Upper-respiratory tract infections (URTI) are the most common infectious condition affecting athletes undertaking intensive exercise (2) and may cause serious disruption to training and competition programs (13,21). Epidemiological studies suggest that intensive training over short or long periods of time are associated with an increased risk of URTI, with elite athletes at greater risk than those undertaking more moderate training regimes (13,17,20). As in the general population, the most common cause of URTI in athletes is viral infection (23). This generally accepted belief is mainly based on physician diagnosis of clinical symptoms in athletes but rarely has this notion been supported by serological investigation. In most studies of URTI in athletes, the episodes were self-reported and not physician verified. This has led to uncertainty about the etiology of URTI in elite athletes.
The classical presentation of viral URTI is for upper-respiratory symptoms to begin 1–2 d after infection and to last 1–2 wk (15), although clearance of the virus may take longer (29). The majority of viruses enter the body via the respiratory tract, infect selected cell types, and replicate within those cells to infect other cells (23). The symptoms reflect disruption of both the normal functions of the colonised cells and the response of the immune system to contain the viral infection (23). These processes cause local inflammation and systemic consequences (e.g., aches, fever, and fatigue) that persist after the disappearance of upper-respiratory symptoms (URS).
Athletes with URS do not always present with the classical symptoms of viral URTI. Symptoms may only persist for a few days, and the severity may not be sufficient to interrupt training (observations of authors). At the other end of the spectrum, a small number of elite athletes may experience persistent or recurring URS and/or fatigue associated with intensive training (3,10,13). These clinical observations suggested that the URS may be associated with episodes of viral reactivation in the upper airways as opposed to primary infections. Epstein-Barr virus (EBV) was the most likely candidate as the biology of EBV infection is consistent with the clinical presentation and time course of URS in elite athletes. In a recent study, nonprimary EBV infection has also been proposed as a potential cause of oropharyngotonsillitis in young adults (32). A pilot study, to test this hypothesis of viral reactivation, demonstrated that EBV-DNA was occasionally detected in saliva from elite swimmers over a 7-month training season (9), often before the appearance of URS.
EBV is a member of the human herpes virus group and is predominantly an asymptomatic infection but may present as infectious mononucleosis. Primary infection usually occurs in childhood, with a seroprevalence of approximately 90% in adults. EBV infects epithelial cells and B-lymphocytes (B-cells) in the oropharynx. The virus establishes lifelong persistence by immortalizing normal resting B-lymphocytes and transforming them into permanent lymphoblastoid cell lines (1). The EBV-infected memory B-lymphocytes are controlled by EBV-specific cytotoxic T-lymphocytes (CTLs) (22), and removal of the immune T-lymphocytes (T-cells) allows transformation of the B-lymphocytes and expression of the EBV viral-DNA (1). The latent virus expression into saliva usually lasts 24–48 h (27). Intensive exercise, particularly training programs undertaken by conditioned athletes, causes transient suppression of cytotoxic and suppressor T-lymphocytes (5) and T-cell function (4), as well as suppression of humoral mucosal immunity (7).
Various stressors on the immune system have the potential to disrupt the tight T-lymphocyte control of EBV-transformed B-lymphocytes resulting in EBV shedding. Reactivation of latent herpes viruses, including EBV, has been detected in response to academic stress (6), during prolonged isolation in Antarctica (27), before spaceflight (16,19), and in immune deficiency states (14). The aim of this study was to determine whether the stresses associated with athletic training result in EBV viral reactivation in a cohort of elite swimmers during a month of intensive training. A specific objective was to characterize the time course of appearance of EBV-DNA in saliva and any associations with mucosal immune suppression and the appearance of URS. The study hypothesis was that the transient immune suppression associated with intensive exercise allows window periods for EBV viral reactivation, resulting in the symptoms that mimic viral URTI.
METHODS
Subjects.
Fourteen highly trained men aged 18–27 yr (mean age 21.4 yr, SD 2.3), from the Australian Institute of Sport (AIS) Swimming Team were recruited to the study. All the swimmers had trained competitively for at least 5 yr and competed successfully at international or national level competition. The mean world ranking of the group was 49 (ranging from number 2 to number 100 world ranked). The study was conducted with the approval of the ethics committees of the Australian Institute of Sport and the University of Newcastle. Written informed consent was obtained from all subjects.
Study design.
To address the hypothesis, the swimmers were studied while undertaking 30 d of intensive training during the southern-hemisphere spring-summer. The training mileage for the study ranged from 197 to 232 km over the 30 d, equating to a mean volume of 8000–9300 m of swimming per day (i.e., 3–4 h·d−1 of physical exercise). The study was designed to examine the time course of detection of EBV-DNA in saliva in relation to the appearance of URS and suppression of mucosal immunity, examined by measuring salivary IgA levels. Unstimulated whole-mixed saliva was collected every second or third day by gently spitting (1 mL) into a 5-mL plastic collection tube. Some swimmers were not available on all days for saliva collection and the number of samples available for assay from each subject is indicated in Table 1 (maximum of 13 d). The samples were collected 2 h postprandially immediately before the afternoon training session (1600 h) and stored frozen at −70°C. Saliva was assayed for both total IgA and EBV-DNA. A standardized daily illness log developed by the AIS was used to assess the symptoms associated with each episode of upper-respiratory illness. Blood was collected by standard venepuncture at the beginning of the study period and serum separated for EBV serology.
;)
Table 1: EBV serology status, number and percentage of days with EBV-DNA detected, number of episodes, and number of days of URS consistent with URTI in a cohort of elite swimmers over a 1-month intensive-training period.
EBV serology.
The EBV serology status of each swimmer was determined before the study. The serum was tested for evidence of recent or past infection with EBV. Commercial ELISA kits (Gull Laboratories, Salt Lake City, UT) for IgM and IgG antibodies to EBV viral capsid antigen and IgG antibody to EBV nuclear antigen-1 were assayed using a BEPIII Immunoanalyser (Dade-Behring, Marburg, Germany).
EBV-DNA.
EBV-DNA in saliva was semiquantitated using a modification of the PCR amplification and electrophoresis detection methods developed by Rozenberg and Lebon (24). Genomic DNA was isolated from 250 μL of each saliva specimen by using a modified version of the gram negative protocol specified for the Wizard Genomic DNA purification kit (Promega, Madison, WI). Initial centrifugation at 14,000 g at 4°C was extended to 30 min. Commercial primers (Gibco-BRL, Glen Waverley, Australia) were used to amplify viruses of the Herpesviridae family, including EBV, cytomegalovirus (CMV), and herpes simplex virus (HSV) types 1 and 2. The oligonucleotide primers amplified a selected conserved fragment of the DNA polymerase genes in the herpes virus group (sense: 5′ CGA CTT TGC CAG CCT GTA CC; antisense: 5′ AGT CCG TGT CCC CGT AGA TG). A 5-μL sample of DNA was added to 45 μL of PCR reaction mixture containing 3 μL of MgCl2 (25 mM, Promega), 5 μL of concentrated (×10) Promega reaction buffer, 1 μL of each deoxynucleotides (dATP, dGPT, dTTP, and dCTP) (10 mM, Promega), 2.5 μL of each oligonucleotide primer (20 μM, Gibco-BRL), 0.2 μL of Taq DNA polymerase (Promega), and 31 μL of diethyl pyrocarbonate (Sigma-Aldrich, Sydney, Australia) treated distilled water (DEPC). The Promega reaction buffer contained Tris-HCl (50 mM, pH 8.0), NaCl (100 mM), ethylene diamine tetra-acetic acid (0.1 mM, EDTA), 50% glycerol, and 1% Triton X-100. The mixture was overlaid with 50 μL of mineral oil (Sigma-Aldrich) and amplified using an automated thermal cycler (Hybaid Omnigene, Ashford, Middlesex, England). The cycle consisted of 1 min for denaturing at 94°C, 1 min for annealing at 60°C, and 1 min for elongation at 72°C, and repeated for 40 cycles. A 2-μL sample of loading buffer (Promega) was added to 8 μL of each amplified DNA product and loaded onto a 2% agarose gel (Progen, Darra, Australia) and electrophoresed at 125 V for 110 min (Pharmacia, Upsalla, Sweden). The gel was stained with a 1 μg·mL−1 ethidium bromide (BDH, Kilsyth, Australia) solution, viewed under an UV transilluminator (Ultra-Lum, Claremont, CA) and photographed for permanent record using a Polaroid MP-4 camera (Polaroid Corporation, Cambridge, MA). and Polaroid Polapan 665 PN film exposed for 30 s. A 100-bp ladder (Promega) was included on the gel to determine the molecular weight of the DNA product. The EBV-DNA in positive samples was graded visually as +/++/+++ against the concentration in the EBV positive-culture control to give a semiquantitative measure of the amount of viral product detected on the electrophoresis gel.
Samples with detectable DNA product were retested before and after digestion with the restriction enzyme Bam H1 (Promega) to distinguish EBV-DNA from HSV-DNA of similar molecular weight. A 10 μL sample of PCR product was added to 1 μL of Bam H1 enzyme, 2 μL of enzyme buffer, and 7 μL of DEPC-treated water and incubated for 4 h at 37°C. The enzyme buffer (Promega) contained Tris-HCl (10 mM, pH7.4), KCl (300 mM), MgCl2 (5 mM), EDTA (0.1 mM), dithiothreitol (1 mM), 0.5 mg·mL−1 bovine serum albumin, and 50% glycerol. This restriction analysis cleaves the 524-bp EBV amplified product into two fragments of 277 and 247 bp. HSV-1 is indigestible by Bam H1, whereas HSV-2 is cleaved into two fragments of 293 and 225 bp. The Bam H1 patterns clearly distinguish EBV from HSV. The 589-bp CMV is distinguishable without the use of restriction enzymes. A 100-bp ladder (Promega) and the EBV positive culture control were included on the gel to determine the molecular weight of the DNA products.
Precautions were taken to avoid contamination in the PCR procedure by physical separation of the pre-PCR and post-PCR processes, avoidance of aerosols during pipetting steps, and contamination from technicians who may be EBV seropositive. Each swimmer’s saliva was processed separately to avoid the risk of cross-contamination between subjects. EBV-positive controls were extracted after other samples had been capped and removed from the work area. Positive controls included in each assay were saliva from a patient with a recent diagnosis of infectious mononucleosis and an EBV culture from the EBV-producing, monkey cell line B95-8 (American Type Culture Collection, Rockville, MD). Saliva from seronegative subjects was used as a negative control.
Salivary IgA.
Total salivary IgA was assayed using a modification of a previously described in-house ELISA (11). The total salivary IgA measure is the average of triplicate samples. The calibration material was Human Protein Calibrator (Dako Immunoglobulins, Marburg, Germany) referenced against BCR CRM470. High- and low-level saliva controls were included in all assays. In addition, saliva from an IgA-deficient individual was included as a negative IgA control. The between-run coefficients of variation for the assay were 12% at 40 mg·L−1 and 8% at 150 mg·L−1. The limit of detection was 5 mg·L−1.
Infection records.
Symptoms of URTI were recorded daily by interview at the time of the saliva collection and included sore throat, cough, runny nose, fever, and headache. The attending sports physician excluded URS not consistent with viral URTI. Two subjects had hay fever on 5 days but no evidence of URTI.
Statistical analysis.
Fisher’s exact test was used to assess the correlation between EBV serology and URS. This test is the appropriate measure of association when sample sizes are small. The study sample size of 14 swimmers provided 80% power to detect a 50% or greater difference in the proportions of swimmers with URS in the EBV seropositive group compared with the EBV seronegative group. The log IgA values at different time periods were assessed using repeated measures ANOVA for swimmers with complete data. Logged values were used so that the data were normally distributed, as required for ANOVA analysis. A repeated-measures analysis took into account the possibility that some IgA measures were correlated, because there were multiple measures from each swimmer. The study design provided a minimum of 80% power to detect a doubling of salivary IgA levels (100% increase) from pre-URS levels to peak-URS levels. Pairwise comparisons of the IgA values were made using the Wilcoxon matched-pairs signed rank test, the nonparametric equivalent to the paired t-test, appropriate for the nonnormal distribution of the IgA values. Significance levels were set at P < 0.05 for all tests.
RESULTS
EBV serology.
The results of the EBV serology indicated that 11 of 14 swimmers (79%) had evidence of past infection with EBV and three had no evidence of prior or current EBV infection (Table 2).
Table 2: Preexercise salivary IgA levels before the appearance of URS, peak and trough IgA levels after the URS, and days to peak and trough levels in swimmers recording URS.
EBV-DNA.
The restriction enzyme digests revealed no evidence of CMV, HSV1, or HSV2 in any of the saliva samples. EBV-DNA was detected in saliva on at least one occasion during the study period in seven swimmers, all of whom were seropositive for EBV (Table 1). The remaining four seropositive swimmers, and the three seronegative swimmers had no detectable EBV-DNA in their saliva throughout the study period. EBV-DNA was detected in saliva between 1 and 13 d in the positive swimmers (Table 1) but not necessarily on consecutive days (Figure 1). Subject 01 had consistently high levels of EBV-DNA detected throughout the entire study period. In other swimmers with a high detection rate (nos. 02 and 03), the semiquantitative rating of the viral EBV load indicated an increase in the viral load over several days, followed by a decrease in subsequent days (Fig. 1, no. 02).
FIGURE 1: Preexercise salivary IgA levels over the 1-month training period for subject 02 are indicated by the solid line (♦). A semiquantitative rating (0–3) of EBV-DNA detected in each saliva sample is represented as a solid bar. First appearance of URS is indicated.
Associations with episodes of URS.
Episodes of URS were reported in 9 of the 14 (64%) swimmers (Table 1). There were no episodes of URS in the three seronegative swimmers during the study period. Although the subject numbers were small, there was a significant positive correlation between EBV seropositivity and the incidence of URS (P = 0.027). Seven of the 11 EBV seropositive swimmers had EBV-DNA detected during the study, but only six of these seven swimmers reported URS.
In the subjects where it was possible to determine the time course of appearance of URS in relation to EBV-DNA expression (Fig. 1, no. 02), the viral EBV-DNA was detected in the saliva samples collected before the appearance of URS (nos. 01, 02, 03, and 04). It was not possible to determine the time course in two subjects (nos. 05 and 06) who already had URS on the first day of the study. However, EBV-DNA was detected before a second episode of URS in one of these subjects (no. 06). The average time between first detection of EBV-DNA and the appearance of URS was 12 d (range: 4–18 d).
Associations with salivary IgA.
Salivary IgA levels increased after the appearance of URS before returning to pre-URS levels (Fig. 2). A repeated-measures ANOVA indicated that changes in salivary IgA levels during the course of URS were significantly different over time (P < 0.0001). The salivary IgA levels immediately before the first appearance of URS (geometric mean (g) = 36.3 mg·L−1, 95% confidence interval (CI): 22.4, 52.5 mg·L−1) were significantly lower (P = 0.01) than the peak levels occurring during the subsequent immune response (g = 112.5 mg·L−1, 95% CI: 70.9, 178.4 mg·L−1). The salivary IgA levels peaked on average 6 d (range: 3–7 d) after the appearance of URS and returned to lower levels on average 11 d (range: 6–19 d) after the first URS (Table 2). There was no significant difference between the salivary IgA levels before the URS and the trough levels (g = 25.0 mg·L−1, 95% CI: 15.0, 41.6 mg·L−1) after the immune response (P = 0.12).
FIGURE 2: Change in salivary IgA after onset of URS, from the baseline level before appearance of URS, to the peak IgA immune response (median = 6 d), and the trough levels post-URS (median = 11 d) for eight episodes in seven subjects. The median response (- - -) is indicated.
DISCUSSION
This study provides the first conclusive evidence of significant latent EBV viral shedding in athletes undertaking intensive training. The data also indicate that symptoms of upper-respiratory illness may be a common occurrence in conditioned athletes during periods of intensive training. The significance of the latent EBV shedding in a high proportion of the swimmers is uncertain. There are two possible explanations for the data. The time course of detection of EBV-DNA in saliva in relation to the URS suggests that latent EBV viral shedding precedes the appearance of URS and may be a contributor to the symptoms. Alternatively, the EBV-DNA may be unrelated to the URS but rather is a reflection of the subclinical immune dysregulation known to be caused by intensive training. The low levels of salivary IgA that preceded the appearance of URS indicated transient mucosal immune suppression in this cohort in association with each episode of upper-respiratory illness.
The high level of EBV seropositivity (79%) in this cohort of swimmers is consistent with previous investigations of elite athletes at the Australian Institute of Sport (79%) (9) and the 80–90% prevalence in general adult population. The finding that there were no reported URS in any of the EBV seronegative swimmers and that 9 of 11 EBV seropositive swimmers had URS episodes resulted in a statistically significant association between prior infection with EBV and URS. However, this should be viewed with caution due to the small numbers of subjects and requires further investigation with larger population size studies.
There are very few prior relevant studies of the effect of stressors on latent viral reactivation and all studies used less-frequent sampling protocols. This limitation may have prevented identification of the full extent of EBV shedding in these prior studies. The proportion of swimmers with evidence of latent EBV viral shedding (11 of 14, 64%) was higher than our previous study of swimmers (3 of 14, 21%) investigated at monthly intervals over a 7-month training season (9). The detection rate on any one occasion for astronauts before spaceflights (18%) (19) and for expeditioners studied during isolation in Antarctica (17%) (16) were of a similar order of magnitude. The Antarctic studies indicated 79–100% of subjects had EBV-DNA detected on at least one occasion (16). EBV-DNA was detected more frequently in saliva from the astronauts before space shuttle missions (18%) than during (9%) or after (6%) the spaceflights (19), and the conclusion was that the increased frequency of shedding was associated with preflight psychological stress. The increased EBV shedding in expeditioners during prolonged Antarctic isolation was shown to be associated with alterations in T-cell function (16) and a polyclonal expansion of the latent EBV infected B-cell population in peripheral blood (27). Another relevant study, using EBV serology, examined military cadets (6) and showed no effect of 6 wk of basic training on reactivation of herpes group viruses but a significant reactivation of latent EBV in response to academic stress. The impact of psychological stressors was not evaluated in our study of elite swimmers and therefore cannot be excluded as a significant cause of the viral reactivation. However, this study was conducted at a time well away from any major competition to avoid any competitive stress.
Given the tight cytotoxic T-cell control required to maintain suppression of the EBV infected B-cells (1), any stressor on the immune system may disturb this equilibrium. Transient suppression of cytotoxic T-cells is known to occur after intensive exercise (5). The viral shedding detected in this cohort of elite swimmers may be the result of diminished T-cell control and expansion of the latent EBV-infected B-cell population. The low levels of salivary IgA before the URS lends support to the concept of transient immunosuppression in athletes undertaking intensive training and is consistent with previous reports of an increased risk of URTI in elite swimmers undertaking intensive training who have low salivary IgA levels (12).
The time sequence of salivary IgA suppression, detection of EBV-DNA, and appearance of URS highlights the potential role of EBV viral reactivation in some episodes of URTI in elite athletes. The increase in salivary IgA after the URS to peak levels approximately 5 d later reflects the normal pattern of mucosal immune response to antigenic stimulation or infection (8). This pattern suggests that despite the transient immune suppression the swimmers were capable of responding to antigenic challenge. A significant finding in this study was the detection of EBV-DNA in saliva before the appearance of the URS. This time sequence suggests that latent EBV viral shedding may be a contributing factor in the symptoms associated with URTI in elite swimmers and the subsequent mucosal immune response. The consistent pattern in this study was (i) low levels of salivary IgA, (ii) detection of EBV-DNA in saliva, (iii) appearance of URS consistent with viral URTI, and (iv) a mucosal immune response of elevated salivary IgA with (v) resolution to lower levels within 6–19 d.
Virus-specific secretory IgA (SIgA) antibodies were not measured in this study but are usually absent from saliva of most asymptomatic shedders of the virus (30). However, SIgA plays a role in the process of EBV transport into mucosal epithelium via secretory component in the chronic active state (26), and the fluctuations of SIgA observed in the swimmers may be relevant to this process. Persistent infection with EBV causes major distortions within the T-cell receptor memory repertoire (25) and abnormal T-cell expansion in the chronic active state (18). Treatment of patients with infectious mononucleosis (28) and of chronic virus carriers (31) with the antiviral therapy, acyclovir, suppressed oropharyngeal EBV shedding, but did not affect the frequency of detection of circulating latent EBV-infected B-lymphocytes. Specific antiviral therapy may prove a suitable treatment option for athletes with evidence of persistent EBV-DNA and recurrent URS that could potentially interfere with training and competition performance.
Given the limited sample size in this study, interpretation of the data can only be speculative, but the data are strongly suggestive of EBV as a mechanism linking intensive exercise with URS. One hypothesis that explains the observed sequence of biological events is that alteration in T-cell functions as a result of intensive training affects the underlying mucosal immune control mechanisms. The altered T-cell function after exercise may cause suppression of the EBV-specific CTLs resulting in expansion of the latent EBV-infected B-cells and viral shedding into the oral cavity. The same T-cell suppression may also lower mucosal antibody production and reduce the levels of IgA in saliva. Regardless of whether the latent EBV viral shedding contributes to the URS, the subsequent mucosal immune response of increased IgA could assist with neutralization of the viral DNA (26).
Alternatively, the appearance of EBV-DNA may reflect alterations in CTL control of the EBV infected memory B-lymphocytes and be important in the immunosurveillance process for long-term suppression of EBV lymphoblastoid cell lines (25) and the prevention of EBV-associated malignancies (22). The random appearance of EBV-DNA in saliva may be an epiphenomenon of little clinical significance or relevance to otherwise healthy athletes. The significance of the EBV-DNA in saliva to elite athletes remains speculative at this stage but warrants further investigation of its role in URS, not only in swimmers but other populations of athletes undertaking high-intensity endurance training. Persistent or recurring expression of latent EBV viral shedding may also have wider implications for understanding the causes of chronic fatiguing illnesses or recurrent URTI of unknown etiology in nonathletic populations. The potential for therapeutic or nutritional interventions to control viral reactivation or prevent mucosal immune suppression also warrant investigation.
The participation of the AIS swimmers in this study and the support of their coaches are gratefully appreciated. The authors also wish to acknowledge the assistance of Ms. Gail Cox (AIS) with collection of the saliva samples, Ms. Deborah Capper (HIU) for analysis of salivary IgA, Dr. Terry Grissel (University of Newcastle) for assistance with development of the EBV-PCR, and Ms. Jennifer Hutchings (HIU) for preparation of this manuscript. The Virology Unit, Hunter Area Pathology Service, John Hunter Hospital, New Lambton, Australia, performed the EBV serology tests. This study was funded by matched grants from the Australian Institute of Sport and the University of Newcastle (Australia).
Address for correspondence: Dr. Maree Gleeson, Hunter Area Pathology Service, Immunology, Locked Bag 1, Hunter Region Mail Centre NSW 2310, Australia; E-mail: mgleeson@ mail.newcastle.edu.au.
REFERENCES
1. Crutchley, A. T., D. M. Williams, G. Niedobitek, and L. S. Young. Epstein-Barr virus: biology and disease. Oral Dis. 2: S156–S163, 1997.
2. Fricker, P. A. Infectious problems in athletes: an overview. In: Medical Problems in Athletes, K. B. Fields and P. A. Fricker (Eds.). Malden, MA: Blackwell Science, 1997, pp. 3–5.
3. Fricker, P. A., W. A. McDonald, M. Gleeson, and R. L. Clancy. Exercise-associated hypogammaglobulinemia. Clin. J. Sport Med. 9: 46–49, 1999.
4. Fry, R. W., A. R. Morton, G. P. M. Crawford, and D. Keast. Cell numbers and in vitro responses of leucocytes and lymphocyte subpopulations following maximal exercise and interval training sessions of different intensities. Eur. J. Appl. Physiol. 64: 218–227, 1992.
5. Gabriel, H., and W. Kindermann. The acute immune response to exercise: what does it mean? Int. J. Sports Med. 18: S28–S45, 1997.
6. Glaser, R., S. B. Friedman, J. Smyth, et al. The differential impact of training stress and final examination stress on herpesvirus latency at the United States Military Academy at West Point. Brain Behav. Immunity 13: 240–251, 1999.
7. Gleeson, M., and D. B. Pyne. Exercise effects on mucosal immunity. Immunol. Cell Biol. 78: 536–544, 2000.
8. Gleeson, M., A. J. Dobson, D. W. Firman, et al. The variability of immunoglobulins and albumin in salivary secretions of children. Scand. J. Immunol. 33: 533–541, 1991.
9. Gleeson, M., C. Koina, R. L. Clancy, et al. Epstein-Barr virus reactivation in elite swimmers. Int. J. Sports Med. 21 (Suppl. 1): 83, 2000.
10. Gleeson, M., E. Ginn, and J. L. Francis. Salivary immunoglobulin monitoring in an elite kayaker. Clin. J. Sport Med. 10: 206–208, 2000.
11. Gleeson, M., J. L. Francis, D. J. Lugg, et al. A year in Antarctica: mucosal immunity at three Australian stations. Immunol. Cell Biol. 78: 616–622, 2000.
12. Gleeson, M., W. A. McDonald, D. P. Pyne, et al. Salivary IgA levels and infection risk in elite swimmers. Med. Sci. Sports Exerc. 31: 67–73, 1999.
13. MacKinnon, L. T. Exercise and resistance to infectious illness. In: Advances in Exercise Immunology, L. T. MacKinnon (Ed.). Champaign, IL: Human Kinetics, 1999, pp. 1–26.
14. Matter, L., M. Gorgievski, A. Montandon, D. Germman, and A. Telenti. Epstein-Barr virus (EBV) antibody patterns and circulating EBV in immunocompromised hosts. Biotest Bull. 5: 24–32, 1993.
15. McDonald, W. A. Upper respiratory tract infections. In: Medical Problems in Athletes, K. B. Fields and P. A. Fricker (Eds.). Malden, MA: Blackwell Science, 1997, pp. 6–10.
16. Mehta, S. K., D. L. Pierson, H. Cooley, R. Dubow, and D. Lugg. Epstein-Barr virus reactivation associated with diminished cell-mediated immunity in antarctic expeditioners. J. Med. Virol. 61: 235–240, 2000.
17. Neiman, D. C. Exercise, infection and immunity. Int. J. Sports Med. 15: S131–S141, 1994.
18. Ohga, S., N. Kimura, H. Takada, et al. Restricted diversification of T-cells in chronic active Epstein-Barr virus infection: potential inclination to T-lymphoproliferative disease. Am. J. Hematol. 61: 26–33, 1999.
19. Payne, D. A., S. K. Mehta, S. K. Tyring, R. P. Stowe, and D. L. Pierson. Incidence of Epstein-Barr virus in astronaut saliva during spaceflight. Aviat. Space Environ. Med. 70: 1211–3, 1999.
20. Peters, E. M. Exercise, immunology and upper respiratory tract infections. Int. J. Sports Med. 18: S69–S77, 1997.
21. Pyne, D., M. Gleeson, W. A. McDonald, R. L. Clancy, C. Perry, and P. A. Fricker. Training strategies to maintain immunocompetence in athletes. Int. J. Sports Med. 21 (Suppl. 1): S51–S60, 2000.
22. Rickinson, A. B., and D. J. Moss. Human cytotoxic T lymphocyte responses to Epstein-Barr virus infection. Ann. Rev. Immunol. 15: 405–431, 1997.
23. Roberts, J. A. Viral illnesses and sports performance. Sports Med. 3: 296–303, 1986.
24. Rozenberg, F., and P. Lebon. Amplification and characterization of herpesvirus DNA in cerebrospinal fluid from patients with acute encephalitis. J. Clin. Microbiol. 29: 2412–2417, 1991.
25. Silins, S. L., S. M. Cross, K. G. Krauer, D. J. Moss, C. W. Schmidt, and I. S. Misko. A functional link for major TCR expansions in healthy adults caused by persistent Epstein-Barr virus infection. J. Clin. Invest. 102: 1551–1558, 1998.
26. Sixby, J. W., and Q-Y. Yao. Immunoglobulin A-induced shift of Epstein-Barr virus tissue tropism. Science 255: 1578–1580, 1992.
27. Tingate, T. R., D. J. Lugg, H. K. Muller, R. P. Stowe, and D. L. Pierson. Antarctic isolation: immune and viral studies. Immunol. Cell Biol. 75: 275–283, 1997.
28. Tynell, E., E. Aurelius, A. Brandell, et al. Acyclovir and prednisolone treatment of acute infectious mononucleosis: a multicenter, double-blind, placebo-controlled study. J. Infect. Dis. 174: 324–331, 1996.
29. Winther, B., J. M. Gwaltney, Jr., N. Mygind, R. B. Turner, and J. O. Hendley. Sites of rhinovirus recovery after point inoculation of the upper airway. JAMA 256: 1763–1767, 1986.
30. Yao, Q-Y., M. Rowe, A. J. Morgan, et al. Salivary and serum IgA antibodies to the Epstein-Barr virus glycoprotein gp340: incidence and potential for virus neutralization. Int. J. Cancer 48: 45–50, 1991.
31. Yao, Q-Y., P. Ogan, M. Rowe, M. Wood, and A. B. Rickinson. Epstein-Barr virus-infected B cells persist in the circulation of acyclovir-treated virus carriers. Int. J. Cancer 43: 67–71, 1989.
32. Yoda, K., T. Sata, T. Kurata, and H. Aramaki. Oropharyngotonsillitis associated with nonprimary Epstein-Barr virus infection. Arch. Otolaryngol. 126: 185–193, 2000.
