Also included are the number of lethal UV hits to the AIV genome required to reach peak IFN-inducing capacity. They are characteristic of each strain.
The histogram in Fig. 8 provides a direct comparison of the maximal IFN-inducing capacity of the two strains of AIV that express genetically compromised NS1 and their NS gene-competent counterparts, both as active and UV-irradiated virus. Note that parental strains require more UV hits to achieve maximum IFN-inducing capacity than do the NS1-compromised strains.
IFN induction dose-response curves induced by UV-irradiated influenza virus. Several stocks of influenza virus were UV irradiated to achieve maximal IFN-inducing capacity and then used to generate IFN induction dose-response curves. The curves were similar to those shown in Fig. 2 but displayed maximal yields of IFN in keeping with the enhanced IFN-inducing capacity acquired following irradiation, and hence the data are not shown. Suffice it to note that the concentration of IFP calculated from these curves was comparable to those found in active virus. This indicates that the increased IFN-inducing capacity of UV-irradiated virus reflects an increase in the quantum yield of IFN that an IFP can induce rather than the conversion of more particles to IFN-inducing status.
The IFN-inducing capacity of heat-inactivated influenza virus. Figure 9 illustrates the survival curves for both IFN-inducing capacity and PFP activity as a function of time at 50°C for A/TK/OR/71—a relatively weak inducer of IFN that phenotypically functions as an ISP (Fig. 5). Exposure to 50.0°C generated an exponential loss of infectivity (PFP) of –10-fold in 12 min. In contrast, the IFN-inducing capacity of the heated virus increased almost twofold and maintained that capacity following at least 24 min of exposure to heat. These results essentially confirm those that led to the discovery of IFN (26). Heat inactivation in the borate buffer used by Isaacs and Lindenmann (26) produced virus with IFN-inducing capacity comparable to that obtained with the NCI solution we used equilibrated in air. The order of magnitude differences in the yields of IFN reported originally and those recorded here are attributed in part to the greatly enhanced production of IFN from developmentally aged CEC, with 40.5°C as an optimal temperature for IFN induction/production, the enhanced sensitivity of current IFN assays, and the IFN-inducing capacity intrinsic to each strain (cf. Fig. 3).
DISCUSSION
The induction and/or production of IFN by eight different strains of influenza virus and its suppression by two of those strains were examined under conditions where developmentally aged primary CEC hyperproduce IFN when induced (70), the threshold of induction is a single molecule of dsRNA per cell (45, 50), single-stranded RNA (ssRNA) does not induce IFN (45, 50), and virus populations can be quantified for their content of both IFP (46) and ISP (52).
The production of IFN by active AIV generally displayed a 2- to 4-h lag followed by a steady increase in accumulation of secreted IFN. Maximal yields were observed after about 20 ± 5 h postinfection of cells incubated at 40.5°C, the optimal temperature for production of ChIFN- (73). The eight AIV strains revealed a 100-fold range for the maximum accumulated yield of IFN (85 to 8,750 U/107 cells). This compares with a 10,000-fold variation observed in the IFN-inducing capacity of scores of different isolates of vesicular stomatitis virus (VSV), reflecting the quasispecies nature of this RNA virus and the plasticity of the induction arm of the IFN system (48, 55, 58).
Analysis of IFP and ISP activity requires that the input multiplicity be preserved, especially at low multiplicities where the early release of newly produced virus might spuriously increase the multiplicity of infection. In this context, egg-derived AIV contains activated fusion peptide, but CEC-derived virions do not (35). Following IFN induction, newly generated progeny virus will be fusion incompetent and noninfectious, thus preserving the input multiplicity. The TK/ONT and WSN strains are exceptions in that they replicate in CEC without trypsin to generate the active fusion protein.
Data from IFN induction dose-response curves reveal that influenza virus populations contain a ratio of IFP to PFP of 20:1, indicating that noninfectious IFP appear to make up over 90% of influenza virus populations. Any undercounting of infectious particles would bring the ratio closer to the observed physical/infectious particle ratio of about 11 (12). Other viruses have been reported to produce an excess of noninfectious IFP to PFP. These include Newcastle disease virus (7:1) (56), VSV (10:1) (51), and ARV (60:1) (85).
The detection of noninfectious IFP is of interest. Although influenza virus gene segments have signals that select them for packaging (17), this mechanism does not guarantee that all eight unique segments will be packaged, nor does it exclude the packaging of multiple copies of some genes at the cost of precluding one or more of the unique eight required for infectivity (14, 15). Additional evidence, including the observed ratio of physical to infectious particles (12), the prediction from bin theory that the packaging of 12 gene segments per virion would ensure that about 10% of the virus particles contained the unique eight genes for infectivity (15, 71), the generation of defective-interfering particles (63), the transient resistance to IFN action that is lost upon repackaging of the genes (71), and the genetic multiplicity reactivation common to orthomyxoviruses (2), supports the view that imperfect packaging may produce an excess of noninfectious virions. Analysis of IFN induction dose-response curves can detect and quantify these particles as IFP (46). The high ratio of noninfectious IFP to PFP intrinsic to populations of active AIV, coupled with an even greater capacity to produce IFN following UV irradiation, and to a lesser degree after heat inactivation, demonstrates that virus replication is not required for IFN induction by AIV. Furthermore, since the polymerase of influenza virus is labile at 50°C, even in the presence of viral RNA promoter sequences (5; G. G. Brownlee, University of Oxford, Oxford, United Kingdom, personal communication), the IFN inducer moiety of influenza virus appears extant in virions prior to heat inactivation. We infer that any ISP activity intrinsic to the virion was inactivated by heat, allowing the expression of the IFP phenotype. However, following UV irradiation, it is possible that limited primary viral transcription from the 3' end of any of the gene segments packaged into noninfectious IFP virions suffices to generate dsRNA as the IFN inducer moiety within the cell. This is the case for VSV in CEC (51). If a comparable situation prevails in AIV, a higher UV dose would be required to prevent dsRNA production because there is an average of 12 gene segments per virion (15, 71), each independently transcribed (1, 66, 74). In that context, a loss of 90% of the IFN-inducing capacity of VSV (51) and AIV (Fig. 6) was observed at about 20 and 80 lethal UV hits, respectively. Interestingly, strains of AIV with genetically compromised NS genes require many fewer UV hits to render them incapable of inducing IFN. We infer from these data that virions of influenza virus may contain a preformed inducer of IFN—most likely viral dsRNA (45), acquired during virion assembly or produced during primary transcription, if not replication (3, 42, 43). Recent reports of the induction of IFN by ssRNA acting through TLR7 receptors in specialized dendritic cells (11, 24, 41) do not appear applicable to CEC cultures, since the delivery of ssRNA in the form of [-]RNA defective-interfering particles (DIP) over a range of 1 to 80 molecules per cell did not induce IFN, yet 1 [±]RNA molecule did (45, 50). Nor do viral proteins appear to be involved since both types of DIP contained identical polypeptides encoded by the same helper virus, yet differed qualitatively in their IFN inducing capacity (45, 50). Whether the same situation pertains to AIV remains to be determined.
Shortly after the discovery of IFN, Lindenmann reported that live influenza virus prevented the induction of IFN by inactivated virus, an effect he termed "inverse interference" (37). In this context, two of the AIV strains, A/TK/OR/71 and A/PR/8/34, were shown to be efficient suppressors of IFN production, i.e., demonstrated inverse interference. The exponential loss of IFN-inducing capacity as a function of ISP multiplicity indicates that a single ISP per cell suffices to block IFN induction and/or production completely (52). Noninfectious ISP appear to make up over 90% of the virus population. Suppression of IFN induction was demonstrable against both a homologous virus (UV-TK/WI/66) and the dsRNA-containing heterologous virus (UV-ARV), indicating that suppression is directed at the host cell. ISP need not be infectious, and hence expression of the ISP phenotype does not require the eight unique gene segments needed for infectivity. The production of large numbers of ISP during influenza virus infection might block activation of the IFN system and not bode well for the outcome of the disease.
As a function of UV dose, all eight influenza virus strains displayed three distinct phases with respect to their IFN-inducing capacity: (i) an initial rapid increase, followed by (ii) a peak or plateau, followed by (iii) a decline to virtually no production. Although these three phases differed quantitatively for the eight strains, there was a striking common feature: even the lowest doses of UV radiation enhanced IFN-inducing capacity. Importantly, there is essentially no threshold in the amount of UV radiation required to enhance IFN inducibility, meaning that at low doses virtually every UV hit to the genome of the noninfectious IFP improves its IFN-inducing capacity. Eventually, a UV dose is reached that maximizes that capacity—up to 100-fold—and thereafter inactivates it. This statement applies even to the two strains already genetically compromised in NS1 expression. Notably, these strains require fewer UV hits (5) to their genome to achieve maximal IFN-inducing capacity, whereas the parental strains require 30 UV hits.
We infer that UV inactivation of two targets in the AIV genome is required to inhibit IFN induction-suppressing activity and achieve maximal IFN-inducing capacity. A small UV target is attributed to the multifunctional NS gene and its NS1-encoded protein, which may block IFN induction by sequestering dsRNA (4, 19, 40) or the protein kinase PKR (81). In addition, it may prevent the 3'-end processing of cellular pre-mRNA because the effector region sequesters the 30-kDa subunit of cleavage and polyadenylation specificity factor(CPSF) and poly(A)-binding protein II (PABII) (16, 33, 61). The marked increase in the IFN-inducing capacity observed at very low doses of UV radiation seen in all strains, with or without an intact NS gene, implicates a role for a gene(s) with a large target. UV inactivation of the large target in AIV converts a weak or good inducer of IFN into a much better inducer—in effect converting an ISP into an IFP. Both activities reside in noninfectious particles.
To identify the gene(s) that constitute the large UV target, we compared the observed rate at which IFN-inducing capacity is enhanced as a function of UV dose, with the expected fraction of inactivation for different genes or combinations thereof based on a Poisson distribution of UV hits to the AIV population. We reasoned that the rate of increase in the IFN-inducing capacity of the AIV population would be most directly related to the rate of inactivation of the gene(s) that comprises the large UV target. The larger the gene, the fewer UV hits it would take to inactivate it. We sought to identify a gene, or combination of genes, whose expected rate of inactivation would best fit the rate at which UV radiation enhanced IFN-inducing capacity. Only one strain has been analyzed thus far: TK/OR/71(delNS1[1-124]). It was chosen because the small gene target already was compromised genetically, meaning that following UV irradiation, any further increase in IFN-inducing capacity could be relegated to a large gene target. Figure 10 (curve A) shows that the expected rate of inactivation of the NS gene in a population of AIV provides a poor fit to the actual rate at which UV radiation enhances IFN-inducing capacity. A somewhat better fit is observed if a gene for one of the three polymerase subunits is considered to be the UV target (PB1, PB2, or PA, which are close to the same size) (curve B). The fit is much improved if the target is assumed to consist of two or three of the polymerase genes and that a UV hit to any one of the two (curve C) or three (curve D) genes suffices to maximally enhance IFN-inducing capacity. The expected curves for UV inactivation of these genes assume that the AIV population is homogeneous with respect to gene content. However, a noninfectious AIV population may contain virions with only two active polymerase genes, accounting for some of the variable IFN-inducing capacity intrinsic to a strain of AIV (14, 35). We cannot distinguish between two or three polymerase subunits as the likely UV target, but for mechanistic reasons, we propose as a working hypothesis that the effective target is the sum of the three polymerase genes that encode the endonuclease complex responsible for snatching caps from newly transcribed cellular pre-mRNAs. Since all three polymerase subunits are required for cap snatching (36, 65), one UV hit to any of the genes encoding a polymerase subunit would block this process and allow production of translatable IFN mRNA, observed as enhanced IFN production. Strong support for this hypothesis comes from the report that the reduced number of cellular RNA polymerase II-derived transcripts that continue to be synthesized after AIV infection do not appear in the cytoplasm as mature mRNA, most likely because the AIV cap-snatching endonuclease renders the decapped mRNAs susceptible to nucleases (29). This process would compromise IFN production.
In addition, any delay in the posttranscriptional processing of cellular pre-mRNA at the 3' end through sequestering of cellular factors CPSF and PABII by the effector region of NS1 (33) would exacerbate the ISP activity of AIV by restricting pre-IFN mRNA to the nucleus, thereby increasing exposure to the cap-snatching endonuclease and further degradation by nucleases (16, 61, 68).
By this model, UV irradiation inhibits the dominant ISP phenotype by inactivating two targets: (i) a large target which consists of all three of the polymerase genes; and (ii) a small target, the NS gene. The important role of NS1A as a key defense mechanism of AIV may be gleaned from the preferential transcription of the NS gene in chicken cells and the appearance of NS1 in the cell within 1 h after infection, ensuring a rapid first response against the IFN system (74). The cumulative effect of inactivating both anti-IFN induction and production mechanisms would be to maximize expression of the IFP phenotype.
At very high UV doses, there is a marked decline in AIV IFN-inducing capacity. Since the hemagglutinating activity of AIV is not diminished by the highest UV doses used, virus attachment and entry into the cell are not likely to be affected, although further processing might be. Most strains require UV doses of about 8,000 ergs/mm2 (100 lethal hits) to achieve loss of virtually all IFN-inducing capacity. Kim and colleagues reported that a dose of UV which delivered about 90 lethal hits (our calculation) to an AIV strain prevented the activation of IRF-3 (30). This high dose of UV may prevent formation of an IFN inducer (dsRNA?) as noted above, and in turn the activation of NF-B (34) and IRF-3 (80).
If dsRNA preexists in AIV, as in the reovirus genome (85), or as a molecule of covalently linked self-complementary [±]RNA in some kinds of VSV defective-interfering particles (45, 47, 50), IFN-induction will not be compromised by UV irradiation or heat (45, 50). However, if dsRNA must be generated anew within the cell by the inducing virus, then heat or high doses of UV radiation will inactivate their IFN-inducing capacity (51). When influenza virus infectivity is inactivated by heat, some strains gain the capacity to induce IFN, providing the experimental basis for the discovery of IFN (26). As first described by Burke and Buchan (7) and reported in detail here, UV irradiation invariably initially enhances the IFN-inducing capacity of AIV. The enhanced IFN-inducing capacity observed in all of the influenza virus strains subjected to UV radiation is consistent with the presence of preformed dsRNA and the inactivation of virus genes represented by the small and the large UV targets. Alternatively, primary transcription by AIV may generate dsRNA which leads to NF-B (34) and IRF-3 activation (30). A model is favored in which low doses of UV irradiation inactivate the ISP activity of AIV, the large UV target, while allowing primary transcripts of sufficient length to produce dsRNA and induce IFN. At high UV doses, the size of the primary transcripts would become rate limiting for dsRNA formation (51), activation of transcription factors, and hence IFN induction (20).
Recent reports show that ssRNA viruses like AIV and VSV, or heat-inactivated AIV, can induce IFN in TLR7-containing cells of dendritic origin when delivered through endocytosis (11, 24, 41). While comparable experiments have not been reported for CEC, we note that in the absence of helper virus, otherwise identical [±] and [–]RNA DIP of VSV in CEC do and do not, respectively, induce IFN. These data show that in CEC the threshold for IFN induction is one molecule of dsRNA and that ssRNA, up to 80 molecules per cell delivered by endocytosis, does not induce IFN (45, 50), strengthening the case that the IFN-inducer moiety in, or generated by, AIV in CEC is dsRNA. Whether helicase (RIG I) (87) is involved in processing such dsRNA is yet to be determined in chicken cells.
AIV are intrinsically sensitive to IFN action (3, 21, 22, 23, 26, 67, 69, 76), most likely through the action of the MX system and inhibition of primary transcription (3, 31, 32). To circumvent this sensitivity, influenza viruses have evolved mechanisms to defeat host cell defenses by blocking sequential events in IFN production, namely, processing of both the 5' and 3' ends of IFN pre-mRNA (16, 33, 84). However, the broad spectrum of IFN-inducing capacities displayed by different AIV strains and differences in the extent to which UV irradiation and heat affect that capacity demonstrate that prevention of IFN induction and/or production is not always successful. Indeed, the IFP and ISP phenotypes may have a profound effect on the outcome of AIV infection, both in mice (13, 21) and in chickens (A. N. Cauthen and D. L. Suarez, personal communication). The potential for IFN to enhance the immune response (28) behooves us to define the IFP/ISP phenotypes of influenza virus strains and to determine whether the excess of noninfectious IFP or ISP produced during virus replication affects the immunobiology of AIV infection (78), pathogenesis of the disease, and the efficacy of attenuated live AIV vaccines.
ACKNOWLEDGMENTS
This research was supported by USDA grant 58-1940-0-007 through the Center for Excellence in Vaccine Research at the University of Connecticut. The study benefited from the use of the Animal Cell Culture Facility of the Biotechnology-Bioservices Center of the University of Connecticut.
We thank Adolfo García-Sastre for pointing out the advantages of using the developmental immaturity of the IFN system in 6- to 7-day-old embryonated eggs (70) to prepare stocks of AIV from strains prone to induce IFN because of genetically compromised expression of NS1A (86). We thank Robert M. Krug for insightful discussions.
This is contribution XXIV in a series entitled "Interferon Induction by Viruses."
Present address: Scripps Research Institute, 10550 N. Torrey Pines Rd., La Jolla, CA 92037
REFERENCES
Abraham, G. 1979. The effect of ultraviolet radiation on the primary transcription of influenza virus messenger RNAs. Virology 97:177-182.
Barry, R. D. 1961. The multiplication of influenza virus. II. Multiplicity reactivation of ultraviolet-irradiated virus. Virology 14:398-405.
Bean, W. J., and R. W. Simpson. 1973. Primary transcription of the influenza virus genome in permissive cells. Virology 56:646-651.
Bergmann, M., A. García-Sastre, E. Carnero, H. Pehamberger, K. Wolff, P. Palese, and T. Muster. 2000. Influenza virus NS1 protein counteracts the PKR-mediated inhibition of replication. J. Virol. 74:6203-6206.
Brownlee, G. G., and J. L. Sharps. 2002. The RNA polymerase of influenza A virus is stabilized by interaction with its viral RNA promoter. J. Virol. 76:7103-7113.
Budowsky, E. I., S. E. Bresler, E. A. Friedman, and N. V. Zheleznova. 1981. Principles of selective inactivation of viral genome. I. UV-induced inactivation of influenza virus. Arch. Virol. 68:239-247.
Burke, D. C., and A. Buchan. 1965. Interferon production in chick embryo cells. I. Production by ultraviolet-inactivated virus. Virology 26:28-35.
Burke, D. C., and A. Isaacs. 1958. Further studies on interferon. Br. J. Exp. Pathol. 39:78-84.
Burke, D. C., and A. Isaacs. 1958. Some factors affecting the production of interferon. Br. J. Exp. Pathol. 39:452-458.
Cox, N. J., and K. Subbarao. 1999. Influenza. Lancet 354:1277-1282.
Diebold, S. S., T. Kaisho, H. Hemmi, S. Akira, and C. R. E. Sousa. 2004. Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA. Science 303:1529-1531.
Donald, H. B., and A. Isaacs. 1954. Counts of influenza virus particles. J. Gen. Microbiol. 10:457-464.
Donelan, N. R., C. F. Basler, and A. García-Sastre. 2003. A recombinant influenza A virus expressing an RNA-binding-defective NS1 protein induces high levels of beta interferon and is attenuated in mice. J. Virol. 77:13257-13266.
Duhaut, S. D., and J. W. Mccauley. 1996. Defective RNAs inhibit the assembly of influenza virus genome segments in a segment-specific manner. Virology 216:326-337.
Enami, H., G. Sharma, C. Benham, and P. Palese. 1991. An influenza virus containing nine different gene segments. Virology 185:291-298.
Enserink, M., and J. Kaiser. 2004. Avian flu finds new mammal hosts. Science 305:1385.
Fortez, P., A. Beloso, and J. Ortín. 1994. Influenza virus NS1 protein inhibits pre-mRNA splicing and blocks RNA nucleocytoplasmic transport. EMBO J. 13:704-712.
Fujii, Y., H. Goto, T. Watanabe, T. Yoshida, and Y. Kawaoka. 2003. Selective incorporation of influenza virus RNA segments into virions. Proc. Natl. Acad. Sci. USA 100:2002-2007.
Gaccione, C., and P. I. Marcus. 1989. Interferon induction by viruses. XVIII. Vesicular stomatitis virus-New Jersey: a single infectious particle can both induce, and suppress, interferon production. J. Interferon Res. 9:603-614.
García-Sastre, A., A. Egorov, D. Matassov, S. Brandt, D. E. Levy, J. E. Durbin, P. Palese, and T. Muster. 1998. Influenza A virus lacking the NS1 gene replicates in interferon-deficient systems. Virology 252:324-330.
Geiss, G. K., M. Salvatore, T. M. Tumpey, V. S. Carter, X. Wang, C. F. Basler, J. K. Taubenberger, R. E. Bumgarner, P. Palese, M. G. Katze, and A. García-Sastre. 2002. Cellular transcriptional profiling in influenza A virus infected lung epithelial cells: role of the nonstructural NS1 protein in the evasion of the host innate defense and its potential contribution to pandemic influenza. Proc. Natl. Acad. Sci. USA 99:10736-10741.
Gresser, I., M. G. Tovey, C. Maury, and M. T. Bandu. 1976. Role of interferon in the pathogenesis of virus diseases in mice as demonstrated by the use of anti-interferon serum. II. Studies with herpes simplex, Maloney sarcoma, vesicular stomatitis, Newcastle disease and influenza viruses. J. Exp. Med. 144:1316-1327.
Gutman, N. R., and L. S. Manakhova. 1981. The interferon-inducing capacity of influenza virus strains and their sensitivity to the effect of exogenous interferon. Vopr. Virusol. 2:168-171.
Haller, O., M. Frese, and G. Kochs. 1998. Mx proteins: mediators of innate resistance to RNA viruses. Rev. Sci. Tech. 17:220-230.
Heil, F., H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, G. Lipford, H. Wagner, and S. Bauer. 2004. Species-specific recognition of single-stranded RNA via toll-like receptor 7 and 8. Science 303:1526-1529.
Hinshaw, V. S., C. W. Olsen, N. Dybdahl-Sissoko, and D. Evans. 1994. Apoptosis: a mechanism of cell killing by influenza A and B viruses. J. Virol. 68:3667-3673.
Isaacs, A., and J. Lindenmann. 1957. Virus interference. I. The interferon. Proc. R. Soc. Ser. B 147:258-267.
Isaacs, A., J. Lindenmann, and R. C. Valentine. 1957. Virus interference. II. Some properties of interferon. Proc. R. Soc. Ser. B 147:268-273.
Julkunen, I., T. Sareneva, J. Pirhonen, T. Ronni, K. Melén, and S. Matikainen. 2001. Molecular pathogenesis of influenza A virus infection and virus-induced regulation of cytokine gene expression. Cytokine Growth Factor Rev. 12:171-180.
Katze, M. G., and R. M. Krug. 1984. Metabolism and expression of RNA polymerase II transcripts in influenza virus-infected cells. Mol. Cell. Biol. 4:2198-2206.
Kim, M.-J., A. G. Latham, and R. M. Krug. 2002. Human influenza viruses activate an interferon-independent transcription of cellular antiviral genes: the outcome with influenza A virus is unique. Proc. Natl. Acad. Sci. USA 99:10096-10101.
Ko, J.-H., H.-K. Jin, A. Asano, A. Takada, A. Ninomiya, H. Kida, H. Hokiyama, M. Ohara, M. Tsuzuki, M. Nishibori, M. Mitzutani, and T. Watanabe. 2002. Polymorphisms and the differential antiviral activity of the chicken Mx gene. Genome Res. 12:595-601.
Krug, R. M., M. Shaw, B. Broni, G. Shapiro, and O. Haller. 1985. Inhibition of influenza viral mRNA synthesis in cells expressing the interferon-induced Mx gene product. J. Virol. 56:201-206.
Krug, R. M., W. Yuan, D. L. Noah, and A. G. Latham. 2003. Intracellular warfare between human influenza viruses and human cells: the roles of the viral NS1 protein. Virology 309:181-189.
Kumar, A., J. Haque, J. Lacoste, J. Hiscott, and B. R. G. Williams. 1994. Double-stranded RNA-dependent protein kinase activates transcription factor NF-B by phosphorylating IB. Proc. Natl. Acad. Sci. USA 91:6288-6292.
Lamb, R. A., and R. M. Krug. 2001. Orthomyxoviridae: the viruses and their replication, p. 725-769. In D. M. Knipe and P. M. Howley (ed.), Fundamental virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Li, M. L., P. Rao, and R. M. Krug. 2001. The active sites of the influenza cap-dependent endonuclease are on different polymerase subunits. EMBO J. 20:2078-2086.
Lindenmann, J. 1960. Interferon und inverse interferenz. Z. Hyg. Infectionskr. Med. Mikrobiol. Immunol. Virol. 146:287-309.
Lindenmann, J., D. C. Burke, and A. Isaacs. 1957. Studies on the production, mode of action and properties of interferon. Br. J. Exp. Pathol. 38:551-562.
Lipatov, A. S., E. A. Govorkova, R. J. Webby, H. Ozaki, M. Peiris, Y. Guan, L. Poon, and R. G. Webster. 2004. Influenza: emergence and control. J. Virol. 78:8951-8959.
Lu, Y., M. Wambach, M. G. Katze, and R. M. Krug. 1995. Binding of the influenza virus NS1 protein to double-stranded RNA inhibits the activation of the protein kinase that phosphorylates the eIF-2 translation initiation factor. Virology 214:222-228.
Lund, J. M., L. Alexopoulou, A. Sato, M. Karaow, N. C. Adams, N. W. Gale, A. Iwasaki, and R. A. Flavell. 2004. Recognition of single-stranded RNA viruses by toll-like receptor 7. Proc. Natl. Acad. Sci. USA 101:5598-5603.
Majde, J. A. 2000. Viral double-stranded RNA, cytokines, and the flu. J. Interferon Cytokine Res. 20:259-272.
Majde, J. A., N. Guha-Thakurta, Z. Chen, S. Bredow, and J. M. Krueger. 1998. Spontaneous release of stable viral double-stranded RNA into the extracellular medium by influenza virus-infected MDCK endothelial cells: implications for the viral acute-phase response. Arch. Virol. 143:2371-2380.
Marcus, P. I. 1982. Interferon induction by viruses. IX. Antagonistic activities of virus particles modulate interferon production. J. Interferon Res. 2:511-518.
Marcus, P. I. 1983. Interferon induction by viruses: one molecule of dsRNA as the threshold for interferon induction, p. 115-180. In I. Gresser (ed.), Interferon 5, Academic Press, New York, N.Y.
Marcus, P. I. 1986. Interferon induction: dose-response curves. Methods Enzymol. 119:106-114.
Marcus, P. I., and C. Gaccione. 1989. Interferon induction by viruses. XIX. Vesicular stomatitis virus-New Jersey: high multiplicity passages generate interferon-inducing, defective-interfering particles. Virology 171:630-633.
Marcus, P. I., L. L. Rodriguez, and M. J. Sekellick. 1998. Interferon induction as a quasispecies marker of vesicular stomatitis virus populations. J. Virol. 72:542-549.
Marcus, P. I., and M. J. Sekellick. 1975. Cell killing by viruses. II. Cell killing by vesicular stomatitis virus: a requirement for virion-derived transcription. Virology 63:176-190.
Marcus, P. I., and M. J. Sekellick. 1977. Defective-interfering particles with covalently-linked [±] RNA induce interferon. Nature 266:815-819.
Marcus, P. I., and M. J. Sekellick. 1980. Interferon induction by viruses. III. Vesicular stomatitis virus: interferon-inducing particle activity requires partial transcription of gene N. J. Gen. Virol. 47:89-96.
Marcus, P. I., and M. J. Sekellick. 1985. Interferon induction by viruses. XIII. Detection and assay of interferon induction-suppressing particles. Virology 142:411-415.
Marcus, P. I., and M. J. Sekellick. 1987. Interferon induction by viruses. XV. Biological characteristics of interferon induction-suppressing particles of vesicular stomatitis virus. J. Interferon Res. 7:269-284.
Marcus, P. I., and M. J. Sekellick. 2001. Combined sequential treatment with interferon and dsRNA abrogates virus resistance to interferon action. J. Interferon Cytokine Res. 21:423-427.
Marcus, P. I., M. J. Sekellick, and S. T. Nichol. 1992. Interferon induction by viruses. XXI. Vesicular stomatitis virus: interferon inducibility as phylogenetic marker. J. Interferon Res. 12:297-305.
Marcus, P. I., C. Svitlik, and M. J. Sekellick. 1983. Interferon induction by viruses. X. A model for interferon induction by Newcastle disease virus. J. Gen. Virol. 64:2419-2431.
Marcus, P. I., L. van der Heide, and M. J. Sekellick. 1999. Chicken interferon action on avian viruses. I. Oral administration of chicken interferon- ameliorates Newcastle disease. J. Interferon Cytokine Res. 19:881-885.
Mattana, P., and G. C. Viscomi. 1998. Variations in the IFN-inducing capacity of Sendai virus subpopulations. J. Interferon Cytokine Res. 18:399-405.
Meegan, J. M., and P. I. Marcus. 1989. Double-stranded ribonuclease coinduced with interferon. Science 244:1089-1091.
Mo, C. W., Y. C. Cao, and B. L. Lim. 2001. The in vivo and in vitro effects of chicken interferon alpha on infectious bursal disease virus and Newcastle disease virus infection. Avian Dis. 45:389-399.
Noah, D. L., K. Y. Twu, and R. M. Krug. 2003. Cellular antiviral responses against influenza A virus are countered at the postranscriptional level by the viral NS1A protein via its binding to a cellular protein required for the 3' end processing of cellular pre-mRNAs. Virology 307:386-395.
Norton, G. P., T. Tanaka, K. Tobita, S. Nakada, D. A. Buonagurio, D. A. Greenspan, M. Krystal, and P. Palese. 1987. Infectious influenza A and B virus variants with long carboxyl terminal deletions in the NS1 polypeptides. Virology 156:204-213.
Odagiri, T., and M. Tashiro. 1997. Segment-specific noncoding sequences of the influenza virus genome RNA are involved in the specific competition between defective interfering RNA and its progenitor RNA segment at the virion assembly step. J. Virol. 71:2138-2145.
Pei, J., M. J. Sekellick, P. I. Marcus, I.-S. Choi, and E. W. Collisson. 2001. Chicken interferon type I inhibits infectious bronchitis virus (IBV) replication and associated respiratory illness. J. Interferon Cytokine Res. 21:1071-1077.
Plotch, S. J., M. Bouloy, I. Ulmanen, and R. M. Krug. 1981. A unique cap (m7GpppXm)-dependent influenza virion endonuclease cleaves capped RNAs to generate the primers that initiate viral RNA transcription. Cell 23:847-858.
Pons, M. W., and O. M. Rochovansky. 1979. Ultraviolet inactivation of influenza virus RNA in vitro and in vivo. Virology 97:183-189.
Portnoy, J., and T. C. Merigan. 1971. The effect of interferon and interferon inducers on avian influenza. J. Infect. Dis. 124:545-552.
Qiu, Y., and R. M. Krug. 1994. The influenza virus NS1 protein is a poly(A)-binding protein that inhibits nuclear export of mRNAs containing poly(A). J. Virol. 68:2425-2432.
Richman, D. D., B. R. Murphy, S. Baron, and C. Uhlendorf. 1976. Three strains of influenza A virus (H3N2): interferon sensitivity in vitro and interferon production in volunteers. J. Clin. Microbiol. 3:223-226.
Sekellick, M. J., W. J. Biggers, and P. I. Marcus. 1990. Development of the interferon system. I. In chicken cells development in ovo continues on time in vitro. In Vitro Cell. Dev. Biol. 26:997-1003.
Sekellick, M. J., S. A. Carra, A. Bowman, D. A. Hopkins, and P. I. Marcus. 2000. Transient resistance of influenza virus to interferon action attributed to random multiple packaging and activity of NS genes. J. Interferon Cytokine Res. 20:963-970.
Sekellick, M. J., A. F. Ferandino, D. A. Hopkins, and P. I. Marcus. 1994. Chicken interferon gene: cloning, expression, and analysis. J. Interferon Res. 13:413-418.
Sekellick, M. J., and P. I. Marcus. 1986. Induction of high titer chicken interferon. Methods Enzymol. 119:115-125.
Smith, G. L., and A. J. Hay. 1982. Replication of the influenza virus genome. Virology 118:96-108.
Staeheli, P., F. Puehler, K. Schneider, T. W. G?bel, and B. Kaspers. 2001. Cytokines of birds: conserved functions—a largely different look. J. Interferon Cytokine Res. 21:993-1010.
Stewart, W., II, E. Declercq, A. Billiau, J. Desmyter, and P. DeSomer. 1972. Increased susceptibility of cells treated with interferon to the toxicity of polyriboinosinic acid-polyribocytidylic acid. Proc. Natl. Acad. Sci. USA 69:1851-1854.
Suarez, D. L., and M. L. Perdue. 1998. Multiple alignment comparison of the non-structural genes of influenza A viruses. Virus Res. 54:59-69.
Suarez, D. L., and S. Schultz-Cherry. 2000. Immunobiology of avian influenza virus: a review. Dev. Comp. Immunol. 24:269-283.
Takizawa, T., S. Matsukkawa, Y. Higuchi, S. Nakamura, Y. Nakanishi, and R. Fukuda. 1993. Induction of programmed cell death (apoptosis) by influenza virus infection in tissue culture cells. J. Gen. Virol. 74:2347-2355.
Talon, J., C. M. Horvath, R. Polley, C. F. Basler, T. Muster, P. Palese, and A. García-Sastre. 2000. Activation of interferon regulatory factor 3 is inhibited by the influenza A virus NS1 protein. J. Virol. 74:7989-7996.
Tan, S. L., and M. G. Katze. 1998. Biochemical and genetic evidence for complex formation between the influenza A virus NS1 protein and the interferon-induced PKR protein kinase. J. Interferon Cytokine Res. 18:757-766.
Tanaka, N., M. Sato, M. S. Lamphier, H. Nozawa, E. Oda, S. Noguchi, R. D. Schreiber, Y. Tsujimoto, and T. Taniguchi. 1998. Type I interferons are essential mediators of apoptotic death in virally infected cells. Genes Cells 3:29-37.
Van Campen, H., B. C. Easterday, and V. S. Hinshaw. 1989. Virulent influenza A viruses: their effect on avian lymphocytes and macrophages. J. Gen. Virol. 70:2887-2895.
Wang, W., and R. M. Krug. 1996. The RNA-binding and effector domains of the viral NS1 protein are conserved to different extents among influenza A and B viruses. Virology 223:41-50.
Winship, T. R., and P. I. Marcus. 1980. Interferon induction by viruses. VI. Reovirus virion genome dsRNA as the interferon inducer in aged chicken embryo cells. J. Interferon Res. 1:155-167.
Yewdell, J., and A. García-Sastre. 2002. Influenza still surprises. Curr. Opin. Microbiol. 5:414-418.
Yoneyama, M., M. Kikuchi, T. Natsukawa, N. Shinobu, T. Imaizumi, M. Miyagishi, K. Taira, S. Akira, and T. Fujita. 2004. The RNA helicase RIG-I had an essential function in double-stranded RNA-induced innate antiviral responses. Nat. Immunol. 5:730-737.(Philip I. Marcus, Jillian)
Department of Pathobiology and Veterinary Science
Center for Excellence in Vaccine Research, University of Connecticut, Storrs, Connecticut
ABSTRACT
Developmentally aged chicken embryo cells which hyperproduce interferon (IFN) when induced were used to quantify IFN production and its suppression by eight strains of type A influenza viruses (AIV). Over 90% of the IFN-inducing or IFN induction-suppressing activity of AIV populations resided in noninfectious particles. The IFN-inducer moiety of AIV appears to preexist in, or be generated by, virions termed IFN-inducing particles (IFP) and was detectable under conditions in which a single molecule of double-stranded RNA introduced into a cell via endocytosis induced IFN, whereas single-stranded RNA did not. Some AIV strains suppressed IFN production, an activity that resided in a noninfectious virion termed an IFN induction-suppressing particle (ISP). The ISP phenotype was dominant over the IFP phenotype. Strains of AIV varied 100-fold in their capacity to induce IFN. AIV genetically compromised in NS1 expression induced about 20 times more IFN than NS1-competent parental strains. UV irradiation further enhanced the IFN-inducing capacity of AIV up to 100-fold, converting ISP into IFP and IFP into more efficient IFP. AIV is known to prevent IFN induction and/or production by expressing NS1 from a small UV target (gene NS). Evidence is presented for an additional downregulator of IFN production, identified as a large UV target postulated to consist of AIV polymerase genes PB1 + PB2 + PA, through the ensuing action of their cap-snatching endonuclease on pre-IFN-mRNA. The products of both the small and large UV targets act in concert to regulate IFN induction and/or production. Knowledge of the IFP/ISP phenotype may be useful in the development of attenuated AIV strains that maximally induce cytokines favorable to the immune response.
INTRODUCTION
Almost 50 years ago Alick Isaacs and Jean Lindenmann exposed fragments of chorioallantoic membranes from chicken eggs to a strain of influenza virus A (AIV) that had been inactivated by heat (26). This treatment rendered the virus noninfectious, yet capable of stimulating cells to produce a secreted factor which, when added to fresh cells for 24 h, interfered with the replication of an active preparation of that same virus. With this experiment, they discovered the active component of the most common type of viral interference and aptly termed it interferon (IFN). Many of the basic properties of IFN and its induction and/or production in chicken cells soon were established in Isaacs' laboratory (7, 8, 27, 38). These seminal studies launched a new era in our understanding of viral interference, representing as they did the first report of the antiviral activity of an inducible, secreted, and isolable cytokine.
The intrinsic sensitivity of influenza virus to the antiviral action of IFN in vitro in both chicken and mammalian cells has been established (3, 22, 23, 67, 69, 71). This provides further impetus to determine the full complement of viral genes and gene products that regulate the capacity of AIV to induce and/or produce IFN, a first step in the activation of the IFN action pathways. The critical role of the IFN system during influenza virus infection was first demonstrated in a classical experiment in which mice infected with influenza virus and injected with anti-IFN serum were observed to produce higher titers of virus and an exacerbated pathology (21). Subsequent studies with knockout mice defective in the IFN system have confirmed this basic observation. Recent studies in which the IFN-inducing capacity varied in otherwise genetically related AIV showed that the strain of AIV capable of producing the most IFN was attenuated in its pathogenicity in mice (13) and in chickens (A. N. Cathuen and D. Suarez, Southeast Poultry Research Laboratory, Athens, Ga.; personal communication).
Although type A influenza viruses (AIV) are intrinsically sensitive to the action of the IFN they induce, the virus continues to expand its propensity to cause pandemics and epornitics of major public health concern. The perennial worldwide appearance of influenza as an infectious disease attests to its potential to decimate both human and avian populations. Consider that up to 47,000 deaths are attributed to influenza in the U.S. each year (10), coupled with the unprecedented deaths of humans in Asia from H5N1 avian-derived influenza virus and the extermination of more than 100 million fowl (39). And now, the emergence of a pathogenic species of the H5N1 avian virus that kills waterfowl—heretofore considered naturally resistant hosts (39)—and felines (15a) alerts us to the increasingly broader host range for lethality acquired by influenza virus. With chickens the largest source of animal protein, the periodic depopulation of millions of chickens as a means of containing AIV-based epornitics adds to the urgency to understand the role of the IFN system in regulating the pathogenicity of this virus—especially since IFN acts independently of viral antigenic variation, can enhance the immune response (reviewed in reference 75), and may be effective against several avian viruses (57, 60, 64).
We have revisited the influenza virus-chicken cell system to address biological features of IFN induction and/or production by AIV in light of new advances in our understanding of the molecular features of this important pathogen (33, 86). This report examines eight strains of AIV for two biological attributes not previously quantified for populations of influenza viruses, namely their content of virions that (i) induce IFN (IFN-inducing particles [IFP]) (46), or (ii) suppress the induction of IFN in cells otherwise programmed to produce it (IFN induction-suppressing particles [ISP]) (52). These two antagonistic phenotypes (44) were measured in populations of AIV that were (i) active, (ii) genetically compromised for NS1 expression, (iii) UV inactivated, or (iv) heat inactivated. The nature of the molecule inducing IFN in the AIV-chicken cell systems is examined, and evidence is presented for a new means by which AIV downregulates IFN induction/production.
(This work was presented in part at the 23rd Meeting of the American Society for Virology, Montreal, Quebec, Canada, 13 July 2004 [ASV Abstr., p. 200, 2004].)
MATERIALS AND METHODS
Cells and media. The preparation and developmental aging of primary chicken embryo cells (CEC) prepared from 9-day-old embryonated eggs (Charles River SPAFAS, Inc., North Franklin, Conn.) has been described (70, 73). Briefly, confluent monolayers of primary chicken embryo cells were established overnight and "developmentally aged" for 7 to 10 days: i.e., incubated at 37.5°C in NCI medium plus 6% calf serum (attachment solution) without a medium change. When appropriately induced and incubated at 40.5°C, these cells will produce levels of IFN many-fold higher than nonaged cells (73). Chick cells aged 4 to 5 days were used for the IFN assay.
Viruses: source, preparation, and assay. Colleagues kindly provided seed stocks of the following AIV: of avian origin; A/TK/ONT/7732/66 (H5N9) (Virginia Hinshaw, University of California, Davis) (83); A/ONT/7732/66(Clone 1B) (H5N9), which was plaque derived from the former stock after one passage on CEC treated with 50 U of recombinant chicken alpha IFN (rChIFN-) per ml for 24 h (71); A/TK/WI/66 (H9N2) (Theodore Girshick, Charles River SPAFAS, Inc.); A/TK/OR/71(H7N3); and A/TK/OR/71(delNS1[1-124]), which contains the DSR-binding region but with deletion of amino acids 125 to 230 and produces NS1 protein lacking the effector region at the C terminus (D. Suarez, Southeast Poultry Research Laboratory) (62, 77). The seed stocks of AIV of human origin were A/WSN/33 (H1N1), A/PR/8/34 (H1N1), and A/PR/8/34/delNS1 (H1N1) (A. García-Sastre and P. Palese, Mount Sinai School of Medicine, New York, N.Y.). The delNS1 strain has a deletion in the NS gene which eliminates all of the NS1 open reading frame except for the first 10 amino acids, which are shared with the viral nuclear export protein NS2 (19); thus, it lacks both the DSR-binding and effector regions. For convenience, this PR/8 strain is termed NS1 to distinguish it from the strain of A/TK/OR delNS1 with the truncated C terminus lacking amino acids 125 to 230 (77).
All stocks of influenza virus were grown in the amniotic/allantoic membranes of 7-to 9-day-old embryonated chicken eggs from specific-pathogen-free flocks from Charles River SPAFAS, Inc.. Each egg was injected with 0.1 ml containing 103 infectious particles, incubated for 48 to 72 h at 34°C with forced air circulation and egg rotation, and held at 4°C for 24 h before harvesting. The amniotic/chorioallantoic fluid was harvested separately from each egg. Those with high hemagglutinating activity were pooled, and aliquots were prepared and stored at –80°C.
Strain A/TK/WI/66 stocks were assayed as 50% egg infective doses (EID50) (in units per milliliter) by Theodore Girshick of Charles River SPAFAS, Inc. Infectivities of A/TK/ONT/7732/66 and WSN/33 stocks were measured by plaque assay in the absence of trypsin. All other strains required washing the monolayers twice with serum-free medium followed by 1.4 to 1.6 μg of trypsin per ml (code T 1426; Sigma-Aldrich, Co., St. Louis, Mo.) in the agarose overlay to achieve consistent plaque formation. Serum was omitted from the overlay during all trypsin-based assays.
IFN induction and assay. Detailed protocols have been described for the induction and assay of chicken IFN in primary CEC (73). UV-irradiated avian reovirus (UV-ARV) was used as a standard IFN inducer (85). It induced 10,000 U/107 aged CEC as determined from over 50 assays. This inducer was used as a standard to assess the intrinsic capacity of each batch of developmentally aged primary CEC to produce IFN. No basal levels of IFN were ever detected in noninduced cells (<1 U/ml.) Aliquots of a partially purified natural type I ChIFN (72) were prepared and used as a standard to assess the sensitivity of each batch of primary CEC to the action of IFN (73) and were usually assayed at 8,000 U/ml. For each batch of primary CEC, the intrinsic capacity to produce IFN and to respond to its action were normalized to these two standards. All IFN inductions started with a 1-h attachment of virus in 300 μl of attachment solution (NCI medium plus 6% calf serum) to a monolayer of 107 primary CEC that had been aged in vitro for 7 to 10 days in 50-mm-diameter petri dishes. Following incubation at 40.5°C, the 3 ml of medium bathing the cells was removed and processed for acid-stable, type I IFN and assayed in a 96-well system as described previously (73).
Assay of IFP and ISP. The assay of IFP was carried out as described earlier by generating and analyzing IFN induction dose (multiplicity)-response (IFN yield) curves (46) under conditions where a single defective-interfering particle which contains 1 molecule of self-complementary [±]RNA covalently linked in the middle induces a full yield of IFN and a related defective-interfering particle with [–] RNA induces little or no IFN (45, 50). This assay measures the IFN-inducing capacity of a virus particle independent of infectivity. The ISP content of influenza virus stocks was determined as previously reported (52). Briefly, a monolayer of CEC was exposed simultaneously to an IFN-inducing virus at a multiplicity high enough to ensure that all cells produce IFN and at increasing multiplicities of the putative ISP. After 24 h at 40.5°C, the medium was assayed for IFN. The fraction of surviving yield of IFN as a multiplicity of ISP was calculated, and the concentration of ISP in the stock was determined as described in Results.
UV irradiation and heat inactivation. UV irradiation was carried out as previously described (49). Vesicular stomatitis virus (VSV) was used as an actinometer. A survival curve of VSV plaque-forming particle (PFP) activity generated as a function of time exposed to UV radiation (254 nm) revealed an exponential loss of infectivity with D37 = 52.5 ergs/mm2, as measured originally with a Laterjet instrument. For influenza virus, D37 = 82.8 ergs/mm2 as measured by us. This compares with a stated value of 72 ergs/mm2 reported previously (1) and a value of 87 ergs/mm2 calculated by us from the data presented (1). Heat inactivation followed the regimen used by Isaacs and Lindenmann (26). However, 50°C was used rather than 56°C because the rates of inactivation were too rapid at the higher temperature to measure accurately for the strains of virus tested. NaHCO3-buffered NCI solution equilibrated with air was used to achieve pH 8.4. The pH 8.5 borate buffer described by Isaacs and Lindenmann (26) was used with comparable results. The relatively low yields of IFN they reported are attributed in part to the use of chorioallantoic membranes from insufficiently mature embryos and Earle's saline as a rinse—which in our hands usually lowered the yield of IFN.
RESULTS
Time course of IFN production in CEC infected with active influenza viruses. Figure 1 shows the time course of IFN production from developmentally aged primary CEC cultures induced by five strains of active influenza virus, including the contrasting responses generated by parental virus and their genetically related NS1-compromised counterparts. Virus multiplicities were sufficiently high to initially infect all cells in the monolayer with particles capable of inducing IFN. All IFN induction experiments reported in this study were carried out in the absence of trypsin, a protease required for five of the eight strains to sustain plaque formation—the two TK/ONT-derived isolates and WSN being exceptions. Since trypsin-dependent virus strains were capable of inducing IFN, it follows that the egg-derived virus used in these studies contained activated fusion protein and was able to attach to and enter cells (35) to induce IFN. However, the fusion protein of any newly released virus was not cleaved by the primary CEC, thus preventing secondary infection. This condition preserves the input multiplicity and allows analysis of IFN induction dose-response curves as previously reported (46).
Figure 1A is representative of most time course curves generated from good inducers of IFN. IFN was detectable in the medium after a lag of about 2 to 4 h. This lag was followed by a linear increase in the accumulation of IFN in the growth medium. Significant differences were observed in the rate at which maximum yield of IFN was obtained. However, peak yields generally were reached by about 15 to 25 h. Figures 1B and C compare the time course of IFN production by two parental viruses and their genetically related strains with compromised expression of NS1 protein. Figure 1B shows the parent A/TK/OR/71 is a weak inducer of IFN, but its genetically-related strain, A/TK/OR/71(delNS1[1-124]), which produces a C-terminus-truncated NS1 protein, is an excellent inducer of IFN. Figure 1C illustrates a comparable pair of strains, the A/PR/8/34 parent and its genetically derived mutant with the severely ablated NS1 gene. For some strains, the maximal accumulation of IFN was followed by a gradual loss in activity, attributed in part to the release of cellular proteases which accompany the apoptosis-based cytopathic effects we observed at higher multiplicities, and has been reported for influenza viruses (25, 54, 79, 82).
IFN induction dose-response curves generated by different strains of active influenza virus. Figure 2 illustrates as a function of the input concentration of active virus, the yield of IFN produced by monolayers of aged CEC 24 h after induction and incubation at 40.5°C. AIV strains with truncated (Fig. 2A) or deleted (Fig. 2B) NS1 genes are compared with their genetically related parents. These results are in keeping with the compromised replication of NS1-deficient AIV observed in IFN-competent cells (4, 13, 19). Dose-response curves are representative of two to three separate experiments. An IFN dose-response curve typical of the other four strains tested is shown and analyzed in depth in Fig. 4. An inducible double-stranded RNase (dsRNase) was coproduced with the IFN (59; data not shown).
Maximal yields of IFN from different strains of active AIV. Figure 3 shows that the maximal yields of IFN produced by aged CEC monolayers induced with eight different active allantoic fluid-derived stocks of AIV covered a 100-fold range, from about 85 to 8,500 U/107 cells. It documents the differences that characterize their intrinsic IFN-inducing capacity as averaged from three or four independent determinations.
The IFP titer of active influenza virus. In the most common form of the IFN induction dose-response curve, the yield of IFN increases as a function of virus multiplicity until it reaches a plateau. Based on a Poisson distribution of virus particles, the plateau is attained when virtually all cells in the monolayer are infected, i.e., induced to produce a quantum yield of IFN characteristic of that strain (46). The plateau is maintained at higher multiplicities unless the virus has a deleterious effect on the cell, in which case a decline in IFN production may be observed. This may be due to the apoptosis that commonly accompanies influenza virus infection (25, 79) or the action of ISP present at higher multiplicities (see below). The number of particles in a virus population capable of inducing IFN, i.e., IFP, can be calculated from dose-response curves as illustrated in Fig. 4. From the Poisson distribution of virus particles attached to cells in a monolayer, the dilution of virus at a 0.63 maximum yield of IFN represents the fraction of the cell population, 0.37, that by chance has not received a virus particle and hence has not been induced. This situation pertains when the multiplicity of IFN-inducing particles (mIFP) = 1. Since the number of cells in the monolayer is known (107) and virus attachment is essentially 100%, the titer of IFP becomes (mIFP = 1)(107 cells)(virus dilution factor = 36) = IFP/ml. In the example illustrated, the stock virus was calculated to contain 3.6 x 108 IFP/ml. The same stock was determined to contain 2.2 x 107 infectious particles (IP)/ml (from an EID50 assay). Thus, the IFP/IP ratio = 16. This indicates that about 95% of the particles of this AIV population are IFP and are noninfectious. There is generally about a 10- to 20-fold excess of noninfectious IFP to IP in influenza virus populations with analyzable dose-response curves. Comparable measurements for virus stocks that were weak inducers of IFN were not possible because a full yield of IFN per cell was not realized: i.e., was variable due to the presence of an antagonistic activity expressed by ISP (see below).
ISP activity of influenza viruses. Two of the strains of AIV, A/TKOR/71 and A/PR8/34, induced low levels of IFN (Fig. 1 to 3) and hence were tested for their capacity to suppress IFN induction because of earlier reports that weak or noninducers of IFN could downregulate IFN production (18, 44, 52, 53). Suppression of IFN induction is a dominant phenotype that can be quantified in terms of ISP. Figure 5 demonstrates that influenza virus also can express an ISP phenotype and illustrates a representative assay using influenza virus A/TK/OR/71 as the ISP. Monolayers of CEC were infected with a multiplicity of UV-A/TK/WI/66 sufficiently high to induce maximal yields of IFN from virtually all cells in the monolayer. The absolute yield induced was 105,000 U/107 cells. This value represents the maximum yield of IFN produced in the absence of any ISP, i.e., 1.0 (upper dashed line). The highest level of IFN induced by A/TK/OR/71, the virus being tested for ISP activity (lower dashed line), was 600 U/107 cells, i.e., 0.006 of the maximum IFN yield induced by the IFP UV-A/TK/WI/66. The test curve (solid line) shows the yield of IFN from CEC monolayers simultaneously infected with the multiplicity of UV-A/TK/WI/66 that by itself induces a maximal yield of IFN, along with increasing multiplicities of A/TK/OR/71, the putative ISP. There is an initial steep decline in the production of IFN at low multiplicities of A/TK/OR/71, which resulted in an 95% suppression of IFN yield. About 5% of the IFN produced was 25-fold more resistant to suppression, as deduced from the final slope of the survival curve for IFN yield.
The titer of A/TK/OR/71 ISP was calculated from the initial slope of the curve which represents the fraction of surviving yield of IFN (52). Based on a Poisson distribution of ISP in the cell monolayer, as in the calculation for IFP described above, the fraction of the IFN yield that is 0.37 of the maximum is presumed to contain, on average, 1 ISP. The reciprocal of the virus dilution that resulted in 0.37 survival of the control yield of IFN was 200. Thus, the ISP titer = (1)(1 x 107)(200) = 2 x 109 ISP/ml. Since this stock of virus contained 4 x 107 PFP/ml, the ratio of ISP to PFP = 50, indicating a large excess of ISP to PFP and that influenza virus particles need not be infectious to be scored as ISP. Comparable results were obtained when the IFN-inducing virus was UV-ARV, demonstrating that the IFN-inducing capacity of both homologous and heterologous virus species could be suppressed by ISP of influenza virus origin. This suggests a common action of ISP on the host cell independent of the nature of the inducing virus (53). A/PR/8/34 acted similarly as an ISP. The actual ratio of ISP to PFP may be lower if the plaquing efficiency of AIV strains that require trypsin to activate the fusion protein is not optimal.
The IFN-inducing capacity of UV-irradiated influenza virus. UV-irradiated influenza virus has been shown to induce IFN (7-9, 38). These observations were extended to generate a profile of the IFN-inducing capacity of eight different AIV strains as a function of UV dose in order to gain insight into the size of the UV target(s) involved in converting a weak inducer or noninducer into a good inducer or a good inducer into a better inducer. In the case of influenza virus, each gene segment has been shown to be independently transcribed, with the effective size of the UV target proportional to the number of nucleotides in the gene. Thus, the longer the length of the sequence, the larger the UV target and the lower the dose required to inactivate it (1, 6, 66, 74).
Virus stocks were irradiated with various UV doses, and their IFN-inducing capacity was determined at a multiplicity of virus comparable to that of the active virus, which induced peak or plateau yields of IFN (cf. Fig. 2). The IFN yield recorded at time zero represents the maximum amount of IFN induced by the unirradiated virus. Figure 6 reveals a qualitatively uniform response of AIV upon exposure to UV radiation irrespective of the intrinsic IFN-inducing capacity of the active virus. The viruses displayed three phases as a function of increasing UV dose with respect to IFN-inducing capacity: (i) an initial rapid increase, (ii) a peak yield, and (iii) a marked decline. Data from the parent and corresponding strains with the truncated (Fig. 6C) and deleted (Fig. 6D) NS1 genes are plotted on the same scale for ease of comparison. The increase and then decline in the IFN-inducing capacity observed for all strains of AIV following UV radiation confirm and extend the original observations (8, 9, 38).
The histogram in Fig. 7 shows the maximum yields of IFN produced by the eight strains of AIV as active virus and following UV irradiation. Also included are the number of lethal UV hits to the AIV genome required to reach peak IFN-inducing capacity. They are characteristic of each strain.
The histogram in Fig. 8 provides a direct comparison of the maximal IFN-inducing capacity of the two strains of AIV that express genetically compromised NS1 and their NS gene-competent counterparts, both as active and UV-irradiated virus. Note that parental strains require more UV hits to achieve maximum IFN-inducing capacity than do the NS1-compromised strains.
IFN induction dose-response curves induced by UV-irradiated influenza virus. Several stocks of influenza virus were UV irradiated to achieve maximal IFN-inducing capacity and then used to generate IFN induction dose-response curves. The curves were similar to those shown in Fig. 2 but displayed maximal yields of IFN in keeping with the enhanced IFN-inducing capacity acquired following irradiation, and hence the data are not shown. Suffice it to note that the concentration of IFP calculated from these curves was comparable to those found in active virus. This indicates that the increased IFN-inducing capacity of UV-irradiated virus reflects an increase in the quantum yield of IFN that an IFP can induce rather than the conversion of more particles to IFN-inducing status.
The IFN-inducing capacity of heat-inactivated influenza virus. Figure 9 illustrates the survival curves for both IFN-inducing capacity and PFP activity as a function of time at 50°C for A/TK/OR/71—a relatively weak inducer of IFN that phenotypically functions as an ISP (Fig. 5). Exposure to 50.0°C generated an exponential loss of infectivity (PFP) of –10-fold in 12 min. In contrast, the IFN-inducing capacity of the heated virus increased almost twofold and maintained that capacity following at least 24 min of exposure to heat. These results essentially confirm those that led to the discovery of IFN (26). Heat inactivation in the borate buffer used by Isaacs and Lindenmann (26) produced virus with IFN-inducing capacity comparable to that obtained with the NCI solution we used equilibrated in air. The order of magnitude differences in the yields of IFN reported originally and those recorded here are attributed in part to the greatly enhanced production of IFN from developmentally aged CEC, with 40.5°C as an optimal temperature for IFN induction/production, the enhanced sensitivity of current IFN assays, and the IFN-inducing capacity intrinsic to each strain (cf. Fig. 3).
DISCUSSION
The induction and/or production of IFN by eight different strains of influenza virus and its suppression by two of those strains were examined under conditions where developmentally aged primary CEC hyperproduce IFN when induced (70), the threshold of induction is a single molecule of dsRNA per cell (45, 50), single-stranded RNA (ssRNA) does not induce IFN (45, 50), and virus populations can be quantified for their content of both IFP (46) and ISP (52).
The production of IFN by active AIV generally displayed a 2- to 4-h lag followed by a steady increase in accumulation of secreted IFN. Maximal yields were observed after about 20 ± 5 h postinfection of cells incubated at 40.5°C, the optimal temperature for production of ChIFN- (73). The eight AIV strains revealed a 100-fold range for the maximum accumulated yield of IFN (85 to 8,750 U/107 cells). This compares with a 10,000-fold variation observed in the IFN-inducing capacity of scores of different isolates of vesicular stomatitis virus (VSV), reflecting the quasispecies nature of this RNA virus and the plasticity of the induction arm of the IFN system (48, 55, 58).
Analysis of IFP and ISP activity requires that the input multiplicity be preserved, especially at low multiplicities where the early release of newly produced virus might spuriously increase the multiplicity of infection. In this context, egg-derived AIV contains activated fusion peptide, but CEC-derived virions do not (35). Following IFN induction, newly generated progeny virus will be fusion incompetent and noninfectious, thus preserving the input multiplicity. The TK/ONT and WSN strains are exceptions in that they replicate in CEC without trypsin to generate the active fusion protein.
Data from IFN induction dose-response curves reveal that influenza virus populations contain a ratio of IFP to PFP of 20:1, indicating that noninfectious IFP appear to make up over 90% of influenza virus populations. Any undercounting of infectious particles would bring the ratio closer to the observed physical/infectious particle ratio of about 11 (12). Other viruses have been reported to produce an excess of noninfectious IFP to PFP. These include Newcastle disease virus (7:1) (56), VSV (10:1) (51), and ARV (60:1) (85).
The detection of noninfectious IFP is of interest. Although influenza virus gene segments have signals that select them for packaging (17), this mechanism does not guarantee that all eight unique segments will be packaged, nor does it exclude the packaging of multiple copies of some genes at the cost of precluding one or more of the unique eight required for infectivity (14, 15). Additional evidence, including the observed ratio of physical to infectious particles (12), the prediction from bin theory that the packaging of 12 gene segments per virion would ensure that about 10% of the virus particles contained the unique eight genes for infectivity (15, 71), the generation of defective-interfering particles (63), the transient resistance to IFN action that is lost upon repackaging of the genes (71), and the genetic multiplicity reactivation common to orthomyxoviruses (2), supports the view that imperfect packaging may produce an excess of noninfectious virions. Analysis of IFN induction dose-response curves can detect and quantify these particles as IFP (46). The high ratio of noninfectious IFP to PFP intrinsic to populations of active AIV, coupled with an even greater capacity to produce IFN following UV irradiation, and to a lesser degree after heat inactivation, demonstrates that virus replication is not required for IFN induction by AIV. Furthermore, since the polymerase of influenza virus is labile at 50°C, even in the presence of viral RNA promoter sequences (5; G. G. Brownlee, University of Oxford, Oxford, United Kingdom, personal communication), the IFN inducer moiety of influenza virus appears extant in virions prior to heat inactivation. We infer that any ISP activity intrinsic to the virion was inactivated by heat, allowing the expression of the IFP phenotype. However, following UV irradiation, it is possible that limited primary viral transcription from the 3' end of any of the gene segments packaged into noninfectious IFP virions suffices to generate dsRNA as the IFN inducer moiety within the cell. This is the case for VSV in CEC (51). If a comparable situation prevails in AIV, a higher UV dose would be required to prevent dsRNA production because there is an average of 12 gene segments per virion (15, 71), each independently transcribed (1, 66, 74). In that context, a loss of 90% of the IFN-inducing capacity of VSV (51) and AIV (Fig. 6) was observed at about 20 and 80 lethal UV hits, respectively. Interestingly, strains of AIV with genetically compromised NS genes require many fewer UV hits to render them incapable of inducing IFN. We infer from these data that virions of influenza virus may contain a preformed inducer of IFN—most likely viral dsRNA (45), acquired during virion assembly or produced during primary transcription, if not replication (3, 42, 43). Recent reports of the induction of IFN by ssRNA acting through TLR7 receptors in specialized dendritic cells (11, 24, 41) do not appear applicable to CEC cultures, since the delivery of ssRNA in the form of [-]RNA defective-interfering particles (DIP) over a range of 1 to 80 molecules per cell did not induce IFN, yet 1 [±]RNA molecule did (45, 50). Nor do viral proteins appear to be involved since both types of DIP contained identical polypeptides encoded by the same helper virus, yet differed qualitatively in their IFN inducing capacity (45, 50). Whether the same situation pertains to AIV remains to be determined.
Shortly after the discovery of IFN, Lindenmann reported that live influenza virus prevented the induction of IFN by inactivated virus, an effect he termed "inverse interference" (37). In this context, two of the AIV strains, A/TK/OR/71 and A/PR/8/34, were shown to be efficient suppressors of IFN production, i.e., demonstrated inverse interference. The exponential loss of IFN-inducing capacity as a function of ISP multiplicity indicates that a single ISP per cell suffices to block IFN induction and/or production completely (52). Noninfectious ISP appear to make up over 90% of the virus population. Suppression of IFN induction was demonstrable against both a homologous virus (UV-TK/WI/66) and the dsRNA-containing heterologous virus (UV-ARV), indicating that suppression is directed at the host cell. ISP need not be infectious, and hence expression of the ISP phenotype does not require the eight unique gene segments needed for infectivity. The production of large numbers of ISP during influenza virus infection might block activation of the IFN system and not bode well for the outcome of the disease.
As a function of UV dose, all eight influenza virus strains displayed three distinct phases with respect to their IFN-inducing capacity: (i) an initial rapid increase, followed by (ii) a peak or plateau, followed by (iii) a decline to virtually no production. Although these three phases differed quantitatively for the eight strains, there was a striking common feature: even the lowest doses of UV radiation enhanced IFN-inducing capacity. Importantly, there is essentially no threshold in the amount of UV radiation required to enhance IFN inducibility, meaning that at low doses virtually every UV hit to the genome of the noninfectious IFP improves its IFN-inducing capacity. Eventually, a UV dose is reached that maximizes that capacity—up to 100-fold—and thereafter inactivates it. This statement applies even to the two strains already genetically compromised in NS1 expression. Notably, these strains require fewer UV hits (5) to their genome to achieve maximal IFN-inducing capacity, whereas the parental strains require 30 UV hits.
We infer that UV inactivation of two targets in the AIV genome is required to inhibit IFN induction-suppressing activity and achieve maximal IFN-inducing capacity. A small UV target is attributed to the multifunctional NS gene and its NS1-encoded protein, which may block IFN induction by sequestering dsRNA (4, 19, 40) or the protein kinase PKR (81). In addition, it may prevent the 3'-end processing of cellular pre-mRNA because the effector region sequesters the 30-kDa subunit of cleavage and polyadenylation specificity factor(CPSF) and poly(A)-binding protein II (PABII) (16, 33, 61). The marked increase in the IFN-inducing capacity observed at very low doses of UV radiation seen in all strains, with or without an intact NS gene, implicates a role for a gene(s) with a large target. UV inactivation of the large target in AIV converts a weak or good inducer of IFN into a much better inducer—in effect converting an ISP into an IFP. Both activities reside in noninfectious particles.
To identify the gene(s) that constitute the large UV target, we compared the observed rate at which IFN-inducing capacity is enhanced as a function of UV dose, with the expected fraction of inactivation for different genes or combinations thereof based on a Poisson distribution of UV hits to the AIV population. We reasoned that the rate of increase in the IFN-inducing capacity of the AIV population would be most directly related to the rate of inactivation of the gene(s) that comprises the large UV target. The larger the gene, the fewer UV hits it would take to inactivate it. We sought to identify a gene, or combination of genes, whose expected rate of inactivation would best fit the rate at which UV radiation enhanced IFN-inducing capacity. Only one strain has been analyzed thus far: TK/OR/71(delNS1[1-124]). It was chosen because the small gene target already was compromised genetically, meaning that following UV irradiation, any further increase in IFN-inducing capacity could be relegated to a large gene target. Figure 10 (curve A) shows that the expected rate of inactivation of the NS gene in a population of AIV provides a poor fit to the actual rate at which UV radiation enhances IFN-inducing capacity. A somewhat better fit is observed if a gene for one of the three polymerase subunits is considered to be the UV target (PB1, PB2, or PA, which are close to the same size) (curve B). The fit is much improved if the target is assumed to consist of two or three of the polymerase genes and that a UV hit to any one of the two (curve C) or three (curve D) genes suffices to maximally enhance IFN-inducing capacity. The expected curves for UV inactivation of these genes assume that the AIV population is homogeneous with respect to gene content. However, a noninfectious AIV population may contain virions with only two active polymerase genes, accounting for some of the variable IFN-inducing capacity intrinsic to a strain of AIV (14, 35). We cannot distinguish between two or three polymerase subunits as the likely UV target, but for mechanistic reasons, we propose as a working hypothesis that the effective target is the sum of the three polymerase genes that encode the endonuclease complex responsible for snatching caps from newly transcribed cellular pre-mRNAs. Since all three polymerase subunits are required for cap snatching (36, 65), one UV hit to any of the genes encoding a polymerase subunit would block this process and allow production of translatable IFN mRNA, observed as enhanced IFN production. Strong support for this hypothesis comes from the report that the reduced number of cellular RNA polymerase II-derived transcripts that continue to be synthesized after AIV infection do not appear in the cytoplasm as mature mRNA, most likely because the AIV cap-snatching endonuclease renders the decapped mRNAs susceptible to nucleases (29). This process would compromise IFN production.
In addition, any delay in the posttranscriptional processing of cellular pre-mRNA at the 3' end through sequestering of cellular factors CPSF and PABII by the effector region of NS1 (33) would exacerbate the ISP activity of AIV by restricting pre-IFN mRNA to the nucleus, thereby increasing exposure to the cap-snatching endonuclease and further degradation by nucleases (16, 61, 68).
By this model, UV irradiation inhibits the dominant ISP phenotype by inactivating two targets: (i) a large target which consists of all three of the polymerase genes; and (ii) a small target, the NS gene. The important role of NS1A as a key defense mechanism of AIV may be gleaned from the preferential transcription of the NS gene in chicken cells and the appearance of NS1 in the cell within 1 h after infection, ensuring a rapid first response against the IFN system (74). The cumulative effect of inactivating both anti-IFN induction and production mechanisms would be to maximize expression of the IFP phenotype.
At very high UV doses, there is a marked decline in AIV IFN-inducing capacity. Since the hemagglutinating activity of AIV is not diminished by the highest UV doses used, virus attachment and entry into the cell are not likely to be affected, although further processing might be. Most strains require UV doses of about 8,000 ergs/mm2 (100 lethal hits) to achieve loss of virtually all IFN-inducing capacity. Kim and colleagues reported that a dose of UV which delivered about 90 lethal hits (our calculation) to an AIV strain prevented the activation of IRF-3 (30). This high dose of UV may prevent formation of an IFN inducer (dsRNA?) as noted above, and in turn the activation of NF-B (34) and IRF-3 (80).
If dsRNA preexists in AIV, as in the reovirus genome (85), or as a molecule of covalently linked self-complementary [±]RNA in some kinds of VSV defective-interfering particles (45, 47, 50), IFN-induction will not be compromised by UV irradiation or heat (45, 50). However, if dsRNA must be generated anew within the cell by the inducing virus, then heat or high doses of UV radiation will inactivate their IFN-inducing capacity (51). When influenza virus infectivity is inactivated by heat, some strains gain the capacity to induce IFN, providing the experimental basis for the discovery of IFN (26). As first described by Burke and Buchan (7) and reported in detail here, UV irradiation invariably initially enhances the IFN-inducing capacity of AIV. The enhanced IFN-inducing capacity observed in all of the influenza virus strains subjected to UV radiation is consistent with the presence of preformed dsRNA and the inactivation of virus genes represented by the small and the large UV targets. Alternatively, primary transcription by AIV may generate dsRNA which leads to NF-B (34) and IRF-3 activation (30). A model is favored in which low doses of UV irradiation inactivate the ISP activity of AIV, the large UV target, while allowing primary transcripts of sufficient length to produce dsRNA and induce IFN. At high UV doses, the size of the primary transcripts would become rate limiting for dsRNA formation (51), activation of transcription factors, and hence IFN induction (20).
Recent reports show that ssRNA viruses like AIV and VSV, or heat-inactivated AIV, can induce IFN in TLR7-containing cells of dendritic origin when delivered through endocytosis (11, 24, 41). While comparable experiments have not been reported for CEC, we note that in the absence of helper virus, otherwise identical [±] and [–]RNA DIP of VSV in CEC do and do not, respectively, induce IFN. These data show that in CEC the threshold for IFN induction is one molecule of dsRNA and that ssRNA, up to 80 molecules per cell delivered by endocytosis, does not induce IFN (45, 50), strengthening the case that the IFN-inducer moiety in, or generated by, AIV in CEC is dsRNA. Whether helicase (RIG I) (87) is involved in processing such dsRNA is yet to be determined in chicken cells.
AIV are intrinsically sensitive to IFN action (3, 21, 22, 23, 26, 67, 69, 76), most likely through the action of the MX system and inhibition of primary transcription (3, 31, 32). To circumvent this sensitivity, influenza viruses have evolved mechanisms to defeat host cell defenses by blocking sequential events in IFN production, namely, processing of both the 5' and 3' ends of IFN pre-mRNA (16, 33, 84). However, the broad spectrum of IFN-inducing capacities displayed by different AIV strains and differences in the extent to which UV irradiation and heat affect that capacity demonstrate that prevention of IFN induction and/or production is not always successful. Indeed, the IFP and ISP phenotypes may have a profound effect on the outcome of AIV infection, both in mice (13, 21) and in chickens (A. N. Cauthen and D. L. Suarez, personal communication). The potential for IFN to enhance the immune response (28) behooves us to define the IFP/ISP phenotypes of influenza virus strains and to determine whether the excess of noninfectious IFP or ISP produced during virus replication affects the immunobiology of AIV infection (78), pathogenesis of the disease, and the efficacy of attenuated live AIV vaccines.
ACKNOWLEDGMENTS
This research was supported by USDA grant 58-1940-0-007 through the Center for Excellence in Vaccine Research at the University of Connecticut. The study benefited from the use of the Animal Cell Culture Facility of the Biotechnology-Bioservices Center of the University of Connecticut.
We thank Adolfo García-Sastre for pointing out the advantages of using the developmental immaturity of the IFN system in 6- to 7-day-old embryonated eggs (70) to prepare stocks of AIV from strains prone to induce IFN because of genetically compromised expression of NS1A (86). We thank Robert M. Krug for insightful discussions.
This is contribution XXIV in a series entitled "Interferon Induction by Viruses."
Present address: Scripps Research Institute, 10550 N. Torrey Pines Rd., La Jolla, CA 92037
REFERENCES
Abraham, G. 1979. The effect of ultraviolet radiation on the primary transcription of influenza virus messenger RNAs. Virology 97:177-182.
Barry, R. D. 1961. The multiplication of influenza virus. II. Multiplicity reactivation of ultraviolet-irradiated virus. Virology 14:398-405.
Bean, W. J., and R. W. Simpson. 1973. Primary transcription of the influenza virus genome in permissive cells. Virology 56:646-651.
Bergmann, M., A. García-Sastre, E. Carnero, H. Pehamberger, K. Wolff, P. Palese, and T. Muster. 2000. Influenza virus NS1 protein counteracts the PKR-mediated inhibition of replication. J. Virol. 74:6203-6206.
Brownlee, G. G., and J. L. Sharps. 2002. The RNA polymerase of influenza A virus is stabilized by interaction with its viral RNA promoter. J. Virol. 76:7103-7113.
Budowsky, E. I., S. E. Bresler, E. A. Friedman, and N. V. Zheleznova. 1981. Principles of selective inactivation of viral genome. I. UV-induced inactivation of influenza virus. Arch. Virol. 68:239-247.
Burke, D. C., and A. Buchan. 1965. Interferon production in chick embryo cells. I. Production by ultraviolet-inactivated virus. Virology 26:28-35.
Burke, D. C., and A. Isaacs. 1958. Further studies on interferon. Br. J. Exp. Pathol. 39:78-84.
Burke, D. C., and A. Isaacs. 1958. Some factors affecting the production of interferon. Br. J. Exp. Pathol. 39:452-458.
Cox, N. J., and K. Subbarao. 1999. Influenza. Lancet 354:1277-1282.
Diebold, S. S., T. Kaisho, H. Hemmi, S. Akira, and C. R. E. Sousa. 2004. Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA. Science 303:1529-1531.
Donald, H. B., and A. Isaacs. 1954. Counts of influenza virus particles. J. Gen. Microbiol. 10:457-464.
Donelan, N. R., C. F. Basler, and A. García-Sastre. 2003. A recombinant influenza A virus expressing an RNA-binding-defective NS1 protein induces high levels of beta interferon and is attenuated in mice. J. Virol. 77:13257-13266.
Duhaut, S. D., and J. W. Mccauley. 1996. Defective RNAs inhibit the assembly of influenza virus genome segments in a segment-specific manner. Virology 216:326-337.
Enami, H., G. Sharma, C. Benham, and P. Palese. 1991. An influenza virus containing nine different gene segments. Virology 185:291-298.
Enserink, M., and J. Kaiser. 2004. Avian flu finds new mammal hosts. Science 305:1385.
Fortez, P., A. Beloso, and J. Ortín. 1994. Influenza virus NS1 protein inhibits pre-mRNA splicing and blocks RNA nucleocytoplasmic transport. EMBO J. 13:704-712.
Fujii, Y., H. Goto, T. Watanabe, T. Yoshida, and Y. Kawaoka. 2003. Selective incorporation of influenza virus RNA segments into virions. Proc. Natl. Acad. Sci. USA 100:2002-2007.
Gaccione, C., and P. I. Marcus. 1989. Interferon induction by viruses. XVIII. Vesicular stomatitis virus-New Jersey: a single infectious particle can both induce, and suppress, interferon production. J. Interferon Res. 9:603-614.
García-Sastre, A., A. Egorov, D. Matassov, S. Brandt, D. E. Levy, J. E. Durbin, P. Palese, and T. Muster. 1998. Influenza A virus lacking the NS1 gene replicates in interferon-deficient systems. Virology 252:324-330.
Geiss, G. K., M. Salvatore, T. M. Tumpey, V. S. Carter, X. Wang, C. F. Basler, J. K. Taubenberger, R. E. Bumgarner, P. Palese, M. G. Katze, and A. García-Sastre. 2002. Cellular transcriptional profiling in influenza A virus infected lung epithelial cells: role of the nonstructural NS1 protein in the evasion of the host innate defense and its potential contribution to pandemic influenza. Proc. Natl. Acad. Sci. USA 99:10736-10741.
Gresser, I., M. G. Tovey, C. Maury, and M. T. Bandu. 1976. Role of interferon in the pathogenesis of virus diseases in mice as demonstrated by the use of anti-interferon serum. II. Studies with herpes simplex, Maloney sarcoma, vesicular stomatitis, Newcastle disease and influenza viruses. J. Exp. Med. 144:1316-1327.
Gutman, N. R., and L. S. Manakhova. 1981. The interferon-inducing capacity of influenza virus strains and their sensitivity to the effect of exogenous interferon. Vopr. Virusol. 2:168-171.
Haller, O., M. Frese, and G. Kochs. 1998. Mx proteins: mediators of innate resistance to RNA viruses. Rev. Sci. Tech. 17:220-230.
Heil, F., H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, G. Lipford, H. Wagner, and S. Bauer. 2004. Species-specific recognition of single-stranded RNA via toll-like receptor 7 and 8. Science 303:1526-1529.
Hinshaw, V. S., C. W. Olsen, N. Dybdahl-Sissoko, and D. Evans. 1994. Apoptosis: a mechanism of cell killing by influenza A and B viruses. J. Virol. 68:3667-3673.
Isaacs, A., and J. Lindenmann. 1957. Virus interference. I. The interferon. Proc. R. Soc. Ser. B 147:258-267.
Isaacs, A., J. Lindenmann, and R. C. Valentine. 1957. Virus interference. II. Some properties of interferon. Proc. R. Soc. Ser. B 147:268-273.
Julkunen, I., T. Sareneva, J. Pirhonen, T. Ronni, K. Melén, and S. Matikainen. 2001. Molecular pathogenesis of influenza A virus infection and virus-induced regulation of cytokine gene expression. Cytokine Growth Factor Rev. 12:171-180.
Katze, M. G., and R. M. Krug. 1984. Metabolism and expression of RNA polymerase II transcripts in influenza virus-infected cells. Mol. Cell. Biol. 4:2198-2206.
Kim, M.-J., A. G. Latham, and R. M. Krug. 2002. Human influenza viruses activate an interferon-independent transcription of cellular antiviral genes: the outcome with influenza A virus is unique. Proc. Natl. Acad. Sci. USA 99:10096-10101.
Ko, J.-H., H.-K. Jin, A. Asano, A. Takada, A. Ninomiya, H. Kida, H. Hokiyama, M. Ohara, M. Tsuzuki, M. Nishibori, M. Mitzutani, and T. Watanabe. 2002. Polymorphisms and the differential antiviral activity of the chicken Mx gene. Genome Res. 12:595-601.
Krug, R. M., M. Shaw, B. Broni, G. Shapiro, and O. Haller. 1985. Inhibition of influenza viral mRNA synthesis in cells expressing the interferon-induced Mx gene product. J. Virol. 56:201-206.
Krug, R. M., W. Yuan, D. L. Noah, and A. G. Latham. 2003. Intracellular warfare between human influenza viruses and human cells: the roles of the viral NS1 protein. Virology 309:181-189.
Kumar, A., J. Haque, J. Lacoste, J. Hiscott, and B. R. G. Williams. 1994. Double-stranded RNA-dependent protein kinase activates transcription factor NF-B by phosphorylating IB. Proc. Natl. Acad. Sci. USA 91:6288-6292.
Lamb, R. A., and R. M. Krug. 2001. Orthomyxoviridae: the viruses and their replication, p. 725-769. In D. M. Knipe and P. M. Howley (ed.), Fundamental virology, 4th ed. Lippincott Williams & Wilkins, Philadelphia, Pa.
Li, M. L., P. Rao, and R. M. Krug. 2001. The active sites of the influenza cap-dependent endonuclease are on different polymerase subunits. EMBO J. 20:2078-2086.
Lindenmann, J. 1960. Interferon und inverse interferenz. Z. Hyg. Infectionskr. Med. Mikrobiol. Immunol. Virol. 146:287-309.
Lindenmann, J., D. C. Burke, and A. Isaacs. 1957. Studies on the production, mode of action and properties of interferon. Br. J. Exp. Pathol. 38:551-562.
Lipatov, A. S., E. A. Govorkova, R. J. Webby, H. Ozaki, M. Peiris, Y. Guan, L. Poon, and R. G. Webster. 2004. Influenza: emergence and control. J. Virol. 78:8951-8959.
Lu, Y., M. Wambach, M. G. Katze, and R. M. Krug. 1995. Binding of the influenza virus NS1 protein to double-stranded RNA inhibits the activation of the protein kinase that phosphorylates the eIF-2 translation initiation factor. Virology 214:222-228.
Lund, J. M., L. Alexopoulou, A. Sato, M. Karaow, N. C. Adams, N. W. Gale, A. Iwasaki, and R. A. Flavell. 2004. Recognition of single-stranded RNA viruses by toll-like receptor 7. Proc. Natl. Acad. Sci. USA 101:5598-5603.
Majde, J. A. 2000. Viral double-stranded RNA, cytokines, and the flu. J. Interferon Cytokine Res. 20:259-272.
Majde, J. A., N. Guha-Thakurta, Z. Chen, S. Bredow, and J. M. Krueger. 1998. Spontaneous release of stable viral double-stranded RNA into the extracellular medium by influenza virus-infected MDCK endothelial cells: implications for the viral acute-phase response. Arch. Virol. 143:2371-2380.
Marcus, P. I. 1982. Interferon induction by viruses. IX. Antagonistic activities of virus particles modulate interferon production. J. Interferon Res. 2:511-518.
Marcus, P. I. 1983. Interferon induction by viruses: one molecule of dsRNA as the threshold for interferon induction, p. 115-180. In I. Gresser (ed.), Interferon 5, Academic Press, New York, N.Y.
Marcus, P. I. 1986. Interferon induction: dose-response curves. Methods Enzymol. 119:106-114.
Marcus, P. I., and C. Gaccione. 1989. Interferon induction by viruses. XIX. Vesicular stomatitis virus-New Jersey: high multiplicity passages generate interferon-inducing, defective-interfering particles. Virology 171:630-633.
Marcus, P. I., L. L. Rodriguez, and M. J. Sekellick. 1998. Interferon induction as a quasispecies marker of vesicular stomatitis virus populations. J. Virol. 72:542-549.
Marcus, P. I., and M. J. Sekellick. 1975. Cell killing by viruses. II. Cell killing by vesicular stomatitis virus: a requirement for virion-derived transcription. Virology 63:176-190.
Marcus, P. I., and M. J. Sekellick. 1977. Defective-interfering particles with covalently-linked [±] RNA induce interferon. Nature 266:815-819.
Marcus, P. I., and M. J. Sekellick. 1980. Interferon induction by viruses. III. Vesicular stomatitis virus: interferon-inducing particle activity requires partial transcription of gene N. J. Gen. Virol. 47:89-96.
Marcus, P. I., and M. J. Sekellick. 1985. Interferon induction by viruses. XIII. Detection and assay of interferon induction-suppressing particles. Virology 142:411-415.
Marcus, P. I., and M. J. Sekellick. 1987. Interferon induction by viruses. XV. Biological characteristics of interferon induction-suppressing particles of vesicular stomatitis virus. J. Interferon Res. 7:269-284.
Marcus, P. I., and M. J. Sekellick. 2001. Combined sequential treatment with interferon and dsRNA abrogates virus resistance to interferon action. J. Interferon Cytokine Res. 21:423-427.
Marcus, P. I., M. J. Sekellick, and S. T. Nichol. 1992. Interferon induction by viruses. XXI. Vesicular stomatitis virus: interferon inducibility as phylogenetic marker. J. Interferon Res. 12:297-305.
Marcus, P. I., C. Svitlik, and M. J. Sekellick. 1983. Interferon induction by viruses. X. A model for interferon induction by Newcastle disease virus. J. Gen. Virol. 64:2419-2431.
Marcus, P. I., L. van der Heide, and M. J. Sekellick. 1999. Chicken interferon action on avian viruses. I. Oral administration of chicken interferon- ameliorates Newcastle disease. J. Interferon Cytokine Res. 19:881-885.
Mattana, P., and G. C. Viscomi. 1998. Variations in the IFN-inducing capacity of Sendai virus subpopulations. J. Interferon Cytokine Res. 18:399-405.
Meegan, J. M., and P. I. Marcus. 1989. Double-stranded ribonuclease coinduced with interferon. Science 244:1089-1091.
Mo, C. W., Y. C. Cao, and B. L. Lim. 2001. The in vivo and in vitro effects of chicken interferon alpha on infectious bursal disease virus and Newcastle disease virus infection. Avian Dis. 45:389-399.
Noah, D. L., K. Y. Twu, and R. M. Krug. 2003. Cellular antiviral responses against influenza A virus are countered at the postranscriptional level by the viral NS1A protein via its binding to a cellular protein required for the 3' end processing of cellular pre-mRNAs. Virology 307:386-395.
Norton, G. P., T. Tanaka, K. Tobita, S. Nakada, D. A. Buonagurio, D. A. Greenspan, M. Krystal, and P. Palese. 1987. Infectious influenza A and B virus variants with long carboxyl terminal deletions in the NS1 polypeptides. Virology 156:204-213.
Odagiri, T., and M. Tashiro. 1997. Segment-specific noncoding sequences of the influenza virus genome RNA are involved in the specific competition between defective interfering RNA and its progenitor RNA segment at the virion assembly step. J. Virol. 71:2138-2145.
Pei, J., M. J. Sekellick, P. I. Marcus, I.-S. Choi, and E. W. Collisson. 2001. Chicken interferon type I inhibits infectious bronchitis virus (IBV) replication and associated respiratory illness. J. Interferon Cytokine Res. 21:1071-1077.
Plotch, S. J., M. Bouloy, I. Ulmanen, and R. M. Krug. 1981. A unique cap (m7GpppXm)-dependent influenza virion endonuclease cleaves capped RNAs to generate the primers that initiate viral RNA transcription. Cell 23:847-858.
Pons, M. W., and O. M. Rochovansky. 1979. Ultraviolet inactivation of influenza virus RNA in vitro and in vivo. Virology 97:183-189.
Portnoy, J., and T. C. Merigan. 1971. The effect of interferon and interferon inducers on avian influenza. J. Infect. Dis. 124:545-552.
Qiu, Y., and R. M. Krug. 1994. The influenza virus NS1 protein is a poly(A)-binding protein that inhibits nuclear export of mRNAs containing poly(A). J. Virol. 68:2425-2432.
Richman, D. D., B. R. Murphy, S. Baron, and C. Uhlendorf. 1976. Three strains of influenza A virus (H3N2): interferon sensitivity in vitro and interferon production in volunteers. J. Clin. Microbiol. 3:223-226.
Sekellick, M. J., W. J. Biggers, and P. I. Marcus. 1990. Development of the interferon system. I. In chicken cells development in ovo continues on time in vitro. In Vitro Cell. Dev. Biol. 26:997-1003.
Sekellick, M. J., S. A. Carra, A. Bowman, D. A. Hopkins, and P. I. Marcus. 2000. Transient resistance of influenza virus to interferon action attributed to random multiple packaging and activity of NS genes. J. Interferon Cytokine Res. 20:963-970.
Sekellick, M. J., A. F. Ferandino, D. A. Hopkins, and P. I. Marcus. 1994. Chicken interferon gene: cloning, expression, and analysis. J. Interferon Res. 13:413-418.
Sekellick, M. J., and P. I. Marcus. 1986. Induction of high titer chicken interferon. Methods Enzymol. 119:115-125.
Smith, G. L., and A. J. Hay. 1982. Replication of the influenza virus genome. Virology 118:96-108.
Staeheli, P., F. Puehler, K. Schneider, T. W. G?bel, and B. Kaspers. 2001. Cytokines of birds: conserved functions—a largely different look. J. Interferon Cytokine Res. 21:993-1010.
Stewart, W., II, E. Declercq, A. Billiau, J. Desmyter, and P. DeSomer. 1972. Increased susceptibility of cells treated with interferon to the toxicity of polyriboinosinic acid-polyribocytidylic acid. Proc. Natl. Acad. Sci. USA 69:1851-1854.
Suarez, D. L., and M. L. Perdue. 1998. Multiple alignment comparison of the non-structural genes of influenza A viruses. Virus Res. 54:59-69.
Suarez, D. L., and S. Schultz-Cherry. 2000. Immunobiology of avian influenza virus: a review. Dev. Comp. Immunol. 24:269-283.
Takizawa, T., S. Matsukkawa, Y. Higuchi, S. Nakamura, Y. Nakanishi, and R. Fukuda. 1993. Induction of programmed cell death (apoptosis) by influenza virus infection in tissue culture cells. J. Gen. Virol. 74:2347-2355.
Talon, J., C. M. Horvath, R. Polley, C. F. Basler, T. Muster, P. Palese, and A. García-Sastre. 2000. Activation of interferon regulatory factor 3 is inhibited by the influenza A virus NS1 protein. J. Virol. 74:7989-7996.
Tan, S. L., and M. G. Katze. 1998. Biochemical and genetic evidence for complex formation between the influenza A virus NS1 protein and the interferon-induced PKR protein kinase. J. Interferon Cytokine Res. 18:757-766.
Tanaka, N., M. Sato, M. S. Lamphier, H. Nozawa, E. Oda, S. Noguchi, R. D. Schreiber, Y. Tsujimoto, and T. Taniguchi. 1998. Type I interferons are essential mediators of apoptotic death in virally infected cells. Genes Cells 3:29-37.
Van Campen, H., B. C. Easterday, and V. S. Hinshaw. 1989. Virulent influenza A viruses: their effect on avian lymphocytes and macrophages. J. Gen. Virol. 70:2887-2895.
Wang, W., and R. M. Krug. 1996. The RNA-binding and effector domains of the viral NS1 protein are conserved to different extents among influenza A and B viruses. Virology 223:41-50.
Winship, T. R., and P. I. Marcus. 1980. Interferon induction by viruses. VI. Reovirus virion genome dsRNA as the interferon inducer in aged chicken embryo cells. J. Interferon Res. 1:155-167.
Yewdell, J., and A. García-Sastre. 2002. Influenza still surprises. Curr. Opin. Microbiol. 5:414-418.
Yoneyama, M., M. Kikuchi, T. Natsukawa, N. Shinobu, T. Imaizumi, M. Miyagishi, K. Taira, S. Akira, and T. Fujita. 2004. The RNA helicase RIG-I had an essential function in double-stranded RNA-induced innate antiviral responses. Nat. Immunol. 5:730-737.(Philip I. Marcus, Jillian)