Astrovirus-Induced Synthesis of Nitric Oxide Contributes to Virus Control during Infection

Abstract

Astrovirus is one of the major causes of infant and childhood diarrhea worldwide. Our understanding of astrovirus pathogenesis trails behind our knowledge of its molecular and epidemiologic properties. Using a recently developed small-animal model, we investigated the mechanisms by which astrovirus induces diarrhea and the role of both the adaptive and innate immune responses to turkey astrovirus type-2 (TAstV-2) infection. Astrovirus-infected animals were analyzed for changes in total lymphocyte populations, alterations in CD4⁺/CD8⁺ ratios, production of virus-specific antibodies (Abs), and macrophage activation. There were no changes in the numbers of circulating or splenic lymphocytes or in CD4⁺/CD8⁺ ratios compared to controls. Additionally, there was only a modest production of virus-specific Abs. However, adherent spleen cells from infected animals produced more nitric oxide (NO) in response to ex vivo stimulation with lipopolysaccharide. In vitro analysis demonstrated that TAstV-2 induced macrophage production of inducible nitric oxide synthase. Studies using NO donors and inhibitors in vivo demonstrated, for the first time, that NO inhibited astrovirus replication. These studies suggest that NO is important in limiting astrovirus replication and are the first, to our knowledge, to describe the potential role of innate immunity in astrovirus infection.

Astroviruses were first identified in infants with diarrhea in 1975 by Madeley and Cosgrove (C. R. Madeley and B. P. Cosgrove, Letter, Lancet ii:451-452, 1975) and are now recognized as one of the leading causes of childhood diarrhea worldwide. By the age of five, 90% of children have antibodies against astroviruses (17, 25). In addition to their endemic nature, astroviruses also cause outbreaks of enteritis in schools, geriatric care facilities, children's hospitals, and in immunocompromised individuals (25). In fact, the elderly and the immunocompromised, such as AIDS patients, represent an expanding demographic of astrovirus disease (30).

Astroviruses are transmitted mainly through a fecal-oral route (24). The virus typically has an incubation period of 1 to 4 days and causes an acute gastroenteritis which lasts approximately 4 days (9). Diarrhea is the most common symptom; however, vomiting, abdominal distention, and dehydration can occur (25). Much of what is known about astrovirus-mediated disease comes from epidemiological studies involving routine surveillance for enteric disease agents, following outbreaks, and serologic studies. Observational data from human samples and serological surveys suggest that antibodies are the key mediators of protection (18). However, there is nothing known about the role of the innate or cellular immune responses in astrovirus resistance. This is due to the lack of a small-animal model for astrovirus infection.

We developed a small-animal model using turkey astrovirus type-2 (North Carolina/034/1999) (TAstV-2) to study the mechanisms of viral pathogenesis and immune protection (3, 14, 16). Using young turkeys, we defined the replication, kinetics, and pathogenesis of astrovirus infection. We demonstrated that astrovirus replicated in the intestines with viral titers peaking between days 3 and 5 and dissipating by day 9 postinoculation. Viral antigen was detected in nonintestinal tissues, and infectious virus was isolated from these tissues and blood, primarily between 3 to 5 days postinfection (dpi), indicating viral spread was systemic, although viral replication was only detected in the intestine (3, 14).

In these studies, we examined the role of the adaptive and innate immune responses in the control and clearance of astrovirus infection using the turkey model. Our results demonstrated that T-cell populations and virus-specific antibodies (Abs) were not substantially altered in response to TAstV-2 infection. However, virus infection induced macrophage (Mφ) production of nitric oxide (NO), and NO suppressed viral replication during infection. This is the first experimental evidence of an interaction between astrovirus and Mφs and demonstrates a potentially significant role for innate immunity in primary astrovirus infection.

MATERIALS AND METHODS

Viral propagation.

TAstV-2 was isolated and propagated as described previously (16, 32). Briefly, fluid isolated from the intestines of TAstV-2-infected turkey poults was clarified by centrifugation (500 × g for 10 min), 0.2-μm filtered, diluted 10⁻³ in phosphate-buffered saline (PBS), and inoculated into the yolk sac of 20-day-old specific-pathogen-free turkey embryos. Five days postinoculation, embryo intestines and intestinal fluid were isolated separately. Infected and control intestines were homogenized in 1 ml of Dulbecco's modified Eagle's medium (Cellgro), clarified by centrifugation, and filtered as above, and total protein was determined by the Bradford colorimetric assay (BCA kit; Pierce). Embryo intestinal fluid (EIF) was collected from infected embryos (negative embryos do not contain fluid in their intestines), clarified by centrifugation, filtered as above, and tested for viral load by real-time reverse transcription-PCR (RT-PCR) and limiting dilutions in eggs. EIF typically contained 10¹² viral genomes (VG)/ml and ∼10⁹ infectious particles/ml as determined by limiting dilution followed by indirect immunofluorescent staining of inoculated embryo intestines.

Virus purification.

TAstV-2 was purified by size-exclusion low-pressure chromatography on a 1.5- by 50-cm (Bio-Rad) Sephacryl CL-6B (Sigma) column preequilibrated with Tris-buffered saline (pH 7.4). The column was run at a flow rate of 0.5 ml/min, and 2-min fractions were collected, with a total run time of 180 min. Fractions were analyzed for total protein concentration by BCA analysis, and the location of TAstV-2 was confirmed by RT-PCR, sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and Western blot analysis. SDS-PAGE and Western blot analysis were performed on a 5-to-20% gradient gel, separated by electrophoresis at 100 V for 60 min, and then either stained for total protein using GelCode Blue stain reagent (Pierce) or transferred to nitrocellulose (Bio-Rad), and viral antigen was detected using a rabbit polyclonal antibody generated to a peptide sequence in the TAstV-2 capsid protein (K₆₇₆-R₆₉₁ [described in reference 14]). Column fractions were also assayed for infectious virus by inoculation into eggs and examined for the accumulation of intestinal fluid, as well as viral replication, by RT-PCR.

Cell culture.

The chicken Mφ cell line (HD11) was kindly provided by Kiek Klasing (University of California, Davis). Cells were cultured with RPMI 1640 (CellGro) medium supplemented with 5% heat-inactivated fetal bovine serum (CellGro) and l-glutamine (2 mM; CellGro) in a humidified incubator with 5.5% CO₂ at 41°C.

Chemicals.

The following chemicals were used: N-p-tosyl-l-phenylalanine chloromethyl ketone (TPCK)-treated trypsin (Pierce) at final concentrations of 1, 5, or 10 μg/ml; lipopolysaccharide (LPS) from Escherichia coli 0127:B8 (Sigma) at 10 ng/ml as a positive control for stimulation of inducible NO synthase (iNOS) activity in vitro; the endotoxin inhibitor polymyxin B sulfate (PMB; Fluka) at a final concentration of 1.5 μg/ml; actinomycin D (ActD; Sigma), an inhibitor of RNA transcription from a DNA template, at a final concentration of 1 μg/ml; N^G-monomethyl-l-arginine (L-NMMA; Calbiochem) at 4 mM, to block NOS activity; a fourfold molar excess of normal l-arginine; the NO donor compound (±)-S-nitroso-N-acetylpenicillamine (SNAP; Calbiochem) used at 500 μM; and the iNOS inhibitor N-[3-(aminomethyl)benzyl]acetamidine, dihydrochloride (1400W; Calbiochem), used at 10 mg/kg of body weight. The concentrations and diluents used had no effect on HD11 viability, as determined by trypan blue exclusion.

RT-PCR.

Total RNA was isolated from EIF (100 μl), tissues (∼100 mg), cells (10⁵ to 10⁶), or purified virus (100 μl) using TRIzol reagent (Invitrogen) following the manufacturer's instructions. Routine detection of TAstV-2 was done by RT-PCR as described previously (15).

TAstV-2 real-time RT-PCR.

TAstV-2 replication and viral load were determined by real-time RT-PCR. Total RNA isolated from cells or tissue was quantitated by spectrophotometry, and equal amounts of total RNA from each sample were added to a 25-μl reaction volume using the one-step RT-PCR kit QuantiTect Probe RT-PCR (Qiagen). The presence of TAstV-2 in experimental samples was detected using primers and probe specific to the polymerase gene (TAV2TMpolFWD, 5′-GAC TGA AAT AAG GTC TGC ACA GGT-3′; TAV2TMpolREV, 5′-AAC CTG CGA ACC CTG CG-3′; TAV2TMpolPRB, 5′-6-carboxyfluorescein-ATG GAC CCC CTT TTT CGG CGG-black hole quencher-1/-3") and quantitated by comparing the samples to a TAstV-2pol RNA standard curve as previously described (14). Primers and probe were designed using the Primer Express version 1.5 (Applied Biosystems) program and constructed by Integrated DNA Technologies, Inc. Reactions were performed using the ABI Prism 7700 Sequence Detector and analyzed using Sequence Detector version 1.7 (Applied Biosystems). All samples and standards were amplified in duplicate. All experiments were performed at least three times.

iNOS RT-PCR.

Increased expression of iNOS RNA was detected by RT-PCR using primers specific to the chicken iNOS gene (accession number U46504). Primers (MKChiNOSFwd, 5′-CTG TGC TTC ATA GCT TCC AG-3′; MKChiNOSRev, 5′-AGG CAC AGA ACT CAG GAT AC-3′) were designed using PRIMER Designer version 2.01 (Scientific and Educational Software). Briefly, total RNA was isolated from HD11 cells treated with medium alone, LPS (10 ng/ml), or 2 × 10⁵ VG/well of purified TAstV-2 following 1, 2, 4, 8, 12, 24, or 48 h of incubation. Equal amounts of isolated RNA were treated with 1 U of DNase I (amplification grade; Invitrogen) for 15 min at room temperature. The enzyme was inactivated by the addition of 1 μl of EDTA (25 mM), and the reaction was heated to 65°C for 10 min. Treated RNA was then brought to a final volume of 20 μl with the addition of first-strand buffer, dithiothreitol, reverse primer, and SuperScript II reverse transcriptase (Invitrogen) following the manufacturer's instructions. RNA was incubated at 45°C for 60 min, at which time 2 μl was removed and used as template in a 50-μl PCR mixture. Following amplification, products were separated by electrophoresis through a 1.5% agarose gel and visualized by ethidium bromide staining. Results are representative of three experiments.

Animals.

Two-day-old unvaccinated British United Turkey of America poults (male and female) were obtained from a commercial hatchery. Control and infected poults were housed in separate BL2 containment facilities in individual Horsfall units with HEPA-filtered inlet and exhaust air valves. Birds were fed routine turkey starter from the University of Georgia and given free access to clean water. After a brief acclimation period, 5-day-old poults were randomly assigned to either a control group or a group infected with astrovirus (n = 60 per group). Poults were orally inoculated with ∼10⁶ genomic units of astrovirus in a 200-μl total volume or with PBS alone, as previously described (16). Birds were monitored daily for signs of clinical disease. On dpi 5, 9, 11, 16, and 21, five random poults per group were euthanized by cervical dislocation and the spleens, intestines, gall bladder, and blood were collected. Spleens from three to five poults were pooled and placed in cold RPMI and homogenized by physical disruption as previously described (20). Heparinized blood from three to five turkeys was pooled and diluted 1:1 in PBS. Leukocytes were then isolated from single-cell suspensions of blood and spleens using Histopaque 1077 (Sigma), following the manufacturer's instructions. Cells from buffy coat were isolated and subsequently tested for TAstV-2 by RT-PCR and egg culture as previously described. Additional cells were seeded in 96-well plates at 10⁶ cells/well and treated with complete RPMI 1640 with and without LPS (10 ng/ml). Following 48 h of stimulation, nitrite levels present in the supernatants were assayed using the Griess assay (34). An additional group of animals intranasally inoculated with 10⁶ 50% egg infectious dose units of the LaSota strain of Newcastle disease virus (NDV; kindly provided by Jack King, Southeast Poultry Research Laboratory) served as a positive control for turkey immune function. The animal experiments were repeated three times with different groups of poults with similar results. All animal experiments were approved by the USDA Animal Care and Use Committee and complied with all federal guidelines.

TAstV-2-specific enzyme-linked immunosorbent assay.

To detect TAstV-2-specific Abs, Immunlon 4 microtiter plates (Dynex) were coated with 2.5 μg of recombinant TAstV-2 capsid protein, 2.5 μg of NDV, or 1% bovine serum albumin (Invitrogen) and incubated with decreasing concentrations of serum or bile from infected and control animals as previously described (20). Briefly, serum immunoglobulin G (IgG) or bile IgA was detected by diluting samples in PBS containing 0.5% Tween 20 (PBST) and incubated for 1 h at room temperature. Plates were washed with PBST, incubated with either alkaline phosphatase-conjugated rabbit anti-chicken/turkey IgG (Zymed) or anti-chicken IgA (Bethyl Laboratories, Inc.), and incubated for 1 h at room temperature. Plates were washed with PBST and detected using 100 μl of FAST p-nitrophenyl phosphate tablets (Sigma) according to the manufacturer's instructions and incubated for 30 min at room temperature in the dark. The presence of TAstV-2 capsid-specific IgG or IgA was then measured at 450 nm on a microplate spectrophotometer. Recombinant TAstV-2 protein was generated using the single-tube protein system 3 (Novagen) and a plasmid containing open reading frame 2 of TAstV-2 (pcDNA3.1⁻/TAstVcap10), and it was purified by anion-exchange column chromatography (Bio-Rad). Samples were called positive if their optical density was at least twice that of the negative controls.

Flow cytometry.

Lymphocytes from peripheral blood and spleen were isolated as described above. Phenotyping of isolated cells was performed as previously described (33). Briefly, cells were washed in cold PBS and then 10⁶ cells were incubated with 0.5 μg of mouse anti-chicken CD4 directly conjugated to fluorescein (clone CT-4), 0.25 μg of mouse anti-chicken CD8α directly conjugated to phycoerythrin (clone 3-298; a generous gift from Southern Biotechnology Associates, Inc.), or corresponding amounts of mouse IgG1 isotype controls directly conjugated to fluorescein or phycoerythrin for 1 h on ice. The cells were then washed with 1 ml of cold PBS, fixed with cold 1% PBS-buffered formalin (Invitrogen), and analyzed on an Epics XL flow cytometer (Beckman Coulter). Lymphocytes were identified by their forward and side scatter properties. Gated lymphocytes were analyzed for CD4⁺ and CD8⁺ cells using both single-color and two-color analyses to ensure proper compensation. The percentage of each phenotype was determined based on cells positive for only one of the two markers. A total of 10,000 cells for each sample was analyzed. The CD4/CD8 ratios reported are the averages of three separate experiments.

Nitrite assay.

Up-regulation and expression of iNOS in HD11 cells was measured indirectly by determining the levels of nitrite in cell culture supernatants using the Griess assay (34). Briefly, 10⁵ cells/well were treated with RPMI alone, RPMI containing LPS, or TAstV-2 infectious material (EIF, homogenized embryo intestines, or column-purified virus) and incubated in a final volume of 100 μl at 41°C for 48 h in a 96-well tissue culture plate. Following incubation, 50 μl of cell-free supernatant was assayed for the presence of nitrite by mixing with equal volumes of 1% sulfanilamide (in 5% phosphoric acid; Sigma) and 0.1% N-1-napthylethylenediamine dihydrochloride (Sigma). Plates were incubated for 15 min in the dark, absorbance was measured at 550 nm using a spectrophotometer, and the nitrite concentration was determined by comparing to a nitrite standard curve. All treatments were done in triplicate, and each experiment was performed at least three times. Media, EIF, and column-purified TAstV-2 were tested for contaminating endotoxins using the limulus amebocyte lysate QCL-1000 kit (BioWhittaker). All reagents tested were found to have less than 1 endotoxin unit/ml. LPS treatment added to HD11 cells contained at least 6 endotoxin units/ml.

HD11 cell infection with TAstV-2.

To determine if HD11 cells support TAstV-2 replication, 5 × 10⁵ cells were seeded into each well of a 24-well plate and incubated overnight. Medium was removed, and cells were inoculated with 5 μl of EIF or 5 μl of PBS in a final volume of 250 μl of serum-free medium (SFM) and incubated for 1 h at 41°C. After 1 h, the inoculum was removed and replaced with either complete medium or SFM containing increasing concentrations of TPCK-treated trypsin (1 to 10 μg/ml). To control for the detection of input virus, replicate cells were fixed by drying at room temperature for 1 h and then inoculated with TAstV-2 as above and incubated with SFM containing 10 μg of trypsin/ml. Cells were monitored for cytopathic effect and viability by trypan blue exclusion. Cells were collected by vigorous pipetting at 24, 48, and 72 h postinoculation and pelleted by brief centrifugation, and supernatants were removed. Whole-cell pellets were lysed in 1 ml of TRIzol reagent. RNA concentrations were determined using spectrophotometry, and 500 ng of each sample RNA was used to detect TAstV-2 genomes using real-time RT-PCR.

Role of NO in ovo.

To examine the role of NO in viral replication in ovo, TAstV-2 was inoculated into specific-pathogen-free turkey embryos in the presence of PBS, NO donor (SNAP), or iNOS inhibitor (1400W), and viral titers were measured by real-time RT-PCR. Briefly, 0.2-μm filtered EIF was diluted in PBS to contain ≥10⁵ embryo infectious units/ml (10⁸ VG/ml), and 50 μl was incubated with either SNAP (500 μM) or 1400W (1.91 mg/ml) for 45 min at room temperature in a final volume of 100 μl. Three eggs were each inoculated with 100 μl of either SNAP plus TAstV-2, 1400W plus TAstV-2, or TAstV-2 plus PBS (positive control). Two eggs were inoculated with 100 μl of SNAP plus PBS, 1400W plus PBS, or PBS alone (negative controls). Eggs were monitored daily for viability and opened at 5 dpi, and intestines were collected. Sections of cecum and duodenal loop were collected from each inoculated embryo, pooled, and either preserved in 10% formalin solution (Fisher) for histological examination or placed into 500 μl of RNAlater (Ambion) for RNA isolation. Pooled tissues for RNA isolation were weighed to ensure that between 80 and 100 mg of tissue was used, per the manufacturer's instructions. Tissues were homogenized in 1 ml of TRIzol reagent, and RNA was resuspended in RNase-free water and quantitated by spectrophotometry. From each sample, 10 ng of total RNA was used to determine the amount of TAstV-2 present by real-time RT-PCR.

Immunostaining.

Tissues from control and infected embryo intestines were fixed in 10% phosphate-buffered formalin overnight, processed, embedded, and sectioned (0.3 μm). Sections were deparaffinized with Citrisolv (Fisher), and antigenic sites were exposed by microwaving for 5 min in a citrate buffer, as previously described (14). The ability of TAstV-2 to replicate in ovo was detected using a rabbit polyclonal Ab generated to a peptide sequence in the TAstV-2 capsid protein, followed by a universal biotin-conjugated Ab from the Vectastain Universal ABC-AP kit (Vector) and detected using streptavidin-labeled Alexa488 (Molecular Probes) as previously described (14). In addition to viral antigen, serial sections were stained for the presence of nitrated tyrosine residues, an in situ indicator of peroxynitrite formation, using a rabbit polyclonal antinitrotyrosine Ab (Molecular Probes) and the Vectastain Universal ABC-AP kit, and they were detected using Fast Red TR/naphthol AS-MX tablets (Sigma) following the manufacturer's instructions. Tissues were counterstained with Harris modified hematoxylin (Fisher).

Statistics.

All error bars represent the 95% confidence interval of the mean, and P values are indicated.

RESULTS

No evidence of adaptive immune response to TAstV-2 infection.

To determine the effects of astrovirus infection on the cellular immune response, we examined leukocytes isolated from infected poults. Leukocytes from peripheral blood and spleens were analyzed for TAstV-2 by RT-PCR and virus isolation using egg culture. Repeated attempts to demonstrate that TAstV-2 was associated with the leukocyte fraction were unsuccessful, suggesting TAstV-2 is not spread to extraintestinal tissues by white blood cells. To evaluate the adaptive immune response to astrovirus, we examined the numbers of leukocytes, induction of astrovirus-specific Abs, and alternations in T-cell populations during infection. There was no difference in peripheral blood leukocyte numbers between infected and control animals (data not shown). To determine the concentration and type of Abs produced in response to TAstV-2-infection, we assayed serum and bile for virus-specific IgG and IgA, respectively (Fig. 1). Although virus-specific Abs were detected, titers were very low. Specific IgG was undetectable at 11 dpi, with a titer of only 8 at 21 dpi (Fig. 1). The levels of IgA were also undetectable at 11 dpi; however, the titers increased fourfold over those of IgG by 21 dpi (Fig. 1). The low titers generated in response to TAstV-2 infection were not due to a general inability to mount a specific Ab response, as age-matched controls infected with NDV produced protective Ab titers (Fig. 1). Subsequent experiments failed to demonstrate the presence of neutralizing Abs in the sera. Furthermore, these animals were not protected when rechallenged with TAstV-2.

FIG. 1.

TAstV-2-specific IgG and IgA responses following infection. Serum and bile were isolated from TAstV-2-infected or mock-infected turkey poults at 11 and 21 dpi. Serial dilutions were incubated with recombinant TAstV-2 capsid protein complexed to microtiter plates, and IgG and IgA were detected using alkaline phosphatase-conjugated goat-anti-chicken IgG or IgA. Enzyme-linked immunosorbent assay titers are reported as the reciprocal of the dilution factor for each sample. Results are representative of three experiments.

In addition to virus-specific Abs, the levels of CD4⁺ and CD8⁺ T cells were measured using the limited tools available for the turkey model. Experiments examining the ratios of CD4⁺ to CD8⁺ cells in the spleen and peripheral blood showed no significant alterations in T-cell populations relative to controls (Table 1). Together these experiments suggest that the adaptive immune response is not critical for viral clearance in primary-infected poults.

TABLE 1.

CD4/CD8 ratios from infected turkey poults

Cell type and infection	CD4/CD8 ratio at:
	5 dpi	9 dpi	16 dpi
Peripheral blood
Control	13a	7	21
TAstV-2	11	8	21
Spleen
Control	14	7	10
TAstV-2	9	11	9

^aRatio of percentage of single positive CD4⁺ cells versus percentage of single positive CD8⁺ cells isolated from peripheral blood or spleens of infected or noninfected poults. Values reported are averages of three experiments.

TAstV-2 infection primes Mφs in vivo.

To investigate the role of Mφs in astrovirus infection, we examined NO production by adherent splenocytes isolated from infected and control poults. Adherent splenocytes from infected poults produced more NO than mock-infected controls when cultured ex vivo and stimulated with LPS. This increase in NO activity over controls was measured between 8 and 11 dpi (Fig. 2), suggesting that astrovirus infection primes Mφs in vivo, making them more readily activated upon ex vivo stimulation with LPS.

FIG. 2.

Adherent splenocytes from TAstV-2-infected poults are more responsive to LPS stimulation. Commercial turkey poults were inoculated at 5 days of age with TAstV-2 or sham inoculated with PBS. At 8 and 11 dpi, five birds from both infected and noninfected groups were sacrificed and spleens were harvested. Adherent cells were selected by culturing overnight on glass plates, collected, and cultured in a 96-well plate at 10⁶ cells/well. Cells were incubated with and without LPS (10 ng/ml) for 48 h, and then 50 μl of supernatant was removed and a Griess assay was performed. Results are expressed as the average micromolar concentration of nitrite present in the supernatant of triplicate wells. Error bars represent the 95% confidence intervals of the means. These data are representative of at least three experiments. a through e designations above bars indicate comparisons had P values of <0.05.

TAstV-2-infected intestines activate Mφs in vitro.

To study the interaction between TAstV-2 and avian Mφs, we examined the ability of TAstV-2 to stimulate the well-characterized chicken Mφ cell line HD11 (8). HD11 cells have been used extensively to examine in vitro interactions between avian Mφs and pathogenic organisms, including several different viruses (8, 22, 29). We first asked if EIF could stimulate HD11 cells to release NO. HD11 cells were treated with various dilutions of EIF and assayed for activation as determined by the Griess assay. EIF increased NO production in a dose-dependent manner (Fig. 3A). To further investigate the effects that TAstV-2 infectious material has on HD11 cell NO production, cells were stimulated with 10 μg of homogenized intestines from infected or mock-infected embryos (5 dpi) and assayed for nitrite in the supernatants. Only the infected intestine homogenate stimulated HD11 cells to produce NO, while the mock-infected intestinal homogenate did not (Fig. 3B). These results suggest the virus, or a host factor up-regulated by infection, stimulates Mφ activation.

FIG. 3.

Stimulation of avian Mφs by TAstV-2 infectious material. (A) HD11 cells (10⁵) were incubated with medium alone, 10 ng of LPS/ml, or EIF added to a final dilution of 1:5, 1:10, 1:50, 1:100, or 1:200. (B) Alternatively, HD11 cells (10⁵) were incubated with medium alone, 10 ng of LPS/ml, or 10 μg of intestinal homogenate protein from either PBS mock-infected control embryos (PBS) or TAstV-2-infected embryos (TAstV-2). All wells were brought to a final volume of 100 μl, cultured for 48 h, and then measured for nitrite using the Griess assay. Results are expressed as the average micromolar concentration of nitrite present in the supernatant of triplicate wells, and error bars represent the 95% confidence intervals of the means. These data are representative of at least three experiments. a and b designations above bars indicate comparisons had P values of <0.001.

Astrovirus directly activates Mφ production of NO.

To determine if the NO activity stimulated by EIF and infected homogenized embryo intestines was due to TAstV-2, we developed a low-pressure chromatography method to purify TAstV-2 from EIF by size exclusion. EIF separated into three major protein peaks when applied to a Sephacryl column (Fig. 4A). Molecular mass markers indicated that the first peak was slightly after the void volume of the column and contained proteins from ∼158 to 670 kDa, the second peak contained proteins from ∼17 to ∼100 kDa, and the third peak was comprised of proteins <17 kDa in size. Representative fractions from all three peaks were tested for TAstV-2 by SDS-PAGE, Western blotting, and RT-PCR. All three assays demonstrated that TAstV-2 eluted in the first peak, specifically from fractions 43 to 114 (Fig. 4B). Inoculation of these fractions into embryonated eggs demonstrated that the column-purified TAstV-2 was infectious. Similar results were obtained using TAstV-2-infected intestinal homogenates. Uninfected intestinal homogenates did not have a corresponding peak (data not shown).

FIG. 4.

Purification of TAstV-2 by size-exclusion low-pressure liquid chromatography. TAstV-2-containing fluid was fractionated using a Sephacryl CL-6B gel filtration column preequilibrated in Tris-buffered saline (pH 7.4). The column was run at a flow rate of 0.5 ml/min and 2-min fractions were collected, with a total run time of 180 min. (A) The line represents the absorbance measured at 280 nm over time. Protein markers were run on the column, and the size distribution is shown above the graph. Bars represent the micromolar concentrations of nitrite present in the supernatants of HD11 cells following 48 h of incubation with 3.5 μg of total protein from each column fraction as determined by Griess assay. Error bars represent the 95% confidence intervals of the means. (B) The presence of virus in each fraction was determined by RT-PCR. Total RNA was isolated from each fraction, and routine RT-PCR for TAstV-2 was performed. Samples were electrophoresed and visualized by ethidium bromide staining. Samples were compared to RNA isolated from TAstV-2-positive embryo intestines. Numbers above lanes represent the time in minutes.

To demonstrate that purified TAstV-2 stimulated HD11 cells, 3.5 μg of total protein from each of the column fractions was added to HD11 cells and assayed for NO (Fig. 4A). NO activity was stimulated by two groups of fractions, one corresponding to the elution of TAstV-2 and another by fractions at the end of the profile containing proteins of <17 kDa (Fig. 4A). Preliminary data suggest that the second group (fractions 200 to 262 [Fig. 4A]) contains interferons, which may account for the NO-inducing activity (data not shown).

To demonstrate that the induction of NO by EIF and column-purified TAstV-2 was not a result of contaminating endotoxin, samples were tested in the presence of the endotoxin inhibitor PMB. PMB inhibited the stimulation of Mφs by LPS but had no effect on purified TAstV-2 (Fig. 5), suggesting that TAstV-2 directly stimulates HD11 cells. There was minimal inhibition of EIF in the presence of PMB. Endotoxins present in the EIF were likely introduced during sample collection, as embryos were bacteria free. To control for the effects of exogenous endotoxin in TAstV-2 samples, PMB was added to all samples (except LPS positive controls) prior to their addition to HD11 cell cultures.

FIG. 5.

TAstV-2-mediated expression of NO is not inhibited by PMB. HD11 cells were incubated for 48 h with medium alone, LPS (10 ng/ml), 20 μl of EIF, or 2 × 10⁵ VG of column-purified TAstV-2 fraction/well with or without the endotoxin inhibitor PMB (1.5 μg/ml). Results are expressed as the average micromolar concentration of nitrite present in the supernatant of triplicate wells, and error bars represent the 95% confidence intervals of the means. These data are representative of at least three experiments. The “a” designation above the bars indicates a P value of <0.001.

TAstV-2 stimulates iNOS up-regulation.

To confirm that TAstV-2-increased NO activity was due to elevated expression of iNOS, we incubated TAstV-2-treated Mφs with inhibitors of NOS activity. Cells treated with the RNA transcription blocker ActD demonstrated a 10-fold inhibition of NO, indicating NOS activity required gene transcription (Fig. 6A). These findings were supported by experiments using the NOS inhibitor L-NMMA. L-NMMA blocked NO release by LPS, EIF, and purified TAstV-2, suggesting that induction of nitrite by TAstV-2 was due to increased NOS enzymatic activity (Fig. 6A). Finally, RT-PCR confirmed an increase in iNOS message following TAstV-2 stimulation. iNOS RNA was elevated within 4 h poststimulation in both the LPS- and TAstV-2-treated cells and remained elevated at 12 h poststimulation in the virus-treated cells (Fig. 6B).

FIG. 6.

TAstV-2-mediated expression of NO requires gene transcription and NOS activity. (A) HD11 cells were incubated for 48 h in the presence of medium alone, LPS (10 ng/ml), 20 μl of EIF, or 2 × 10⁵ VG of column-purified TAstV-2 fraction/well with or without the RNA transcription inhibitor ActD (1 μg/ml) or the NOS inhibitor L-NMMA (4 mM). Results are expressed as the average micromolar concentrations of nitrite present in the supernatants of triplicate wells. Error bars represent the 95% confidence intervals of the means. (B) TAstV-2 induces expression of iNOS RNA. HD11 cells were stimulated with medium alone, LPS (10 ng/ml), or TAstV-2 (2 × 10⁵ VG/well) for 1, 2, 4, 8, 12, 24, or 48 h. At each time point, total RNA was isolated and treated with DNase I. Samples were then analyzed by RT-PCR for the presence of iNOS RNA. Detection of the expected 649-bp product was observed. These data are representative of at least three experiments. a, b, c, and d designations above bars indicate comparisons had P values of <0.001.

Mφ activation is independent of productive TAstV-2 replication.

To determine if TAstV-2-induced iNOS was due to viral replication, cells were inoculated with TAstV-2 in complete medium and SFM in the presence of increasing concentrations of trypsin and monitored for cytopathic effect, viability, and viral replication as determined by real-time RT-PCR. Throughout the course of these experiments, no cytopathic effect was observed. Similarly, no changes in cellular proliferation or viability were detected by 5-bromo-2′-deoxyuridine incorporation or trypan blue exclusion. Examination of the cell pellets for TAstV-2 VG by real-time RT-PCR showed no significant differences in viral load compared to fixed cells (Fig. 7). Although it is not possible to rule out abortive replication, it was clear that there was no productive replication. Additional experiments to detect the presence of viral negative strand by RT-PCR and to detect viral message using in situ hybridization in inoculated HD11 cells were all negative, providing further evidence against productive viral replication.

FIG. 7.

TAstV-2 RNA detected in HD11 cells following 24, 48, and 72 h of culture. HD11 cells (5 × 10⁵) were incubated overnight and infected with TAstV-2, and cells were cultured in 500 μl of SFM containing 1 (open square), 5 (open diamond), or 10 μg (open triangle) of trypsin/ml or 500 μl of complete medium (closed triangle). As a control for input virus, cells were fixed by drying and then infected as above. These cells were incubated in SFM containing 10 μg of trypsin/ml (x). All samples were amplified in duplicate; data are representative of at least three experiments. Results are reported as the log₁₀ of the number of VG detected in 500 ng of total RNA. Cells mock infected with PBS and cultured in complete medium were used as a negative control and were determined to have no detectable virus (data not shown). Error bars represent the 95% confidence intervals of the means.

Astrovirus infection induces NO in vivo.

TAstV-2 induced NO activity in stimulated Mφs in vitro in a replication-independent manner. To determine if TAstV-2 increased NO activity in vivo, embryos were inoculated with EIF and intestinal sections were stained for the presence of nitrotyrosine. An increase in nitrated tyrosine residues directly correlates to increased concentrations of reactive oxygen species (12). Embryos infected with TAstV-2 showed a substantial increase in staining for nitrotyrosine in the lamina propria of the intestines compared to mock-infected embryos (Fig. 8). These results suggest that the host responds to astrovirus infection in part through increased NO activity.

FIG. 8.

Increased nitrotyrosine staining in TAstV-2-infected intestines. Embryos were inoculated with 100 μl of PBS (A) or 100 μl of TAstV-2 containing 10⁸ VG (B) and incubated for 5 days, and then the duodenum was isolated and stained for nitrotyrosine residues followed by FastRed detection. Panels are representative of three separate groups of infected animals. Original magnification was ×20.

NO inhibits TAstV-2 replication.

To determine the role of NO in astrovirus pathogenesis, embryos were inoculated with EIF in the presence of either the NO donor SNAP (37) or the iNOS enzyme-specific inhibitor 1400W (11). Following 5 days of incubation, the embryos were examined for the accumulation of fluid in their intestines. There was a slight reduction in the amount of fluid in the SNAP-treated infected embryos compared to the positive controls and a substantial increase in the amount of fluid in the intestines of 1400W-treated infected embryos compared to the positive controls. To evaluate viral titers, total RNA was isolated from intestines and viral titers were determined using real-time RT-PCR. Analysis of 10 ng of total embryo intestinal RNA showed a greater-than-5-log reduction in TAstV-2 viral RNA in SNAP-treated embryos compared with the positive controls and a 3-log increase in TAstV-2 in the 1400W-treated embryos (Fig. 9). These results were further supported by immunofluorescence data. Staining for viral antigen in these embryo intestines showed a significant increase in fluorescence in the 1400W-treated embryos compared with positive controls and almost no detectable viral staining in the SNAP-treated infected tissues (Fig. 10). These results represent the first description of a potential role for NO in the host response to astrovirus infection and suggest its importance in limiting or preventing viral replication.

FIG. 9.

NO inhibits TAstV-2 replication in ovo. Total RNA was isolated from control and TAstV-2-infected embryo intestines incubated with PBS, SNAP (500 μM), or 1400W (10 mg/kg) and analyzed for TAstV-2 genome levels by real-time RT-PCR. All samples were amplified in duplicate; data are representative of at least three experiments. Results are reported as the log₁₀ of the number of VG detected in 10 ng of total RNA. Error bars represent the 95% confidence intervals of the means. ^a,b,c, P < 0.001.

FIG. 10.

NO affects levels of TAstV-2 antigen staining in ovo. The photomicrograph shows the distribution of specific immunofluorescent staining against astrovirus capsid antigen of 100 μl of PBS (A), 5 × 10⁶ VG of TAstV-2 (B), 5 × 10⁶ VG TAstV-2 plus 500 μM SNAP (C), or 5 × 10⁶ VG TAstV-2 plus 10 mg of 1400W/kg of embryo intestine at 5 dpi (D). Panels are representative of intestines of three embryos. Original magnification was ×10.

DISCUSSION

Our turkey model provides a system to ask basic questions about the host-pathogen interactions of astrovirus and young animals, both in vitro and in vivo. In this study we examined the effects of astrovirus infection on both the innate and adaptive immune responses. In our system, infected animals demonstrated no differences in the numbers of circulating or splenic lymphocytes or the ratios of CD4⁺ to CD8⁺ cells and minimal production of astrovirus-specific Abs. These data indicate a nominal role for adaptive immunity during primary infection. However, we did notice an increase in Mφ activation and release of NO in response to astrovirus infection. Increases in NO activity were measured both in vivo and in vitro and were shown to be involved in limiting viral replication, suggesting that the innate immune system, specifically Mφs and iNOS, play a key role in controlling astrovirus replication.

We were surprised to find little evidence of an adaptive immune response following astrovirus infection in turkeys (Fig. 1; Table 1). Both B cells and T cells respond to human astrovirus infection. Virus-neutralizing Abs are considered key to astrovirus resistance. Human volunteer studies have demonstrated that subjects with preexisting Ab titers do not show signs of astrovirus disease (19). The protective role of virus-specific Abs has also been demonstrated therapeutically. Intravenous immunoglobulin therapy has been used to treat persistent astrovirus infections in immune-compromised patients (4, 38). Molberg et al. demonstrated astrovirus-specific Th1 CD4⁺ T cells in the intestines of healthy adults (26). However, these studies primarily involved healthy adults and demonstrated factors involved in protecting the host from repeated infection. Astrovirus infections are typically associated with immature or infirmed immune systems. In these people, the role of humoral and cellular immunity is hindered or nonexistent; however, astroviruses seldom establish persistent infections. As a result, we were interested in studying the host response to primary infection in a young animal. Understanding the mechanisms involved in viral clearance and disease resolution under these circumstances would greatly advance our understanding of viral enteritis and potential general therapies. The lack of acquired immunity to TAstV-2 infection suggests the turkey model may reflect the host response in the noncompetent immune host.

Our results suggest that in the absence of an adaptive immune response, the innate immune system may be critical in controlling the disease. We observed that adherent splenocytes from the infected animals produced more NO when stimulated with LPS than mock-infected controls (Fig. 2). These results suggested that adherent splenocytes were effectively primed by the astrovirus infection, making them more susceptible to secondary activation with LPS (1). In contrast, HD11 cells respond directly to TAstV-2 inoculation by releasing NO independent of a secondary activation. It is well documented that primary Mφs require two signals for activation, a priming signal and an activating signal. Cells distal to the site of infection would not be activated but would have received a priming signal, making them more reactive to secondary stimulation. In contrast, Mφ cell lines do not require two stimuli for activation and respond to inoculation with viruses (29) and bacterial products by releasing NO, and they are more easily activated than primary avian cells (8).

TAstV-2 increased iNOS activity using EIF, homogenized intestines, and purified virus in vitro (Fig. 2 and 4) through a replication-independent mechanism (Fig. 6). NO is important in several viral diseases, and its effects can range from propathogen to prohost (23). Production of NO enhanced human immunodeficiency virus replication (5) and increased the inflammation and pneumonia associated with influenza infection in mice (2). Inversely, the production of NO inhibited viral replication and delayed death in rabies virus- and coxsackievirus-infected mice (10, 35, 36, 39). The increased expression of iNOS by TAstV-2 led us to speculate that NO may play a role in astrovirus pathogenesis.

To determine the effect of NO on astrovirus infection, we assayed for increased NO activity in infected tissues. Examination of infected and control embryo intestines demonstrated increased staining for nitrotyrosine, indicating reactive oxygen species were generated during infection (Fig. 7). The differences in nitrotyrosine staining between controls and infected embryos were most notable at 5 dpi. This observation was significant, since fluid accumulation in infected embryos was not detected until 5 dpi, suggesting NO may be involved in fluid accumulation by affecting enterocyte barrier function and ion transport (6, 28, 31). The source of the NO in vivo is unknown. It is possible that Mφs may be responsible; however, enterocytes produce iNOS in response to stimuli (27) and express many of the pattern recognition receptors of the innate immune system (7, 13). However, given the lack of phenotypic markers available for turkeys, we are currently unable to identify the cell type responsible for the increased NO activity.

Regardless of the source, NO was important for controlling viral replication and viral clearance. This was supported by our finding that exogenous NO dramatically inhibited the replication of TAstV-2 in ovo, and inhibition of the iNOS response led to increased viral titers (Fig. 8 and 9). We are currently examining the specific mechanisms involved in iNOS expression and the mechanisms by which NO inhibits astrovirus replication. Studies are under way to determine if NO inactivates TAstV-2 directly or if increased NO levels lead to increased expression of interferon or other cytokines, which may establish an antiviral state within the tissue. Preliminary studies suggest EIF contains interferon activity and that exogenous interferon can limit viral replication. In addition, we are examining the role of NO in astrovirus pathogenesis. Previously, we demonstrated levels of active transforming growth factor-β increased significantly in infected poults and embryos. Transforming growth factor-β is typically considered antiinflammatory, and it down-regulates iNOS expression (21). Determining the relationship between these powerful immune-modulating compounds, in the context of astrovirus infection, will increase our understanding of astrovirus disease and of viral enteritis in general.

These data represent the first report, to our knowledge, of an astrovirus-mediated activation of Mφs, as well as a description of an important role for Mφs and innate immunity in the host response to astrovirus infection. The development of effective vaccines has been and remains the most sought after therapeutic or prevention measure to viral diseases. However, given the acute nature of the disease and the immune status of those most affected, therapies based on an understanding of the innate immune response may be more efficacious.