Infection with respiratory syncytial virus (RSV) is the most important respiratory infectious agent in infancy.1,2 3–7
This concept is derived largely from experiences with a formalin-inactivated RSV vaccine preparation approximately 40 years ago. Vaccine recipients developed greater lymphoproliferative responses to RSV antigen and more severe forms of respiratory illness when subsequently infected with RSV than did controls.8 8 9,10
In mouse models, ablation of either CD4 or CD8-positive lymphocytes after RSV infection lessens the severity of illness.11 12 13,14 15 16
Therefore, although increased activity of some component of the inflammatory response to RSV infection may be critical to the development of severe forms of bronchiolitis, the nature of the hypothetical overly responsive mediator remains unknown. In human infants, RSV generally causes a more severe form of LRTI than does influenza virus.17,18
Measurement of Cytokines in NPS
NPS from subjects recruited for cytokine analysis were tested using the Bio-Plex Human Cytokine 17-Plex panel (Bio-Rad Laboratories, Hercules, CA).
IL2, IFN-γ, and IL17.
IL2 was detectable in secretions of only 2 of 36 (5.6%) subjects with RSV infection, but in 8 of 33 (24.2%) subjects with influenza infection (P = 0.04). IFN-γ was detected in 13 of 36 (36.1%) of subjects with RSV infection, and in 32 of 33 (97%) subjects with influenza virus infection (P < 0.0001). IL17 was undetectable in subjects with RSV infection, but was detected in 19 of 33 (57.6%) subjects with influenza infection (P < 0.0001).
IL4, IL5, and IL13.
IL4 was detected in secretions of only 6 of 36 (16.7%) subjects with RSV infection, and 21 of 33 (61.6%) of subjects with influenza infection (P = 0.0001). IL5 and IL13 were found in infrequently (0–24%) in subjects with RSV or influenza infection, with no statistically significant differences between groups (each P ≥ 0.23).
IL12, MCP-1, and IL6.
IL12 (P = 0.04), MCP-1 (P = 0.047) and IL6 (P = 0.015) were detected less frequently in subjects with RSV infection than in those with influenza infection.
IL1β, IL7, and IL10.
IL1b (P < 0.0001), IL7 (P < 0.0001) and IL10 (P = 0.0025) were all less frequently in subjects with RSV infection than in those with influenza infection.
G-CSF, GM-CSF, TNFa, IL8, and MIP-1β.
Each of these mediators was detected with a frequency that did not differ markedly (P ≥ 0.24) between subjects with RSV or influenza virus infection.
Therefore, most cytokines and chemokines were detected more frequently among infants with influenza virus infection and, surprisingly, none were found more commonly in those with RSV infection.
Concentrations of Mediators in Secretions and Type of Virus Infection
We next evaluated the measured quantities of mediators in secretions. Concentrations of mediators in subjects with each type of virus infection followed the same pattern as the frequency of detection of mediators. That is, most cytokines were present in greater quantities in influenza virus infection, particularly T lymphocyte-associated cytokines IL2, IL17 and IFN-γ. In contrast IL4, IL5 and IL13 were only rarely detectable in subjects with either type of virus infection.
Cytokines primarily released by macrophages (IL6, IL12, IL1β, and MCP-1) were in present in greater quantities among individuals with influenza virus infection.
The chemokines G-CSF, GM-CSF, TNFα, IL8, and MIP1β (primarily released from epithelial cells) were present in high concentrations in subjects with either type of virus infection.
Concentrations of Mediators in Secretions Analyzed by Form and Severity of Illness
We next compared the concentrations of mediators in subjects with upper respiratory infection (URI) versus those with bronchiolitis, for both types of viral infection. No mediator was present in significantly higher concentrations among subjects with LRTI than in subjects with URI alone (each P ≥ 0.04, data not shown). This was the case for subjects with RSV infection as well as for those with influenza virus infection.
We next studied the relationship of mediator release to the measured value of oxygen saturation. The concentrations of the mediators studied bore no substantial relationship to the degree of oxygen saturation (each P ≥ 0.14).
Analysis of Autopsy Tissues
Lung tissue samples from fatal RSV (n = 9) and influenza (n = 11) LRTI cases and control tissues from an infants dying of asphyxia with no evidence of viral infection were processed for routine H&E staining and by immunohistochemistry (IHC). Demographic features of infants dying of LRTI are described in Table 2 . These factors were similar to those for subjects described in Table 1 , except that subjects with congenital heart disease and Down syndrome were not excluded. Preexisting conditions present before the onset of LRTI in these infants are listed in Table 2 . Two pathologists who were unaware of the virological diagnosis reviewed all tissues. In all cases, the cause of death was determined to be bronchiolitis. No cases of bacterial infection were identified, and no cases had immunodeficiency as determined by the clinical history and the general autopsy findings.
TABLE 2:
TABLE 1:
Viral Antigen.
RSV antigen was present extensively in the lung, mostly in exfoliated bronchiolar epithelial cells (Fig. 1 , left panel). In contrast, influenza antigen was much more sparse, and was detected primarily in the intact epithelium of larger airways, and only minimally in exfoliated epithelium (Fig. 2 , right panel).
FIGURE 1.:
Immunohistochemical staining for viral antigen in lung tissue from human infants with fatal RSV (left panel) or influenza (right panel) infection. RSV antigen was widely detectable in exfoliated bronchial epithelial cells. Influenza antigen was detected less extensively, and in predominantly intact epithelium.
FIGURE 2.:
Immunohistochemical staining for CD4 (upper panels) and CD8 (lower panels) antigen-positive leukocytes in lung tissue from human infants with fatal RSV (left panels) or influenza (right panels) infection. Few CD4 cells were detectable, with slightly more found in tissues from influenza-infected cases than in specimens from cases of RSV infection. CD8 cells were only rarely identified in tissues from infants infected by either virus.
Inflammatory Cells.
CD4 and CD8 antigen-positive lymphocytes were rarely detected in lung tissues from these fatal cases. There were possibly slightly more CD4 lymphocytes among fatal cases of influenza infection than in RSV cases (Fig. 2 , top), whereas virtually no CD8-positive cells could be identified in subjects with either type of virus infection (Fig. 2 , bottom).
There were also virtually no detectable cells bearing CD56 antigen (natural killer cells) in lung tissues from fatal bronchiolitis cases, whereas cells expressing CD56 were readily observed in controls consisting of lung tissue from an adult with lung cancer. Lung tissues were also stained for the presence of granzyme, a product of cytotoxic lymphocytes. Granzyme positive cells were undetectable in bronchiolitis cases, but were readily detectable in control tissues from the patient with lung cancer (data not shown).
In contrast to the above, cells staining positive for CD16 (macrophages and neutrophils) were abundant in the lungs of fatal bronchiolitis cases. In control tissues from an infant dying from asphyxia, almost no cells expressed CD16.
Markers of Apoptosis.
We next studied the lung for markers of programmed cell death. Caspase3, an intracellular marker of apoptosis, was abundantly present in lung tissues of subjects with infection caused by either type of virus. This was also true for the membrane-associated marker, Fas. However, the cells expressing these markers were strikingly different according to the type of virus causing infection. That is, in cases of RSV infection, apoptotic markers were expressed primarily on epithelial cells. In influenza infection, these markers were expressed almost exclusively by inflammatory cells in the lung and airway.
DISCUSSION
In this study, CD4 and CD8 antigen-positive lymphocytes (and the cytotoxic enzyme, granzyme), were nearly undetectable in autopsy tissues of infants with fatal bronchiolitis. These findings argue strongly that cytotoxic lymphocyte responses do not play any important role in the pathogenesis of bronchiolitis in infancy. In fact, the subjects who had the mildest forms of bronchiolitis (infants surviving influenza virus infection) had the highest amounts of the cytokine IFN-γ, which is secreted by cytotoxic lymphocytes responding to viral infections. In addition, viral antigen was prominent in the airway epithelium in subjects with either RSV or influenza virus infection, but especially in RSV bronchiolitis. Infantile bronchiolitis therefore is characterized not by enhanced cytotoxic lymphocyte responses as has previously been proposed, but rather by an inefficient clearance of virus.
In the absence of the development of adaptive immune responses, innate immune mechanisms are probably responsible for survival from primary infection with these 2 viruses. In this regard, macrophages and neutrophils (but not natural killer cells) were predominant in the lung, and markers of apoptosis were prominently expressed. Recovery from primary RSV and influenza infections therefore presumably depends on the activity of these less efficient modes of viral clearance, which are associated with substantial bystander damage to uninfected cells. This collateral damage may represent the portion of illness, if any, that is related to vigorous immune or inflammatory responses in bronchiolitis. However, it should be noted that cytokine responses were greater among subjects with the milder cases of influenza infection than in those with RSV infection. Cytokines and chemokines released primarily by lymphocytes and by macrophages were present more frequently and in higher concentrations in subjects with influenza virus infection than in those with RSV infection. The pathogenesis of RSV infection is therefore unlikely to be related to enhanced release of inflammatory mediators (at least those studied above). The greater destruction of epithelium in RSV infection may play a more substantial role in disease pathogenesis, but this requires further study.
These findings would seem to have enormous significance to the development of antiviral agents and vaccines against RSV and influenza virus infection. The development of antiviral agents has perhaps been discouraged by the idea that much of the disease caused by RSV infection is immunologically mediated. The autopsy findings, in contrast, show that considerable quantities of virus are present at the height of illness, and antiviral compounds could therefore have a more beneficial role than is presently anticipated.
There is substantial fear of using nonreplicating RSV vaccines in infants, based on the disastrous experience with the intramuscularly-administered formalin-inactivated (FI) RSV vaccine. Infants who received this vaccine developed severe and, in 2 cases, fatal respiratory illness when subsequently infected with RSV naturally.8
Illness induced by FI-RSV was initially believed to be identical clinically to that induced by natural RSV infection. Enhanced illness has been observed following the administration of similar FI-RSV vaccine to animals. Interestingly, enhanced disease following FI-RSV vaccine administration to animals can be abrogated by antibodies to IL4 or IL13.9,10 8 H 2 skewing of immune responses. The mechanism of FI-RSV disease would therefore differ from that of natural bronchiolitis, where eosinophils are not observed in autopsy tissues,13,14 19 H 2 lymphocyte responses, and that the induction of cytotoxic (TH 1) T-lymphocyte responses is desirable. The findings also suggest that it may be feasible to use nonreplicating RSV vaccines in human infants as long as they induce cytotoxic TH 1 responses, and avoid TH 2 skewing.
ACKNOWLEDGMENTS
We are grateful to Roberto P. Garofalo, MD, Karen H. Hintz, BS, Luis Avendano, MD, Katherine Sanchez, MD, Luis Velozo, MD, Hasan S. Jafri, MD, Susana ChĂ¡vez-Bueno, MD, Pearay L. Ogra, MD, and LuAnn McKinney, PhD, without whose invaluable assistance this project could not have been completed.
REFERENCES
1. Leader S, Kohlhase K. Respiratory syncytial virus-coded pediatric hospitalizations, 1997 to 1999.
Pediatr Infect Dis J . 2002;21:629–632.
2. Garenne M, Ronsmans C, Campbell H. The magnitude of mortality from acute respiratory infections in children under 5 years in developing countries.
World Health Stat Q . 1992;45:180–191.
3. Openshaw PJ, Tregoning JS. Immune responses and disease enhancement during respiratory syncytial virus infection.
Clin Microbiol Rev . 2005;18:541–555.
4. Graham BS, Rutigliano JA, Johnson TR. Respiratory syncytial virus immunobiology and pathogenesis.
Virology . 2002;297:1–7.
5. McNamara PS, Smyth R. The pathogenesis of respiratory syncytial virus disease in childhood.
Brit Med Bull . 2002;61:13–28.
6. Varga SM, Braciale TJ. RSV-induced immunopathology: dynamic interplay between the virus and host immune response.
Virology . 2002;295:203–207.
7. Durbin JE, Durbin RK. Respiratory syncytial virus-induced immunoprotection and immunopathology.
Viral Immunol . 2004;17:370–380.
8. Kim HW, Canchola JG, Brandt CD, et al. Respiratory syncytial virus disease in infants despite prior administration of antigenic inactivated vaccine.
Am J Epidemiol . 1969;89:422–434.
9. Connors M, Giese NA, Kulkarni AB, Firestone CY, Morse HC III, Murphy BR. Enhanced pulmonary histopathology induced by respiratory syncytial virus (RSV) challenge of formalin-inactivated RSV-immunized BALB/c mice is abrogated by depletion of interleukin-4 (IL-4) and IL-10.
J Virol . 1994;68:5321–5325.
10. De Swart RL, Kuiken T, Timmerman HH, et al. Immunization of macaques with formalin-inactivated respiratory syncytial virus (RSV) induces interleukin-13-associated hypersensitivity to subsequent RSV infection.
J Virol . 2002;76:11561–11569.
11. Graham BS, Bunton LA, Wright PF, Karzon DT. Role of T lymphocyte subsets in the pathogenesis of primary infection and rechallenge with respiratory syncytial virus in mice.
J Clin Invest . 1991;88:1026–1033.
12. Everard ML, Swarbrick A, Wrightham M, et al. Analysis of cells obtained by bronchial lavage of infants with respiratory syncytial virus infection.
Arch Dis Child . 1994;71:428–432.
13. Neilson KA, Yunis EJ. Demonstration of respiratory syncytial virus in an autopsy series.
Pediatr Pathol . 1990;10:491–502.
14. Wohl ME, Chernick V. State of the art: bronchiolitis.
Am Rev Respir Dis . 1978;118:759–781.
15. Graham BS, Perkins MD, Wright PF, Karzon DT. Primary respiratory syncytial virus infection in mice.
J Med Virol . 1988;26:153–162.
16. DeVincenzo JP, El Saleeby CM, Bush AJ. Respiratory syncytial virus load predicts disease severity in previously healthy infants.
J Infect Dis . 2005;191:1861–1868.
17. Laham FR, Israele V, Casellas JM, et al. Differential production of inflammatory cytokines in primary infection with human metapneumovirus and with other common respiratory viruses of infancy.
J Infect Dis . 2004;189:2047–2056.
18. Wolf DG, Greenberg D, Kalkstein D, et al. Comparison of human metapneumovirus, respiratory syncytial virus and influenza A virus lower respiratory tract infections in hospitalized young children.
Pediatr Infect Dis J . 2006;25:320–324.
19. Garofalo RP, Patti J, Hintz KH, Hill V, Ogra PL, Welliver RC. Macrophage inflammatory-protein-1α (not T helper type 2 cytokines) is associated with severe forms of respiratory syncytial virus bronchiolitis.
J Infect Dis . 2001;184:393–399.
