Background
Viruses of the genus A
lphavirus, family
Togaviridae, are significant human pathogens maintained by an enzootic cycle between mosquitoes and vertebrates. They can be generally divided by geography and pathology into the Old World alphaviruses such as Chikungunya, Ross River, and O’nyong-nyong viruses, which are usually associated with a mild to severe arthralgic disease in humans, and the New World alphaviruses such as western equine encephalitis virus (WEEV), Venezuelan equine encephalitis virus (VEEV), and eastern equine encephalitis virus (EEEV), which cause a mild to severe encephalitis in both humans and equids. Previously, EEEV was further divided into the highly pathogenic North American antigenic group (NA EEEV) and the three Central and South American antigenic groups, which only occasionally cause human disease (SA EEEV) [
1,
2]. Recent changes in taxonomy identify NA EEEV strains as EEEV and SA EEEV are a new species designated Madariaga viruses (II-IV) [
1]
Because of its high mortality (36-75 %) [
3], lack of a licensed human vaccine and effective antiviral therapy, and its potential use as a biological weapon, EEEV is listed as a category B agent by the National Institute of Allergy and Infectious Diseases (NIAID). Therefore, research directed towards understanding the pathogenesis of EEEV, the difference in virulence between EEEV and Madariaga virus strains, and the development of a safe and effective vaccine and antiviral treatment for humans is essential. However, research has been hampered by the fact that mice develop age-dependent resistance to peripheral infection. Recently, EEEV, FL 93–939 was shown to be virulent in adult mice when administered peripherally; however there is limited information regarding the pathogenesis of this strain, with most studies focusing on the mechanism of decreased type I IFN induction in infected mice [
4‐
6]. Characterizing the clinical course and outcome of EEEV strain FL93-939 in mice using various routes of infection is an important first step in understanding the pathogenesis of the NA EEEV strains, which is paramount when developing a vaccine and/or therapeutic to prevent the disease.
A comprehensive study was performed to determine and compare the clinical course of NA EEEV FL93-939 infection of mice following infection by the intranasal (IN), aerosol (AE), and subcutaneous (SC) routes. The AE route is the likely route of exposure to an EEEV bioweapon and one way by which laboratory workers may be accidentally exposed. However, many investigators do not have the equipment necessary to conduct AE studies and often use IN as a proxy exposure method. Identifying any differences in pathogenesis will be important for the proper interpretation of future studies utilizing either of these routes. The SC route of exposure is meant to mimic two unintentional exposure scenarios: laboratory exposure via contaminated needle puncture and natural infection from the bite of an infected mosquito, with the caveat that needle inoculation is not a perfect substitute for a mosquito bite [
7].
Here the clinical course of disease, including clinical sign onset and duration, as well as complete blood count, serum analysis, and tissue distribution of virus over time following infection is presented. A subsequent manuscript will present tissue pathology associated with the various routes of exposure.
Discussion
To better understand the pathogenesis of EEEV strain FL93-939 in mice, several studies were conducted to evaluate the differences between 3 routes of infection: intranasal, aerosol, and subcutaneous. Although the intranasal and aerosol routes were similar, there were important differences noted. While animals in both groups lost weight between 3–4 dpi, those animals in the AE study displayed clinical signs of disease at 3 dpi compared to 4 dpi in the IN study. The clinical signs were so severe in the majority of animals in the AE study at 4 dpi that the study was terminated early. Clinical signs of disease did not appear until 5 dpi in the SC study. Interestingly though, the absolute dose given per mouse was similar between the IN and SC routes, 1300 pfu and 1000 pfu respectively, yet the outcome was drastically different with regard to clinical disease onset and viral tissue distribution and titers. This is likely due to differences in inoculation site, the resultant immune response, and the relative lethal dose given; 100LD
50 in the IN study compared to 30LD
50 in the SQ study. The clinical signs noted in the SC study were similar yet less severe than previously reported [
10]; however, mice in the Vogel et al. study had severe clinical signs by 4 dpi. The difference in onset of clinical signs and severity noted in the Vogel study may be attributed to the virus strain (FL91–4679), higher inoculum (10
5 pfu), the mouse strain (C57/BL6), or the age of the mice (5-weeks) used in that study, or most likely a combination of these factors. The clinical findings in our SC study more closely paralleled those noted recently by Gardner et al. [
11] in which they did not observe signs of disease until 6 dpi following a subcutaneous exposure and weight loss was minimal over the course of the study.
Telemetry was used previously to study non-human primates infected with EEEV by the aerosol route [
12]. However, telemetry had not been used previously to study EEEV infection in mice, or to compare routes of infection. As in the non-human primate study, telemetry proved to be a useful tool in determining onset of clinical disease in mice. Fever was detected at 3 dpi in both the IN and AE studies, which coincided with the onset of mild signs (ruffled fur). Additionally, obvious changes in diurnal patterns in both temperature and activity coincided with the onset of more severe clinical signs of disease. This study highlighted the importance of monitoring such parameters and revealed that the clinical signs in mice, similar to those seen in humans, included fever prior to or at the onset of more obvious clinical signs of disease, such as ruffled fur and lethargy.
A complete blood count was performed on animals from the AE and SC studies; however, the results were not specific or predictive of outcome. Most animals had a leukopenia characterized by a lymphopenia at 2–3 dpi, which can be indicative of a viral infection. These results were similar to those reported by Adams et al. [
13] in which marmosets infected with EEEV had a leukopenia 1 dpi, but a concomitant decrease in neutrophils, lymphocytes and monocytes. In the marmosets, this blood profile rapidly changed to a leukocytosis, characterized by a neutrophilia, lymphocytosis and a monocytosis by 3–4 dpi. While there was a rise in leukocytes in our SC study after 3 dpi, the number of leukocytes remained within the normal range for the study duration. It is difficult to compare CBC results in research models to that observed in human cases, since the infectious dose and time from inoculation to presentation in humans is typically not known.
Cytokine analysis in EEEV infected research models has not been reported. While there were several pro-inflammatory cytokines and chemokines that were elevated to statistically significant levels relative to saline-treated control animals, most of the elevated cyto/chemokine levels were only slightly increased from baseline. However, temporal comparison of cytokine levels between various routes of infection revealed some interesting results. That no significant differences were observed among any of the 25 cytokines and chemokines tested between aerosol and intranasal infection at any time point suggests these routes of infection induced very similar immune responses. However, several differences were noted between subcutaneous and intranasal or aerosol infections. IFN-γ, MIP-1β (a chemoattractant for macrophages and NK cells), RANTES (a chemokine for T-cells), and MIG (a T-cell attractant and activator) were increased one day following subcutaneous infection relative to aerosol or intranasal infection, followed by a return to baseline one day later when the same cyto/chemokines become increased in aerosol or intranasal infection. sCD62E, a soluble analogue of the E-selectin cell adhesion molecule, may also have been increased in both IN and AE infected animals early in infection. However, the relatively high level observed in AE-infected day 0 controls prevents a conclusive association. Overall, these data suggest a mild induction of certain pro-inflammatory cytokines and chemokines after infection. It is possible that the earlier spike in IFN-γ, MIP-1β, RANTES, and MIG levels after subcutaneous infection contributed to the decreased virulence seen in these mice compared to intranasal or aerosol infection. It is noteworthy that this study focused on serum cytokine and chemokine levels, which may reflect proteins produced by a number of sources, including circulating immune cells and various target tissues. Future studies should focus on determining tissue cytokine levels to further elucidate their source and potential role in infection and disease control or exacerbation.
In these studies, virus titration of tissue homogenates was used to characterize the temporal progression of EEEV FL93-939 from the site of exposure/inoculation to the CNS of infected mice. In IN infected animals, virus was first detected in the blood and the brain homogenate at 1 dpi. By 2 dpi, viremia peaked and the virus titer in the brain rapidly increased. These data could be suggestive of a vascular route of neuroinvasion or invasion of the CNS via the olfactory route as observed in guinea pigs exposed to aerosolized EEEV [
14]. However, further studies using IHC strongly indicate EEEV neuroinvasion following IN exposure follows the olfactory route [
15].
The results for the AE study were very similar to the results of the IN study, with the important exception that virus was detected in the brain by titer at only 6 hpi following aerosol exposure. The aerosol delivery method likely allowed more virus contact with olfactory neurons, thus facilitating the earlier viral invasion of the olfactory bulb and subsequent transport to the olfactory tract and beyond [
15]. These results support the clinical findings of more rapid and severe disease onset in this study.
The results of the SC study were less clear. Despite convincing evidence of virus replication near the site of inoculation and viremia in 30-40 % of animals at the early time points, viral infection of the brain was not consistently observed. While virus was first detected in the brain by plaque assay at 1 dpi (1 of 5 animals, 20 %), virus was only detected in the brain of 10 of 60 (17 %) of animals from 3–8 dpi. This may be due to variation in LD50 dose, immune response, or route of neuroinvasion.
In humans, the incubation period following natural EEEV infection is short, usually 4–10 days. Systemic infection is often characterized by abrupt onset of chills and fever followed by malaise, arthralgia, and myalgia. Typically these are difficult parameters to measure in animals; however, telemetry allowed for monitoring of temperature and activity and fever and decreased activity (lethargy or malaise) were noted in many infected animals. Clinical signs of encephalitis in humans include abrupt onset of severe fever, intense headache, irritability, restlessness, drowsiness, anorexia, nausea, vomiting, diarrhea, cyanosis, convulsions, and coma. Again, while most of these clinical signs cannot be evaluated in mice, marked lethargy and tremors were noted in some infected animals. In humans, death usually occurs within 2–14 days after the onset of clinical signs [
16]. In these studies, mice generally were moribund or succumbed to infection within 2–4 days following the onset of clinical signs.
Methods
Mice
Specific pathogen free 8–10 week-old female BALB/c mice (NCI, Frederick, MD) were housed in cages equipped with microisolators and were provided food and water ad libitum throughout the study. The room temperature was maintained at 23 ± 1 °C and periods of light and dark were on a 12 h cycle. Mice were acclimated for 1 week after which 10 animals in each study were surgically implanted with intraperitoneal telemetry devices (TA-F20, Data Sciences International, St. Paul, MN) to monitor body temperature and activity. Animals received 1 week post-operative recovery, thus weighed approximately 20 gm, and were 10–13 weeks old at the time of exposure. For the portions of the study involving live EEEV, mice were housed in a biosafety level 3 (BSL-3) facility. Human end points were used during all mouse studies. Research was conducted under an IACUC approved protocol in compliance with the Animal Welfare Act, Public Health Service Policy, and other Federal statutes and regulations relating to animals and experiments involving animals. The facility where the research was conducted is accredited by the Association for the Assessment and Accreditation of Laboratory Animal Care International and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals, National Research Council, 2011.
Virus
EEEV strain FL93-939 was obtained from Dr. Scott Weaver, UTMB, Galveston, TX. A sucrose-purified working stock was prepared from seed stock (P1) through an additional passage (P2) in Vero cells. Virus titer was determined by standard plaque assay on Vero cell monolayers. Virus was aliquoted and frozen at −70 to −80 °C prior to use. Challenge virus was diluted in either Eagle’s minimum essential medium (EMEM) (Cellgro, Mediatech, Inc., Manassas, VA) or sterilized phosphate buffered saline (PBS) (GIBCO Invitrogen Corp., Grand Island, NY).
Experimental design
Groups of 10 mice were exposed to approximately 100LD
50 of EEEV strain FL93-939 by either the intranasal, aerosol or subcutaneous route. For the intranasal route of exposure, virus dose was prepared in a 20 μL volume in sterilized PBS. Control mice received only sterilized PBS. Mice were briefly anesthetized with isoflurane using the IMPAC
6 (VetEquip, Inc., Pleasanton, CA) and given 10 μL of challenge virus per nostril. For the aerosol route of exposure, virus dose was prepared in a 10 ml volume in EMEM. Control mice were exposed to diluent only. Aerosol exposures were conducted in a whole-body bioaerosol exposure system. A Collison nebulizer (BGI, Inc., Waltham, MA) was used to generate small (1 μm mass median aerodynamic diameter) diameter particles for each acute 10 min exposure. Briefly, mice were placed in wire cages, which were then placed into a chamber where they were exposed to aerosolized virus for 10 min. ‘Presented’ dose was estimated by calculating the respiratory minute volume (V
m) using Guyton’s formula [
17], expressed as V
m = 2.10 x W
b
0.75 where W
b = body weight (gm) based on the average group weights the day of exposure. The presented dose was then calculated by multiplying the estimated total volume (V
t) of experimental atmosphere inhaled by each animal (V
t = V
m x length of exposure) by the empirically determined exposure concentration (C
e) (‘presented dose’ = C
e x V
t). Exposure concentration, expressed in plaque-forming units (PFU)/L, was determined by isokinetic sampling of the chamber with an all-glass impinger (AGI) (Ace Glass, Vineland, NJ). Samples were titrated by standard plaque assay on Vero cell monolayers [
14]. For the subcutaneous route of exposure, virus dose was prepared in a 10 μL volume in EMEM. Mice were inoculated in the left foot pad in order to track viral replication in the surrounding tissue and draining lymph node (popliteal lymph node). Control mice received diluent only. Challenge virus preparations were back-titrated by standard plaque assay using Vero cells. Mice from the intranasal and aerosol studies were euthanized at pre-determined time points: 6, 12, 24, 48, 72, 96, and 120 hours post-infection (hpi). In addition to the previous listed time points, mice in the subcutaneous study were also euthanized at 144, 168, and 192 hpi. At the time of euthanasia, mice were anesthetized with mouse K-A-X (50 mg ketamine (Fort Dodge Animal Health, Fort Dodge, IA), 0.5 mg acepromazine (Boehringer Ingelheim, Ridgefield, CT), and 5.5 mg xylazine (Lloyd Laboratories, Walnut, CA)) given intraperitoneally at a dose of 0.2 ml per 20 gm. Mice were euthanized by exsanguination via cardiac puncture and whole blood samples were collected for CBC analysis, while serum samples were collected for viral titer and cytokine analysis. Five mice from each time point were perfused with PBS and tissues were individually collected and frozen for viral titer analysis.
Acquisition and analysis of telemetry data
All telemetry data was collected using the DSI DataQuest ARTM™ software. The system was programmed to sample body temperature and physical activity for a 20 sec period every 30 min. Baseline data was collected for 2 days. Data collection continued until euthanasia or the end of the study. Pre-exposure temperature data was used to develop a baseline period to fit an autoregressive integrated moving average (ARIMA) model. Forecasted values for the post-exposure period were based on the baseline extrapolated forward in time using SAS ETS (v. 9.2). Residual changes were determined by subtracting the predicted value from the actual value recorded for each time point. For temperature, residual changes greater than two standard deviations were used to compute fever duration (number of hours of significant temperature elevation), fever hours (sum of the significant temperature elevations), and average fever elevation (fever hours divided by fever duration in hours). Only time periods consisting of two or more consecutive time points of elevated temperature were used in the analysis.
CBC analysis
A complete blood count (CBC) was determined on whole blood samples collected at the time of euthanasia. Samples were run on an Abbott CELL-DYN 3700 with veterinary package (Abbott Laboratories, Abbott Park, IL) on the same day as collection. This instrument produces a differential white blood cell count based on size, internal granularity, and nuclear content using optical and impedance technology. T-tests with stepdown Bonferroni adjustment were used to compare mean levels of blood parameters between infected groups and uninfected control groups at various time points.
Cytokine analysis
Cytokines/chemokines were measured on selected serum samples using BD™ Cytometric Bead Array mouse soluble protein flex sets (BD Biosciences, San Jose, CA) read on a BD FACSCanto II flow cytometer as per manufacturer’s instructions. Twenty-five soluble proteins were measured (sCD62E, sCD62L, G-CSF, GM-CSF, IFN-γ, IL-1α, IL-1β, IL-2, I L-3, IL-4, IL-5, IL-6, IL-9, IL-10, IL-12/IL23p40, IL-13, IL-17A, IL-21, KC/CXCL1, MCP-1/CCL2, MIG.CXCL9, MIP-1α/CCL3, MIP-1β/CCL4, RANTES/CCL5, and TNF) and results were analyzed using FCAP Array software (BD Biosciences).
Mucosal secretions
A bronchoalveolar lavage (BAL) and nasopharyngeal flush (NF) were performed using 0.5 ml of sterile PBS for each. Briefly, under deep anesthesia, the trachea was exposed and an 18G needle was inserted toward the lower or upper respiratory tract, respectively. PBS was flushed into the lungs and aspirated for BAL or through the nares and/or oropharynx for nasopharyngeal flushes. Mice were then euthanized by exsanguination via cardiac puncture.
Virus titrations
For determination of EEEV titers, tissue samples were homogenized using a mini-bead beater and 1–2 stainless steel beads (3.2 mm diameter) (BioSpec Products, Inc., Bartlesville, OK) and 500 μL of complete medium. Homogenized samples were centrifuged for 10 min at 10,000 rpm in a table-top centrifuge and supernatants were collected and stored at −70 to −80 °C until virus titration. Titration of virus was performed by standard plaque assay on Vero cell monolayers. Briefly, supernatant from homogenized tissues, serum, BAL, nasal flush, or AGI samples were serially diluted in EMEM ( Cellgro, Mediatech, Inc.) containing 5 % fetal bovine serum (FBS) (GIBCO Invitrogen Corp.), 1 % penicillin (20,000 IU/ml)-streptomycin sulfate (20,000 μg/ml), 1 % non-essential amino acids (NEAA) (Sigma Aldrich Company, Inc., St. Louis, MO), 1 % 200 mM L-glutamine (Thermo Scientific, Logan, UT), and 0.1 % gentamicin solution (Sigma Aldrich Company, Inc.). Diluted samples were then added in duplicate to 6-well plates containing confluent monolayer of Vero cells (African green monkey kidney cells) which were incubated at 37 °C for 1 hour, with rocking every 15 min. Following the incubation period, a 0.5 % agarose overlay in 2x EBME solution (GIBCO, Life Technologies) with HEPES and 10 % FBS, 1 % L-glutamine, 1 % NEAA, 1 % penicillin-streptomycin sulfate, and 0.1 % gentamicin was added, and plated were incubated at 37 °C at 5 % CO2 for 24 hr. Thereafter, a second agarose overlay in 2x EBME containing supplements and 5 % neutral red was added. The plates were again incubated at 37 °C at 5 % CO2 for 24 hr. Defined plaques (neutral red exclusion areas) were then counted. The limit of detection for this assay was 5 pfu/ml.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
SH was responsible for experimental design, experimental execution, data collection, data analysis, and manuscript preparation. EM was responsible for the experimental execution, data collection, data analysis, and manuscript preparation. RB participated in the experimental execution and data collection. DF participated in the data analysis. CL participated in the experimental execution and data collection. JC participated in the experimental execution and data collection. LE participated in the experimental execution and data collection. KS participated in the experimental execution, data collection and data analysis. REC participated in the experimental execution, data collection, and data analysis. SB participated in data collection and data analysis. RM was responsible for the experimental design, data analysis and manuscript preparation. PG was responsible for the experimental design, experimental execution, data collection; data analysis, and manuscript preparation. All authors read and approved the final manuscript.