693
Immunity to and Frequency of Reinfection with Respiratory Syncytial Virus Caroline Breese Hall, Edward E. Walsh, Christine E. Long, and Kenneth C. Schnabel
Departments of Pediatrics and Medicine, University of Rochester School of Medicine and Dentistry. New York
The components of immunity to respiratory syncytial virus (RSV) and its durability are incompletely understood. Humoral antibody has been both indicted as being not protective or contributing to the pathogenesis of disease and vindicated as being protective [1-5]. A reexamination of the role of humoral antibody has become possible with the recent availability of highly purified surface glycoproteins of RSV. The apparent functions of the 70-kDa or F protein as the fusion protein and the 90-kba or G protein as the attachment protein suggest that they possess a key role in the immune response to RSV [6-8]. Children have been shown to develop antibody to both proteins in response to infection, but the role of these antibodies in protecting against infection or disease in humans remains to be defined [5, 9-13]. Furthermore, although young children are known ~to be infected repetitively, even during successive annual outbreaks, the duration of immunity is unknown [14, 15]. We sought, therefore, to determine after natural infection the duration of protection to challenge with a strain of RSV of the same strain group, and to delineate the role of antibodies to the F and G proteins in immunity to reinfection in a group of young, healthy adults.
Methods Subjects. Fifteen young, healthy adults who had acquired natural RSV infection were enrolled.
Received 30 July 1990; revised 9 November 1990. Informed consent was obtained from subjects. Financial support: National Institutes of Health (AI-05049 and 20608). Reprints or correspondence: Dr. Caroline Breese Hall, Box 689, Infectious Diseases, University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave., Rochester, NY 14642. The Journal of Infectious Diseases 1991;163:693-698 © 1991 by The University of Chicago. All rights reserved. 0022-1899/9116304-0002$01.00
Study design. The subjects were challenged intranasally with a safety-tested pool of RSV at increasing intervals of time over a 26month period, starting 2 months after their natural RSV infection. Challenges were administered six times at 2, 4, 8, 14, 20, and 26 months after natural infection. Before the initial challenge the subjects were determined to be in good health by history, physical examination, complete blood cell counts, urinalysis, serum electrolyte and chemistry determinations, pregnancy tests on the women, and pulmonary function testing (spirometry). Before each subsequent challenge the examinations were repeated, and after each challenge the subjects were evaluated daily for 2 weeks by physical examination and history. Nasal washes were obtained before each challenge and daily for the next 2 weeks. A serum sample was obtained at the time of the natural infection, just before each challenge, and 1 month after each challenge. Challenge virus. The virus administered was a safety-tested, live RSV stock obtained from R. Chanock (National Institutes of Health). This was derived from RSV strain A-2, which was multiply passaged in the following tissue cultures: human embryonic kidney (6 passages), calf kidney (l0 passages), and bovine embryonic kidney (4 passages). The stock contained 4.7 10gIO TCIDso/mlRSV. At the time of each challenge the stock virus was titered on HEp-2 cells. Intranasal drops (0.5 ml per nostril) were given for each challenge. Viral isolation. Nasal wash specimens were immediately inoculated onto two cell cultures of each of the followingcell lines: HEp-2, rhesus monkey kidney,MDCK, and human foreskin fibroblasts. Cultures were incubated at 35°C on a roller drum and checked daily for typical cytopathic effect. RSV isolates were confirmed by immunofluorescent testing. Serum antibody determinations. Serum antibodies to the F protein of an RSV group A strain (Long) and to the G proteins of the RSV group A strain (GA ) and of a group B strain (18537) (OB) were determined by ELISA. The F protein from only the group A strain was used because of the extensive cross-reactivity of the F proteins of the two strain groups [11]. The viral glycoproteins were purified by affinity chromatography using monoclonal antibodies [6, 7, 16]. The purity of the F and G proteins had been documented by prior testing with standard sera, which showed that the ELISA detected only antibody to the glycoprotein used, indicating no detectable con-
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
To better understand the duration of immunity against respiratory syncytial virus (RSV) and the role of serum antibodies to the surface glycoproteins, F and G, in susceptibility to reinfection, 15 adults with previous natural RSV infection were challenged with RSV of the same strain group (A) at 2, 4, 8, 14, 20, and 26 months after natural infection. By 2 months about one-half and by 8 months two-thirds of the subjects became reinfected. Each challenge resulted in infection in at least one-fourth of the subjects. Within 26 months 73% had two or more and 47% had three or more infections. The duration of immunity tended to increase after two closely spaced infections. Higher neutralizing, F and GAantibody levels before challenge correlated significantly with protection against infection. However, even in subjects with the highest antibody levels, the risk of reinfection was 25%. Specific nasal 19A antibody titers did not correlate significantly with protection. This suggests that humoral neutralizing, F, and G antibodies correlate with resistance to reinfection, but protection is far from complete and is of short duration.
694
Hall et al.
Results Fifteen adults were identified as having a respiratory illness from natural RSV infection. Five had moderately severe illness characterized by fever (37.9-38.3°C) for 1-4 days, upper respiratory tract symptoms, and productive cough. Six were moderately ill with upper respiratory tract symptoms; including nasal congestion, sore throat and ear pain, and cough lasting ~10 days. Four had mild upper respiratory tract illnesses lasting 3-7 days. These natural infections were
documented by viral isolation, and in 12 viral shedding was sequentially quantitated. In these individuals shedding lasted an average of 4.7 days (range, 1-8), and the mean peak amount of virus shed was 3.5 10gIO TCIDso/ml (range, 1.2-4.7). The mean levels and ranges of ELISA antibody to the F, G A , and G a proteins before and 1 month after the natural infection are shown in table 1. Five of the subjects demonstrated a significant fourfold antibody rise to the F protein, whereas the rest with one exception had a twofold rise. One subject showed a concurrent fourfold rise to G B and one to both G A and Ga. The subsequent challenges with the A strain RSV pool resulted in infection in one-third of the total patient challenges, as defined by a fourfold rise in antibody to the F and/or G A proteins or in neutralizing antibody or by viral isolation. The rate of reinfection was highest after the first challenge, 2 months after the natural infection, but each subsequent challenge produced infection in 25%-30% of the subjects (table 1). Eleven (73 %) were reinfected one or more times during the 26 months after the natural infection, and about 50 % had three or more infections (natural infection plus two or more postchallenge infections) (table 2). The median number of months between infections tended to increase with the number of reinfections (table 2). Of the total infections resulting from challenge, 48 % were asymptomatic and the rest resulted in mild upper respiratory symptoms lasting 1-4 days. The first challenge resulted in the highest proportion of symptomatic infections, with 85 % of the infected subjects developing mild nasal congestion and some a slightly sore throat. Subsequent challenges resulted in ~50 % of the infections being asymptomatic. The first challenge also resulted in the greatest proportion of reinfections associated with viral shedding and with a fourfold antibody rise. In 74% of the reinfections, viral shedding occurred, but the duration and amount of shedding tended to diminish in subsequent challenges (table 1). The average duration of shedding for all infections was 3.4 days (range, 1-7). After the first reinfection the average duration of shed-
Table 1. Proportion of subjects that became infected as detected by viral isolation or antibody rise, associated viral shedding, and mean antibody levels (logz) to F, GA, and GB proteins of respiratory syncytial virus (RSV) before and after natural infection and each RSV challenge. Viral shedding Time of challenge (no. of subjects) Natural infection (15) 2 months (15) 4 months (10) 8 months (14) 14 months (12) 20 months (II) 26 months (10)
% infected
% shedding
Mean duration (days)
100 47 30 29 25 27 30
100 47 0 21 '8 27 30
4.7 4.6 0 4.1 1 2 1
NOTE. Time of challenge is months after natural infection.
Antibody response
Mean peak titer (loglO TCIDso/ml)
Before
After
Before
After
Before
After
3.5 3.4 0 2.5 1.2 1.7 0.7
8.4 10.3 11.4 11.7 11.6 11.7 11.6
10.2 12.0 11.9 11.8 11.7 11.9 11.1
7.6 8.7 9.5 9.6 9.0 9.4 9.6
8.5 9.7 9.9 9.5 9.6 9.7 9.3
8.3 9.2 10.0 10.2 9.9 9.8 9.9
8.7 10.2 10.5 9.9 10.2 10.0 9.9
F
GA
GB
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
tamination by the other surface glycoprotein [12]. The purified proteins were coated to plastic 96-well ELISA plates in bicarbonate buffer. The wells were blocked with PBS with 0.5 % Tween 20, washed with PBS, and incubated for 16 h with serial twofold dilutions of serum in PBS with 0.1% Tween 20 and 5 mM EDTA. After washing with PBS the wells were incubated with alkaline phosphatase-labeled goat anti-human IgG (heavy chain specific) for 3 h, followed by the addition of substrate. Assays were standardized using pools of adult and child sera. Neutralization antibodies were determined by a modification of the microneutralization method of Anderson et al. [17] against strains of both groups of RSV, using the A2 and 18537 strains. Nasal antibody. Secretory IgA antibody to RSV in nasal washes was measured by the indirect immunofl uorescent technique of McIntosh et al. [18] using RSV-infected HEp-2 cells and fluoresceinconjugated antiserum to human IgA. The total IgA concentration in the nasal washes was measured by single radial immunodiffusion with secretory IgA standards [18]. Straingrouping. The wild RSV isolates were determined to belong to groups A or B [19] by indirect immunofluorescent antibody testing on infected HEp-2 cells using a panel of 12 monoclonal antibodies and fluorescein-labeled goat anti-mouse IgG. Three monoclonal antibodies were directed to the F protein of the Long or A2 strains (provided by L. 1. Anderson, Centers for Disease Control), and nine were directed to the G proteins of the Long or 18537 strains. One monoclonal antibody directed to the F protein and four directed to the G protein reacted with both A and B strains; four (three to G and one to F) reacted only with A strains, and three (two to G and one to F) reacted only with B strains.
1ID 1991;163 (April)
JID 1991;163 (April)
Immunity to and Reinfection with RSV
Table 2. Number of reinfections after natural respiratory syncytial virus infection and time in months between infection for each subject. No. of reinfections Natural after natural and first Subject infection reinfection
9 10 II 12 13 14 15
4 4 3 2 2 2 2 I I I I
2 2 2 8 8 2 14 2 8 2
First and second reinfection
Second and third reinfection
Third and fourth reinfection
2 2 2 18 12 6
4 10 16
6 8
0 0 0 0
NOTE. Mean and median values: between natural infection and first reinfection, 5 and 2 months, respectively; first and second reinfection, 8.6 and 4; second and third, 10 and 10; and third and fourth, 7 and 7.
ding was 4.6 days compared with an average of 1.7 days for the subsequent reinfections accompanied by viral shedding (P = .001). Shedding first occurred an average of6 days after challenge (range, 4-11) . No significant correlation existed between the levels of any of the types of serum antibody and the frequency or duration of viral shedding. A similar decline with subsequent challenges was observed in the proportion of subjects with a demonstrable fourfold antibody rise (figure 1) as well as ill the mean rise of the levels of antibodies of all subjects after challenge (table 1). Of the reinfections, 57 % were accompanied by a fourfold or greater rise in neutralizing or ELISA antibody. Most frequently the significant rise was to G A , occurring in 48 % of the reinfections. A heterologous response of neutralizing antibody to the B strain of RSV occurred in 13 % of the total reinfections and to GB in 26 %. This heterologous G B response, however, was accounted for by only three subjects who, with each repetitive infection, produced a significant rise in both G B and G A antibody. The levels of each type of antibody before RSV challenge were examined in relation to whether infection subsequently occurred (table 3). The mean level of each antibody before challenge was lower in those who subsequently became infected than in those who did not become infected. A significant correlation between the prechallenge level of antibody and resistance to infection was present for antibody to the F and G A proteins and for neutralizing antibody to the A2 strain. The effects of the levels of the F .and G A antibodies in subsequent susceptibility to infection were independent of each other but combined were greater than either alone (by logistic regression for F antibody level alone, coefficient = - .52,
P = .003; for G A antibody level alone, coefficient = -.51, P = .002; and for combined F and G A antibody levels, coefficient for F = -.34 and for G = -.39, P = .002). If the infected subjects were divided according to symptomatic or asymptomatic infection, the mean levels of antibody before challenge for each of the antibodies was higher in those who had asymptomatic illness. A significant correlation existed for G A antibody and neutralizing antibody to the B strain with protection against symptomatic illness, and the effect of the G A antibody level was independent of the levels of the other antibodies. If only subjects who demonstrated viral shedding were considered infected, these correlations of prechallenge antibody levels with protection against infection and against symptomatic illness remained significant, except for the correlation with neutralizing antibody to the heterologous B strain. The risk of becoming infected according to the level of F and G A antibody before RSV challenge is shown in figure 2. If all subjects at each challenge were divided according to whether their antibody level to the F protein was ~9.64 log, or above that level, the risk of infection was 75 % versus 28 % (P < .01). Similarly, for subjects with prechallenge levels of G A antibody ~7.64 Iog-, the risk of infection was 55 % versus 26.5 % for those with G A antibody above that level (P < .02). However, even subjects with the highest levels of F and G A antibody before challenge (~12.64 log, and ~10.64 log-, respectively) still had a 25 % chance of becoming infected. IgA antibody to RSV was detectable in the nasal secretions of the subjects before challenge in two-thirds of the total patient challenges. Subjects possessing detectable nasal antibody before challenge tended to become infected less often. In 54 % of the patient challenges that resulted in infection, the subjects had nasal antibody present before challenge compared with 73 % of those who remained uninfected from the challenge (P = .11). The titer of nasal antibody present did not correlate with the risk of subsequent infection. The mean nasal antibody titer before challenge adjusted to 5 mgldl IgA was 1.91og z (range, 0.3-5.8) for those who subsequently became infected, which is not significantly different from the mean titer (2.11og z; range, 0.3-5.4) of those who did not become infected.
Discussion Reinfection with RSV during the first few years of life has been recognized to occur frequently [14, 15]. In day care settings and in the home, young children may acquire RSV infection yearly during their first several years oflife. Reinfection in older age groups has also been documented [20], but the frequency and potential ease of repetitive infection suggested by this study has not been recognized. One possible explanation for the ease of reinfection after natural RSV infection is an antigenic difference between the wild and challenge strains. Only recently has significant strain variation among RSV isolates been confirmed. Two distinct
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
1 2 3 4 5 6 7 8
Months between
695
696
Hall et al.
1ID 1991;163 (April)
50 % 0 F
S
Figure 1. Proportion of subjects
40 30
U B
J E
10
ANTIBODY RISE NEUT. A RISE GA RISE F RISE
0 2
8
4
14
20
26
MONTHS TIME OF CHALLENG E
Table 3. Comparison of mean antibody levels before respiratory syncytial virus (RSV) challenge in subjects who did and did not subsequently become infected and in those who had symptomatic and asymptomatic infection. Mean prechallenge antibody level (Iogz) in subjects subsequently infected who wer e Antibody
F GA GB Neut A Neut B
Infected
P
Not infected
Symptomatic
P
Asymptomatic
10.8 8.3 9.33 11.27 9.18
.005 .002 .07 .03 .11
11.9 9.73 10.26 11.9 9.72
10.55 7.64 8.76 10.80 8 .64
.49 .009 .16 . 10 .0 1
11.09 9.18 9 .95 11.77 9.77
NOTE. Neut A and Ncut B. neutralizing antibody to the A and B 'train, o f RSV . P was determined by nonpaired t test.
'tl
80
C1l
13 C1l
:E
F Antibody
GA Antibody
60
CIl
13 C1l
'Ii'
:>
40
CIl
'0 ~
20 0 < 9.6 >
< 10.6 > < 11.6 >
< 7.6 >
< 8.6 >
< 9.6 >
Antibod y Titer (log 2 ) P•
.0 1
.0 1
. 11
.02
.0 1
.0 7
Figure 2. Proportion of subjects subsequently infected with respiratory syncytial virus divided according to whether their antibody levels to F and GA proteins before respiratory syncytial virus challenge were above (1Zl) or below (_) titer designated on horizontal axis. P was determined by X2 analysis.
strain groups of RSV have been identified by the use of monoclonal antibodies [19, 21-23]. These two groups, A and B, differ primarily in the G surface glycoprotein. The F surface glycoprotein, however, is relatively well conserved, and antibody to the F protein is cross-reactive between the two groups [11, 19,21,22,24]. In our subjects the natural infections were all caused by strains belonging to group A. Thus, since the natural and experimentally induced infections were by strains of the same group, lack of cross-reactive antibodies would be unlikely to account for the susceptibility to reinfection. Although some antigenic differences exist among strains within a group, they seem comparatively minor [19, 21]. Antibodies to these two surface glycoproteins, F and G, have been suggested as playing a major role in immunity to RSV. Neutralizing epitopes are present on both F and G, and
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
C T S
20
who developed a significant (fourfold or greater) rise in level of one or more antibodies accordingto time of respiratory syncytial virus challenge (in months since natural infection) and frequency of significant response of specific antibodies (neutralizing antibodies to the homologous A strain [Neut. A rise] and ELISA antibodies to the G A [GA rise] and F proteins [F rise]) according to number of challenges.
JID 1991;163 (April)
Immunity to and Reinfection with RSV
of maternal antibody present at the time of the initial RSV infection. Passive antibody can affect not only the quantity but also the quality of the humoral antibody response [40, 41]. Secretory antibody to RSV has been shown to be of all immunoglobulin classes and to be directed to both the F and G proteins [12, 42]. However, its role in protecting against RSV infection has not been well delineated. Mills et al. [43] showed that neutralizing activity in the nasalsecretions of volunteers related to protection against RSV challenge. However, McIntosh et al. [18] observed that in infants such neutralizing activity was not mediated by specific antibody and did not relate to the clinical course or viral shedding, whereas specific IgA antibody correlated with the cessation of viral shedding. But infants' secretions may simultaneously contain high titers of virus and antibody and still remain infectious [42]. In our study, the presence, more than the titer, of secretory antibody showed some correlation with resistance to reinfection. However, volunteers possessing no measurable specific IgA antibody, including one with secretory IgA deficiency, resisted challenge while others with high levels of specific secretory antibody became infected. Our findings suggest that subunit vaccines composed of the F and G proteins may offer protection against illness and infection. Infants respond to both of these surface glycoproteins, although the response tends to be less consistent in younger infants, to the G protein, and in the presence of maternal or preexisting antibody [5, 9, 12, 13, 40]. This study also suggests that the expectations for a vaccine must be limited and pragmatic. No matter how efficient a vaccine is in eliciting humoral and secretory antibody responses, protection against infection is unlikely to be complete or durable. Nevertheless, protection against serious infection during an infant's first one to two RSV seasons is a potentially reachable and realistic goal with the currently available knowledge and techniques. Whether a vaccine will have to contain the F as well as the GA and GB proteins remains to be determined, but it would have to be given in the first weeks or months of life, when the potentially inhibiting maternal antibody is present, and as suggested by this study, it may require two or more closely spaced immunizations. References 1. Chanock RM, Kapikian AZ, Mills J, Kim HW, Parrott RH. Influence of immunological factors in respiratory syncytial virus disease of the lower respiratory tract. Arch Environ Health 1970;21:347-55. 2. Glezen WP, Paredes A, Allison JE, Taber LH, Frank AL. Risk of respiratory syncytial virus infection for infants from low-income families in relationship to age, sex, ethnic group, and maternal antibody. J Pediatr 1981;98:708-15. 3. Parrott RH, Kim HW, Arobbio JO, et at. Epidemiology of respiratory syncytial virus infection in Washington, DC. II. Infection and disease with respect to age, immunologic status, race and sex. Am J Epidemiol 1973;98:289-300. 4. Oglivie MM, Vathenen AS, Radford M, Codd J, Key S. Maternal antibody and respiratory syncytial virus infection in infancy.J Med Virol 1981;7:263-71.
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
a fusion epitope is also present on F [24-26]. Passive transfer of monoclonal antibodies to both the F and G proteins protects rodents against RSV challenge [27, 28], as does immunization with these glycoproteins [11, 26, 29-31]. Antibody induced by the F protein has provided protection against challenge with both prototype A and B strains, whereas immunization with the G protein has resulted in little protection against the heterologous strain [11]. Immunization of baboons with purified F protein has similarly induced specific antibody and protection against replication of RSV in the lower respiratory tract after subsequent challenge [32]. However, in chimpanzees recombinant vaccinia viruses expressing the F and G proteins have seemed poorly immunogenic and not protective [33]. In humans little information currently exists concerning the role of antibody to these two surface glycoproteins in protection against infection or illness. Ward et al. [5] showed that the mean level of antibody to the F protein was higher, but not significantly so, in mothers of infants who did not become infected with RSV during their first year of life compared with mothers whose children acquired RSV infection in their first year. Little difference existed between the two groups in the mean level of antibody to the G protein. Mufson et al. [34], in a study of 13 children with two RSV infections 1 year apart, suggested that the children whose initial infection was by A strain viruses were less likely to acquire a second infection from an A strain virus than would be expected epidemiologically. In our study the levels of ELISA antibody against the F and homologous G proteins and neutralizing antibody against the homologous strain correlated with protection against reinfection. Antibody to the G protein also correlated with protection against symptomatic infection. Neutralizing antibodies to both the F and G proteins have'been shown to be protective in vitro and in animals [26, 27, 35] and have been noted in children to correlate inversely with the severity of illness [36]. It has been further suggested that the relative lack of serum neutralizing and fusion.-inhibiting antibodies produced by the formalin-inactivated vaccine played a role in the subsequent enhanced disease observed in vaccinees in the 1960s [37, 38]. Immunization of cotton rats with this inactivated vaccine similarly produced, after RSV challenge, an enhanced pathology in the lung associated with diminished levels of neutralizing and fusion-inhibiting antibodies [39]. A heterologous response to the GB protein occurred in three subjects repetitively, possible suggestingthat an immunologic memory of previous experience with B strains was elicited by the A strain infection. An infant's earliest experiences with RSV may program the type and focus of the immune response to subsequent RSV infections, which could partly explain the variable rate of infection and the degree of illness from a single strain among a group of normal individuals. Modulation of the infant's initial immune response to RSV may be partly dependent upon the type and amount
697
Hall et al.
698
c.«
26. Walsh EE, Hall CB, Briselli M, Brandriss MW, Schlesinger J1. Immunization with glycoprotein subunits of respiratory syncytial virus to protect cotton rats against viral infection. J Infect Dis 1987;155: 1198-1203. 27. Walsh EE, Schlesinger JJ, Brandriss MW. Protection from respiratory syncytial virus infection in cotton rats by passive transfer of monoclonal antibodies. Infect Immun 1984;43 :756-8. 28. Taylor G, Stott EJ, Bew M, et al. Monoclonal antibodies against respiratory syncytial virus infection in mice. Immunology 1984;52:137-42. 29. Elango N, Prince GA, Murphy BR, Venkatesan S, Chanock Moss B. Resistance to human respiratory syncytial virus (RSV) infection induced by immunization of cotton rats with a recombinant vaccinia virus expressing the RSV G glycoprotein. Proc Nat! Acad Sci USA 1986;83:1906-10. 30. Olmsted RA, Elango N, Prince GA, et al. Expression of the F glycoprotein of respiratory syncytial virus by a recombinant vaccinia virus: comparison of the individual contributions of the F and G glycoproteins to host immunity. Proc Nat! Acad Sci USA 1986;83: 7462-6. 31. Stott EJ, Ball LA, Young KK, Furze J, Wertz GW. Human respiratory syncytial virus glycoprotein G expressed from a recombinant vaccinia virus vector protects mice against live-virus challenge. J Virol 1986;60:607-13. 32. Hildreth SW, Baggs RB, Eichberg JW, Johnson C, Arumugham RG, Paradiso PRo A parenterally administered subunit RSV vaccine: safety studies in animals and adult humans [abstract 1066]. Pediatr Res 1989;25:180A. 33. Collins PL, Purcell RH, London LA, et al. Evaluation in chimpanzees of vaccinia virus recombinants that express the surface glycoproteins of human respiratory syncytial virus. Vaccine 1990;8:164-8. 34. Mufson MA, Belshe RB, Orvell C, Norrby E. Subgroup characteristics of respiratory syncytial virus strains recovered from children with two consecutive infections. J Clin Microbiol 1987;25:1535-9. 35. Cote PJ Jr, Fernie BF, Ford EC, Shih JWK, Gerin JL. Monoclonal antibodies to respiratory syncytial virus: detection of virus neutralization and other antigen antibody systems using infected human and murine cells. J Virol Methods 1981;3:137-47. 36. Fernald GW, Almond JR, Henderson FW. Cellular and humoral immunity in recurrent respiratory syncytial virus infections. Pediatr Res 1983; 17:753-8. 37. Murphy BR, Prince GA, Walsh EE, et al. Dissociation between serum neutralizing and glycoprotein antibody responses of infants and children who received inactivated respiratory syncytial virus vaccine. J Clin Microbiol 1986;24:197-202. 38. Murphy BR, Walsh EE. Formalin-inactivated respiratory syncytial virus vaccine induces antibodies to the fusion glycoprotein that are deficient in fusion-inhibiting activity. J Clin MicrobioI1988;26:1595-7. 39. Prince GA, Jenson AB, Hemming VG, et al. Enhancement of respiratory syncytial virus pulmonary pathology in cotton rats against viral infection. J Virol 1986;57:721. 40. Murphy BR, Alling DW, Snyder MH, et al. Effect of age and preexisting antibody on serum antibody response of infants and children to the F and G glycoproteins during respiratory syncytial viral infection. J Clin Microbiol 1986;24:894-8. 41. Murphy BR, Olmsted RA, Collins PL, et al. Passive transfer ofrespiratory syncytial virus (RSV) antiserum suppresses the immune response to the RSV fusion (F) and large (G) glycoproteins expressed by recombinant vaccinia viruses. J Virol 1988;62:3907-10. 42. Kaul TW, Welliver RC, Wong DT, Udwadia RA, Riddlesberger K, Ogra PL. Secretory antibody response to respiratory syncytial virus infection. Am J Dis Child 1981;135:1013-6. 43. Mills VJ, VanKirk JE, Wright PF, Chanock RM. Experimental respiratory syncytial virus infection of adults. J ImmunoI1971;107:123-30.
RM:
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
5. Ward KA, Lambden PR, Oglivie MM, Watt P1. Antibodies to respiratory syncytial virus polypeptides and their significance in human infection. J Gen Virol 1983;64:1867-76. 6. Walsh EE, Schlesinger J1. Brandriss MW. Purification and characterization of GP90, one of the envelope glycoproteins of respiratory syncytial virus. Virology 1984;65:1-7. 7. Walsh EE, Brandriss MW, Schlesinger JJ. Purification and characterization of the respiratory syncytial virus fusion protein. J Gen Virol 1985;66:409-15. 8. Levine S, Klaiber-Franco R, Paradiso PRo Demonstration that glycoprotein G is the attachment protein of respiratory syncytial virus. J Gen Virol 1987;68:2521-4. 9. Hendry RM, Burns JC, Walsh EE, et al. Strain-specific serum antibody responses in infants undergoing primary infection with respiratory syncytial virus. J Infect Dis 1988;157:640-7. 10. Wagner DK, Graham BS, Wright PF, et al. Serum immunoglobulin G antibody subclass responses to respiratory syncytial virus F and G glycoproteins after primary infection. J Clin MicrobioI1986;24:304-6. 11. Johnson PR, Olmsted RA, Prince GA, et al. Antigenic relatedness between glycoproteins of human respiratory syncytial virus subgroups A and B: evaluation of the contributions ofF/G glycoproteins to immunity. J Virol 1987;61:3163-6. 12. Murphy BR, Graham BS, Prince GA, et al. Serum and nasal-wash immunoglobulin G and A antibody response of infants and children to respiratory syncytial virus F and G glycoproteins following primary infection. J Clin Microbiol 1986;23:1009-14. 13. Welliver RC, Sun M, Hildreth SW, Arumugham R, Ogra PL. Respiratory syncytial virus-specific antibody responses in immunoglobulin A and E isolates to the F and G proteins and to intact virus after natural infection. J Clin Microbiol 1989;27:295-9. 14. Henderson FW, Collier AM, Clyde WA, Danny FW. Respiratory syncytial virus infections, reinfections and immunity. N Engl J Med 1979; 300:530-4. 15. Glezen WP, Taber LH, Frank AL, Kasel JA. Risk of primary infection and reinfection with respiratory syncytial virus. Am J Dis Child 1986; 140:543-6. 16. Walsh EE, Cote PJ, Fernie BF, Schlesinger JJ, Brandriss MW. Analysis of the respiratory syncytial virus fusion protein using monoclonal and polyclonal antibodies. J Gen Virol 1986;67:505-13. 17. Anderson U, Hierholzer JC, Bingham PG, Stone YO. Microneutralization test for respiratory syncytial virus based on an enzyme immunoassay. J Clin Microbiol 1985;22:1050-2. 18. McIntosh K, Masters HB, Orr I, Chao RK, Barkin RM. The immunologic response to infection with respiratory syncytial virus in infants. J Infect Dis 1978;138:24-32. 19. Anderson U, Hierholzer JC, Tsou al. Antigenic characterization of respiratory syncytial virus strains with monoclonal antibodies. J Infect Dis 1985;151:626-33. 20. Hall CB, Geiman JM, Biggar R, Kotok 01, Hogan PM, Douglas RG Jr. Respiratory syncytial virus infections within families. N Engl J Med 1976;294:414-9. 21. Mufson MA, Orvell C, Rafnar B, Norrby E. Two distinct subtypes of human respiratory syncytial virus. J Gen Virol 1985;66:2111-24. 22. Gimenez HB, Hardman N, Keir HM, Cash P. Antigenic variation between human respiratory syncytial virus isolates. J Gen Virol 1986; 67:863-70. 23. Hendry RM, Talis AL, Godfrey E, Anderson U, Fernie BF, McIntosh K. Concurrent circulation of antigenically distinct strains of RSV during community outbreaks. J Infect Dis 1986;153:291-7. 24. Walsh EE, Brandriss MW, Schlesinger JJ. Immunological differences between the envelope glycoproteins of two strains of human respiratory syncytial virus. J Gen Virol 1987;68:2169-76. 25. Walsh EE, Hruska 1. Monoclonal antibodies, to respiratory syncytial virus proteins: identification of the fusion protein. J ViroI1983;47:171-7.
JID 1991;163 (April)
Immunity to and Frequency of Reinfection with Respiratory Syncytial Virus Caroline Breese Hall, Edward E. Walsh, Christine E. Long, and Kenneth C. Schnabel
Departments of Pediatrics and Medicine, University of Rochester School of Medicine and Dentistry. New York
The components of immunity to respiratory syncytial virus (RSV) and its durability are incompletely understood. Humoral antibody has been both indicted as being not protective or contributing to the pathogenesis of disease and vindicated as being protective [1-5]. A reexamination of the role of humoral antibody has become possible with the recent availability of highly purified surface glycoproteins of RSV. The apparent functions of the 70-kDa or F protein as the fusion protein and the 90-kba or G protein as the attachment protein suggest that they possess a key role in the immune response to RSV [6-8]. Children have been shown to develop antibody to both proteins in response to infection, but the role of these antibodies in protecting against infection or disease in humans remains to be defined [5, 9-13]. Furthermore, although young children are known ~to be infected repetitively, even during successive annual outbreaks, the duration of immunity is unknown [14, 15]. We sought, therefore, to determine after natural infection the duration of protection to challenge with a strain of RSV of the same strain group, and to delineate the role of antibodies to the F and G proteins in immunity to reinfection in a group of young, healthy adults.
Methods Subjects. Fifteen young, healthy adults who had acquired natural RSV infection were enrolled.
Received 30 July 1990; revised 9 November 1990. Informed consent was obtained from subjects. Financial support: National Institutes of Health (AI-05049 and 20608). Reprints or correspondence: Dr. Caroline Breese Hall, Box 689, Infectious Diseases, University of Rochester School of Medicine and Dentistry, 601 Elmwood Ave., Rochester, NY 14642. The Journal of Infectious Diseases 1991;163:693-698 © 1991 by The University of Chicago. All rights reserved. 0022-1899/9116304-0002$01.00
Study design. The subjects were challenged intranasally with a safety-tested pool of RSV at increasing intervals of time over a 26month period, starting 2 months after their natural RSV infection. Challenges were administered six times at 2, 4, 8, 14, 20, and 26 months after natural infection. Before the initial challenge the subjects were determined to be in good health by history, physical examination, complete blood cell counts, urinalysis, serum electrolyte and chemistry determinations, pregnancy tests on the women, and pulmonary function testing (spirometry). Before each subsequent challenge the examinations were repeated, and after each challenge the subjects were evaluated daily for 2 weeks by physical examination and history. Nasal washes were obtained before each challenge and daily for the next 2 weeks. A serum sample was obtained at the time of the natural infection, just before each challenge, and 1 month after each challenge. Challenge virus. The virus administered was a safety-tested, live RSV stock obtained from R. Chanock (National Institutes of Health). This was derived from RSV strain A-2, which was multiply passaged in the following tissue cultures: human embryonic kidney (6 passages), calf kidney (l0 passages), and bovine embryonic kidney (4 passages). The stock contained 4.7 10gIO TCIDso/mlRSV. At the time of each challenge the stock virus was titered on HEp-2 cells. Intranasal drops (0.5 ml per nostril) were given for each challenge. Viral isolation. Nasal wash specimens were immediately inoculated onto two cell cultures of each of the followingcell lines: HEp-2, rhesus monkey kidney,MDCK, and human foreskin fibroblasts. Cultures were incubated at 35°C on a roller drum and checked daily for typical cytopathic effect. RSV isolates were confirmed by immunofluorescent testing. Serum antibody determinations. Serum antibodies to the F protein of an RSV group A strain (Long) and to the G proteins of the RSV group A strain (GA ) and of a group B strain (18537) (OB) were determined by ELISA. The F protein from only the group A strain was used because of the extensive cross-reactivity of the F proteins of the two strain groups [11]. The viral glycoproteins were purified by affinity chromatography using monoclonal antibodies [6, 7, 16]. The purity of the F and G proteins had been documented by prior testing with standard sera, which showed that the ELISA detected only antibody to the glycoprotein used, indicating no detectable con-
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
To better understand the duration of immunity against respiratory syncytial virus (RSV) and the role of serum antibodies to the surface glycoproteins, F and G, in susceptibility to reinfection, 15 adults with previous natural RSV infection were challenged with RSV of the same strain group (A) at 2, 4, 8, 14, 20, and 26 months after natural infection. By 2 months about one-half and by 8 months two-thirds of the subjects became reinfected. Each challenge resulted in infection in at least one-fourth of the subjects. Within 26 months 73% had two or more and 47% had three or more infections. The duration of immunity tended to increase after two closely spaced infections. Higher neutralizing, F and GAantibody levels before challenge correlated significantly with protection against infection. However, even in subjects with the highest antibody levels, the risk of reinfection was 25%. Specific nasal 19A antibody titers did not correlate significantly with protection. This suggests that humoral neutralizing, F, and G antibodies correlate with resistance to reinfection, but protection is far from complete and is of short duration.
694
Hall et al.
Results Fifteen adults were identified as having a respiratory illness from natural RSV infection. Five had moderately severe illness characterized by fever (37.9-38.3°C) for 1-4 days, upper respiratory tract symptoms, and productive cough. Six were moderately ill with upper respiratory tract symptoms; including nasal congestion, sore throat and ear pain, and cough lasting ~10 days. Four had mild upper respiratory tract illnesses lasting 3-7 days. These natural infections were
documented by viral isolation, and in 12 viral shedding was sequentially quantitated. In these individuals shedding lasted an average of 4.7 days (range, 1-8), and the mean peak amount of virus shed was 3.5 10gIO TCIDso/ml (range, 1.2-4.7). The mean levels and ranges of ELISA antibody to the F, G A , and G a proteins before and 1 month after the natural infection are shown in table 1. Five of the subjects demonstrated a significant fourfold antibody rise to the F protein, whereas the rest with one exception had a twofold rise. One subject showed a concurrent fourfold rise to G B and one to both G A and Ga. The subsequent challenges with the A strain RSV pool resulted in infection in one-third of the total patient challenges, as defined by a fourfold rise in antibody to the F and/or G A proteins or in neutralizing antibody or by viral isolation. The rate of reinfection was highest after the first challenge, 2 months after the natural infection, but each subsequent challenge produced infection in 25%-30% of the subjects (table 1). Eleven (73 %) were reinfected one or more times during the 26 months after the natural infection, and about 50 % had three or more infections (natural infection plus two or more postchallenge infections) (table 2). The median number of months between infections tended to increase with the number of reinfections (table 2). Of the total infections resulting from challenge, 48 % were asymptomatic and the rest resulted in mild upper respiratory symptoms lasting 1-4 days. The first challenge resulted in the highest proportion of symptomatic infections, with 85 % of the infected subjects developing mild nasal congestion and some a slightly sore throat. Subsequent challenges resulted in ~50 % of the infections being asymptomatic. The first challenge also resulted in the greatest proportion of reinfections associated with viral shedding and with a fourfold antibody rise. In 74% of the reinfections, viral shedding occurred, but the duration and amount of shedding tended to diminish in subsequent challenges (table 1). The average duration of shedding for all infections was 3.4 days (range, 1-7). After the first reinfection the average duration of shed-
Table 1. Proportion of subjects that became infected as detected by viral isolation or antibody rise, associated viral shedding, and mean antibody levels (logz) to F, GA, and GB proteins of respiratory syncytial virus (RSV) before and after natural infection and each RSV challenge. Viral shedding Time of challenge (no. of subjects) Natural infection (15) 2 months (15) 4 months (10) 8 months (14) 14 months (12) 20 months (II) 26 months (10)
% infected
% shedding
Mean duration (days)
100 47 30 29 25 27 30
100 47 0 21 '8 27 30
4.7 4.6 0 4.1 1 2 1
NOTE. Time of challenge is months after natural infection.
Antibody response
Mean peak titer (loglO TCIDso/ml)
Before
After
Before
After
Before
After
3.5 3.4 0 2.5 1.2 1.7 0.7
8.4 10.3 11.4 11.7 11.6 11.7 11.6
10.2 12.0 11.9 11.8 11.7 11.9 11.1
7.6 8.7 9.5 9.6 9.0 9.4 9.6
8.5 9.7 9.9 9.5 9.6 9.7 9.3
8.3 9.2 10.0 10.2 9.9 9.8 9.9
8.7 10.2 10.5 9.9 10.2 10.0 9.9
F
GA
GB
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
tamination by the other surface glycoprotein [12]. The purified proteins were coated to plastic 96-well ELISA plates in bicarbonate buffer. The wells were blocked with PBS with 0.5 % Tween 20, washed with PBS, and incubated for 16 h with serial twofold dilutions of serum in PBS with 0.1% Tween 20 and 5 mM EDTA. After washing with PBS the wells were incubated with alkaline phosphatase-labeled goat anti-human IgG (heavy chain specific) for 3 h, followed by the addition of substrate. Assays were standardized using pools of adult and child sera. Neutralization antibodies were determined by a modification of the microneutralization method of Anderson et al. [17] against strains of both groups of RSV, using the A2 and 18537 strains. Nasal antibody. Secretory IgA antibody to RSV in nasal washes was measured by the indirect immunofl uorescent technique of McIntosh et al. [18] using RSV-infected HEp-2 cells and fluoresceinconjugated antiserum to human IgA. The total IgA concentration in the nasal washes was measured by single radial immunodiffusion with secretory IgA standards [18]. Straingrouping. The wild RSV isolates were determined to belong to groups A or B [19] by indirect immunofluorescent antibody testing on infected HEp-2 cells using a panel of 12 monoclonal antibodies and fluorescein-labeled goat anti-mouse IgG. Three monoclonal antibodies were directed to the F protein of the Long or A2 strains (provided by L. 1. Anderson, Centers for Disease Control), and nine were directed to the G proteins of the Long or 18537 strains. One monoclonal antibody directed to the F protein and four directed to the G protein reacted with both A and B strains; four (three to G and one to F) reacted only with A strains, and three (two to G and one to F) reacted only with B strains.
1ID 1991;163 (April)
JID 1991;163 (April)
Immunity to and Reinfection with RSV
Table 2. Number of reinfections after natural respiratory syncytial virus infection and time in months between infection for each subject. No. of reinfections Natural after natural and first Subject infection reinfection
9 10 II 12 13 14 15
4 4 3 2 2 2 2 I I I I
2 2 2 8 8 2 14 2 8 2
First and second reinfection
Second and third reinfection
Third and fourth reinfection
2 2 2 18 12 6
4 10 16
6 8
0 0 0 0
NOTE. Mean and median values: between natural infection and first reinfection, 5 and 2 months, respectively; first and second reinfection, 8.6 and 4; second and third, 10 and 10; and third and fourth, 7 and 7.
ding was 4.6 days compared with an average of 1.7 days for the subsequent reinfections accompanied by viral shedding (P = .001). Shedding first occurred an average of6 days after challenge (range, 4-11) . No significant correlation existed between the levels of any of the types of serum antibody and the frequency or duration of viral shedding. A similar decline with subsequent challenges was observed in the proportion of subjects with a demonstrable fourfold antibody rise (figure 1) as well as ill the mean rise of the levels of antibodies of all subjects after challenge (table 1). Of the reinfections, 57 % were accompanied by a fourfold or greater rise in neutralizing or ELISA antibody. Most frequently the significant rise was to G A , occurring in 48 % of the reinfections. A heterologous response of neutralizing antibody to the B strain of RSV occurred in 13 % of the total reinfections and to GB in 26 %. This heterologous G B response, however, was accounted for by only three subjects who, with each repetitive infection, produced a significant rise in both G B and G A antibody. The levels of each type of antibody before RSV challenge were examined in relation to whether infection subsequently occurred (table 3). The mean level of each antibody before challenge was lower in those who subsequently became infected than in those who did not become infected. A significant correlation between the prechallenge level of antibody and resistance to infection was present for antibody to the F and G A proteins and for neutralizing antibody to the A2 strain. The effects of the levels of the F .and G A antibodies in subsequent susceptibility to infection were independent of each other but combined were greater than either alone (by logistic regression for F antibody level alone, coefficient = - .52,
P = .003; for G A antibody level alone, coefficient = -.51, P = .002; and for combined F and G A antibody levels, coefficient for F = -.34 and for G = -.39, P = .002). If the infected subjects were divided according to symptomatic or asymptomatic infection, the mean levels of antibody before challenge for each of the antibodies was higher in those who had asymptomatic illness. A significant correlation existed for G A antibody and neutralizing antibody to the B strain with protection against symptomatic illness, and the effect of the G A antibody level was independent of the levels of the other antibodies. If only subjects who demonstrated viral shedding were considered infected, these correlations of prechallenge antibody levels with protection against infection and against symptomatic illness remained significant, except for the correlation with neutralizing antibody to the heterologous B strain. The risk of becoming infected according to the level of F and G A antibody before RSV challenge is shown in figure 2. If all subjects at each challenge were divided according to whether their antibody level to the F protein was ~9.64 log, or above that level, the risk of infection was 75 % versus 28 % (P < .01). Similarly, for subjects with prechallenge levels of G A antibody ~7.64 Iog-, the risk of infection was 55 % versus 26.5 % for those with G A antibody above that level (P < .02). However, even subjects with the highest levels of F and G A antibody before challenge (~12.64 log, and ~10.64 log-, respectively) still had a 25 % chance of becoming infected. IgA antibody to RSV was detectable in the nasal secretions of the subjects before challenge in two-thirds of the total patient challenges. Subjects possessing detectable nasal antibody before challenge tended to become infected less often. In 54 % of the patient challenges that resulted in infection, the subjects had nasal antibody present before challenge compared with 73 % of those who remained uninfected from the challenge (P = .11). The titer of nasal antibody present did not correlate with the risk of subsequent infection. The mean nasal antibody titer before challenge adjusted to 5 mgldl IgA was 1.91og z (range, 0.3-5.8) for those who subsequently became infected, which is not significantly different from the mean titer (2.11og z; range, 0.3-5.4) of those who did not become infected.
Discussion Reinfection with RSV during the first few years of life has been recognized to occur frequently [14, 15]. In day care settings and in the home, young children may acquire RSV infection yearly during their first several years oflife. Reinfection in older age groups has also been documented [20], but the frequency and potential ease of repetitive infection suggested by this study has not been recognized. One possible explanation for the ease of reinfection after natural RSV infection is an antigenic difference between the wild and challenge strains. Only recently has significant strain variation among RSV isolates been confirmed. Two distinct
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
1 2 3 4 5 6 7 8
Months between
695
696
Hall et al.
1ID 1991;163 (April)
50 % 0 F
S
Figure 1. Proportion of subjects
40 30
U B
J E
10
ANTIBODY RISE NEUT. A RISE GA RISE F RISE
0 2
8
4
14
20
26
MONTHS TIME OF CHALLENG E
Table 3. Comparison of mean antibody levels before respiratory syncytial virus (RSV) challenge in subjects who did and did not subsequently become infected and in those who had symptomatic and asymptomatic infection. Mean prechallenge antibody level (Iogz) in subjects subsequently infected who wer e Antibody
F GA GB Neut A Neut B
Infected
P
Not infected
Symptomatic
P
Asymptomatic
10.8 8.3 9.33 11.27 9.18
.005 .002 .07 .03 .11
11.9 9.73 10.26 11.9 9.72
10.55 7.64 8.76 10.80 8 .64
.49 .009 .16 . 10 .0 1
11.09 9.18 9 .95 11.77 9.77
NOTE. Neut A and Ncut B. neutralizing antibody to the A and B 'train, o f RSV . P was determined by nonpaired t test.
'tl
80
C1l
13 C1l
:E
F Antibody
GA Antibody
60
CIl
13 C1l
'Ii'
:>
40
CIl
'0 ~
20 0 < 9.6 >
< 10.6 > < 11.6 >
< 7.6 >
< 8.6 >
< 9.6 >
Antibod y Titer (log 2 ) P•
.0 1
.0 1
. 11
.02
.0 1
.0 7
Figure 2. Proportion of subjects subsequently infected with respiratory syncytial virus divided according to whether their antibody levels to F and GA proteins before respiratory syncytial virus challenge were above (1Zl) or below (_) titer designated on horizontal axis. P was determined by X2 analysis.
strain groups of RSV have been identified by the use of monoclonal antibodies [19, 21-23]. These two groups, A and B, differ primarily in the G surface glycoprotein. The F surface glycoprotein, however, is relatively well conserved, and antibody to the F protein is cross-reactive between the two groups [11, 19,21,22,24]. In our subjects the natural infections were all caused by strains belonging to group A. Thus, since the natural and experimentally induced infections were by strains of the same group, lack of cross-reactive antibodies would be unlikely to account for the susceptibility to reinfection. Although some antigenic differences exist among strains within a group, they seem comparatively minor [19, 21]. Antibodies to these two surface glycoproteins, F and G, have been suggested as playing a major role in immunity to RSV. Neutralizing epitopes are present on both F and G, and
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
C T S
20
who developed a significant (fourfold or greater) rise in level of one or more antibodies accordingto time of respiratory syncytial virus challenge (in months since natural infection) and frequency of significant response of specific antibodies (neutralizing antibodies to the homologous A strain [Neut. A rise] and ELISA antibodies to the G A [GA rise] and F proteins [F rise]) according to number of challenges.
JID 1991;163 (April)
Immunity to and Reinfection with RSV
of maternal antibody present at the time of the initial RSV infection. Passive antibody can affect not only the quantity but also the quality of the humoral antibody response [40, 41]. Secretory antibody to RSV has been shown to be of all immunoglobulin classes and to be directed to both the F and G proteins [12, 42]. However, its role in protecting against RSV infection has not been well delineated. Mills et al. [43] showed that neutralizing activity in the nasalsecretions of volunteers related to protection against RSV challenge. However, McIntosh et al. [18] observed that in infants such neutralizing activity was not mediated by specific antibody and did not relate to the clinical course or viral shedding, whereas specific IgA antibody correlated with the cessation of viral shedding. But infants' secretions may simultaneously contain high titers of virus and antibody and still remain infectious [42]. In our study, the presence, more than the titer, of secretory antibody showed some correlation with resistance to reinfection. However, volunteers possessing no measurable specific IgA antibody, including one with secretory IgA deficiency, resisted challenge while others with high levels of specific secretory antibody became infected. Our findings suggest that subunit vaccines composed of the F and G proteins may offer protection against illness and infection. Infants respond to both of these surface glycoproteins, although the response tends to be less consistent in younger infants, to the G protein, and in the presence of maternal or preexisting antibody [5, 9, 12, 13, 40]. This study also suggests that the expectations for a vaccine must be limited and pragmatic. No matter how efficient a vaccine is in eliciting humoral and secretory antibody responses, protection against infection is unlikely to be complete or durable. Nevertheless, protection against serious infection during an infant's first one to two RSV seasons is a potentially reachable and realistic goal with the currently available knowledge and techniques. Whether a vaccine will have to contain the F as well as the GA and GB proteins remains to be determined, but it would have to be given in the first weeks or months of life, when the potentially inhibiting maternal antibody is present, and as suggested by this study, it may require two or more closely spaced immunizations. References 1. Chanock RM, Kapikian AZ, Mills J, Kim HW, Parrott RH. Influence of immunological factors in respiratory syncytial virus disease of the lower respiratory tract. Arch Environ Health 1970;21:347-55. 2. Glezen WP, Paredes A, Allison JE, Taber LH, Frank AL. Risk of respiratory syncytial virus infection for infants from low-income families in relationship to age, sex, ethnic group, and maternal antibody. J Pediatr 1981;98:708-15. 3. Parrott RH, Kim HW, Arobbio JO, et at. Epidemiology of respiratory syncytial virus infection in Washington, DC. II. Infection and disease with respect to age, immunologic status, race and sex. Am J Epidemiol 1973;98:289-300. 4. Oglivie MM, Vathenen AS, Radford M, Codd J, Key S. Maternal antibody and respiratory syncytial virus infection in infancy.J Med Virol 1981;7:263-71.
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
a fusion epitope is also present on F [24-26]. Passive transfer of monoclonal antibodies to both the F and G proteins protects rodents against RSV challenge [27, 28], as does immunization with these glycoproteins [11, 26, 29-31]. Antibody induced by the F protein has provided protection against challenge with both prototype A and B strains, whereas immunization with the G protein has resulted in little protection against the heterologous strain [11]. Immunization of baboons with purified F protein has similarly induced specific antibody and protection against replication of RSV in the lower respiratory tract after subsequent challenge [32]. However, in chimpanzees recombinant vaccinia viruses expressing the F and G proteins have seemed poorly immunogenic and not protective [33]. In humans little information currently exists concerning the role of antibody to these two surface glycoproteins in protection against infection or illness. Ward et al. [5] showed that the mean level of antibody to the F protein was higher, but not significantly so, in mothers of infants who did not become infected with RSV during their first year of life compared with mothers whose children acquired RSV infection in their first year. Little difference existed between the two groups in the mean level of antibody to the G protein. Mufson et al. [34], in a study of 13 children with two RSV infections 1 year apart, suggested that the children whose initial infection was by A strain viruses were less likely to acquire a second infection from an A strain virus than would be expected epidemiologically. In our study the levels of ELISA antibody against the F and homologous G proteins and neutralizing antibody against the homologous strain correlated with protection against reinfection. Antibody to the G protein also correlated with protection against symptomatic infection. Neutralizing antibodies to both the F and G proteins have'been shown to be protective in vitro and in animals [26, 27, 35] and have been noted in children to correlate inversely with the severity of illness [36]. It has been further suggested that the relative lack of serum neutralizing and fusion.-inhibiting antibodies produced by the formalin-inactivated vaccine played a role in the subsequent enhanced disease observed in vaccinees in the 1960s [37, 38]. Immunization of cotton rats with this inactivated vaccine similarly produced, after RSV challenge, an enhanced pathology in the lung associated with diminished levels of neutralizing and fusion-inhibiting antibodies [39]. A heterologous response to the GB protein occurred in three subjects repetitively, possible suggestingthat an immunologic memory of previous experience with B strains was elicited by the A strain infection. An infant's earliest experiences with RSV may program the type and focus of the immune response to subsequent RSV infections, which could partly explain the variable rate of infection and the degree of illness from a single strain among a group of normal individuals. Modulation of the infant's initial immune response to RSV may be partly dependent upon the type and amount
697
Hall et al.
698
c.«
26. Walsh EE, Hall CB, Briselli M, Brandriss MW, Schlesinger J1. Immunization with glycoprotein subunits of respiratory syncytial virus to protect cotton rats against viral infection. J Infect Dis 1987;155: 1198-1203. 27. Walsh EE, Schlesinger JJ, Brandriss MW. Protection from respiratory syncytial virus infection in cotton rats by passive transfer of monoclonal antibodies. Infect Immun 1984;43 :756-8. 28. Taylor G, Stott EJ, Bew M, et al. Monoclonal antibodies against respiratory syncytial virus infection in mice. Immunology 1984;52:137-42. 29. Elango N, Prince GA, Murphy BR, Venkatesan S, Chanock Moss B. Resistance to human respiratory syncytial virus (RSV) infection induced by immunization of cotton rats with a recombinant vaccinia virus expressing the RSV G glycoprotein. Proc Nat! Acad Sci USA 1986;83:1906-10. 30. Olmsted RA, Elango N, Prince GA, et al. Expression of the F glycoprotein of respiratory syncytial virus by a recombinant vaccinia virus: comparison of the individual contributions of the F and G glycoproteins to host immunity. Proc Nat! Acad Sci USA 1986;83: 7462-6. 31. Stott EJ, Ball LA, Young KK, Furze J, Wertz GW. Human respiratory syncytial virus glycoprotein G expressed from a recombinant vaccinia virus vector protects mice against live-virus challenge. J Virol 1986;60:607-13. 32. Hildreth SW, Baggs RB, Eichberg JW, Johnson C, Arumugham RG, Paradiso PRo A parenterally administered subunit RSV vaccine: safety studies in animals and adult humans [abstract 1066]. Pediatr Res 1989;25:180A. 33. Collins PL, Purcell RH, London LA, et al. Evaluation in chimpanzees of vaccinia virus recombinants that express the surface glycoproteins of human respiratory syncytial virus. Vaccine 1990;8:164-8. 34. Mufson MA, Belshe RB, Orvell C, Norrby E. Subgroup characteristics of respiratory syncytial virus strains recovered from children with two consecutive infections. J Clin Microbiol 1987;25:1535-9. 35. Cote PJ Jr, Fernie BF, Ford EC, Shih JWK, Gerin JL. Monoclonal antibodies to respiratory syncytial virus: detection of virus neutralization and other antigen antibody systems using infected human and murine cells. J Virol Methods 1981;3:137-47. 36. Fernald GW, Almond JR, Henderson FW. Cellular and humoral immunity in recurrent respiratory syncytial virus infections. Pediatr Res 1983; 17:753-8. 37. Murphy BR, Prince GA, Walsh EE, et al. Dissociation between serum neutralizing and glycoprotein antibody responses of infants and children who received inactivated respiratory syncytial virus vaccine. J Clin Microbiol 1986;24:197-202. 38. Murphy BR, Walsh EE. Formalin-inactivated respiratory syncytial virus vaccine induces antibodies to the fusion glycoprotein that are deficient in fusion-inhibiting activity. J Clin MicrobioI1988;26:1595-7. 39. Prince GA, Jenson AB, Hemming VG, et al. Enhancement of respiratory syncytial virus pulmonary pathology in cotton rats against viral infection. J Virol 1986;57:721. 40. Murphy BR, Alling DW, Snyder MH, et al. Effect of age and preexisting antibody on serum antibody response of infants and children to the F and G glycoproteins during respiratory syncytial viral infection. J Clin Microbiol 1986;24:894-8. 41. Murphy BR, Olmsted RA, Collins PL, et al. Passive transfer ofrespiratory syncytial virus (RSV) antiserum suppresses the immune response to the RSV fusion (F) and large (G) glycoproteins expressed by recombinant vaccinia viruses. J Virol 1988;62:3907-10. 42. Kaul TW, Welliver RC, Wong DT, Udwadia RA, Riddlesberger K, Ogra PL. Secretory antibody response to respiratory syncytial virus infection. Am J Dis Child 1981;135:1013-6. 43. Mills VJ, VanKirk JE, Wright PF, Chanock RM. Experimental respiratory syncytial virus infection of adults. J ImmunoI1971;107:123-30.
RM:
Downloaded from http://jid.oxfordjournals.org/ at University of Otago on January 13, 2015
5. Ward KA, Lambden PR, Oglivie MM, Watt P1. Antibodies to respiratory syncytial virus polypeptides and their significance in human infection. J Gen Virol 1983;64:1867-76. 6. Walsh EE, Schlesinger J1. Brandriss MW. Purification and characterization of GP90, one of the envelope glycoproteins of respiratory syncytial virus. Virology 1984;65:1-7. 7. Walsh EE, Brandriss MW, Schlesinger JJ. Purification and characterization of the respiratory syncytial virus fusion protein. J Gen Virol 1985;66:409-15. 8. Levine S, Klaiber-Franco R, Paradiso PRo Demonstration that glycoprotein G is the attachment protein of respiratory syncytial virus. J Gen Virol 1987;68:2521-4. 9. Hendry RM, Burns JC, Walsh EE, et al. Strain-specific serum antibody responses in infants undergoing primary infection with respiratory syncytial virus. J Infect Dis 1988;157:640-7. 10. Wagner DK, Graham BS, Wright PF, et al. Serum immunoglobulin G antibody subclass responses to respiratory syncytial virus F and G glycoproteins after primary infection. J Clin MicrobioI1986;24:304-6. 11. Johnson PR, Olmsted RA, Prince GA, et al. Antigenic relatedness between glycoproteins of human respiratory syncytial virus subgroups A and B: evaluation of the contributions ofF/G glycoproteins to immunity. J Virol 1987;61:3163-6. 12. Murphy BR, Graham BS, Prince GA, et al. Serum and nasal-wash immunoglobulin G and A antibody response of infants and children to respiratory syncytial virus F and G glycoproteins following primary infection. J Clin Microbiol 1986;23:1009-14. 13. Welliver RC, Sun M, Hildreth SW, Arumugham R, Ogra PL. Respiratory syncytial virus-specific antibody responses in immunoglobulin A and E isolates to the F and G proteins and to intact virus after natural infection. J Clin Microbiol 1989;27:295-9. 14. Henderson FW, Collier AM, Clyde WA, Danny FW. Respiratory syncytial virus infections, reinfections and immunity. N Engl J Med 1979; 300:530-4. 15. Glezen WP, Taber LH, Frank AL, Kasel JA. Risk of primary infection and reinfection with respiratory syncytial virus. Am J Dis Child 1986; 140:543-6. 16. Walsh EE, Cote PJ, Fernie BF, Schlesinger JJ, Brandriss MW. Analysis of the respiratory syncytial virus fusion protein using monoclonal and polyclonal antibodies. J Gen Virol 1986;67:505-13. 17. Anderson U, Hierholzer JC, Bingham PG, Stone YO. Microneutralization test for respiratory syncytial virus based on an enzyme immunoassay. J Clin Microbiol 1985;22:1050-2. 18. McIntosh K, Masters HB, Orr I, Chao RK, Barkin RM. The immunologic response to infection with respiratory syncytial virus in infants. J Infect Dis 1978;138:24-32. 19. Anderson U, Hierholzer JC, Tsou al. Antigenic characterization of respiratory syncytial virus strains with monoclonal antibodies. J Infect Dis 1985;151:626-33. 20. Hall CB, Geiman JM, Biggar R, Kotok 01, Hogan PM, Douglas RG Jr. Respiratory syncytial virus infections within families. N Engl J Med 1976;294:414-9. 21. Mufson MA, Orvell C, Rafnar B, Norrby E. Two distinct subtypes of human respiratory syncytial virus. J Gen Virol 1985;66:2111-24. 22. Gimenez HB, Hardman N, Keir HM, Cash P. Antigenic variation between human respiratory syncytial virus isolates. J Gen Virol 1986; 67:863-70. 23. Hendry RM, Talis AL, Godfrey E, Anderson U, Fernie BF, McIntosh K. Concurrent circulation of antigenically distinct strains of RSV during community outbreaks. J Infect Dis 1986;153:291-7. 24. Walsh EE, Brandriss MW, Schlesinger JJ. Immunological differences between the envelope glycoproteins of two strains of human respiratory syncytial virus. J Gen Virol 1987;68:2169-76. 25. Walsh EE, Hruska 1. Monoclonal antibodies, to respiratory syncytial virus proteins: identification of the fusion protein. J ViroI1983;47:171-7.
JID 1991;163 (April)
