Complement Is Activated in the Upper Respiratory Tract during Influenza Virus Infection 1- 3
ANN B. BJORNSON, MARTHA A. MELLENCAMP, and GILBERT M. SCHIFF
Introduction
Influenza virus infection produces a local inflammatory response that is characterized by edema and infiltration of inflammatory cells into the airway (1-5). This response occurs in association with desquamation of ciliated epithelial cells induced by replication of the influenza virus. The cells recruited into the airway in response to influenza infection consist of polymorphonuclear leukocytes and mononuclear cells (1, 2, 4, 6, 7). Recruitment of these cells continues throughout the course of infection into the recovery phase. Complement is a central mediator of inflammation and as such may play an important role in the local inflammatory response associated with influenza virus infection. As an essential first step in determining the involvement of complement in this response, we have asked the question: Does complement activation occur in the upper respiratory tract during experimental influenza virus infection in human volunteers? The results of our study document the presence of C3a and C5a in nasal lavage fluids from subjects who develop influenza illness. Methods Subjects Twelve subjects 18 to 40 yr of age were enrolled in this pilot study (nine women and three men with a mean age of 29.3 yr). These subjects were in good health as assessed by physical examination and routine laboratory tests. None of the subjects had a history of significant disease of the respiratory tract, kidney, liver, or heart; excessive alcohol use; hematologic disorders; hypertension; allergies; recent upper respiratory infections; or other diseases or conditions that might put the subject at undue risk or affect the outcome of the study. Pregnancy was ruled out by history and laboratory tests. One subject withdrew from the study prior to administration of the challenge virus. Seven of the remaining subjects were susceptible to the challenge virus as determined by screening serum hemagglutination inhibition (HAl) antibody titers of ~ 1:8 and were assigned to receive 1062
SUMMARY The purpose of this pilot study was to determine whether complement Is activated In the upper respiratory tract during experimental Influenza virus Infection In human volunteers. Seven subjects were challenged with influenza AlBethesda/1/85 (H3N2), and four subjects received placebo. C3a and C5a concentrations were measured by radioimmunoassay In nasal lavage fluids before challenge and for 8 days after challenge. A significant increase (p < 0.05) In C3a and C5a concentrations was demonstrated In lavage fluids from subjects who developed Influenza Illness as compared with unlnfected control subjects and infected subjects who remained asymptomatic. Maximal levels of C3a and C5a were detected during the recovery phase of illness. These results suggest that complement Is activated In the airway In response to influenza Illness. AM REV RESPIR DIS 1991; 143:1062-1066
viral inoculation. The remaining four subjects had screening HAl antibody titers to the challenge virus of 1:32or 1:64and were assigned to the control group. Use of any prescription or nonprescription medication other than contraceptives or vitamins was prohibited during the 48 h prior to the initiation of the study and during the course of the study. Informed consent was obtained from the study subjects, and guidelines for human experimentation of the James N. Gamble Institute of Medical Research were followed in the conduct of the clinical research.
contain 107 .4 tissue culture infectious dose (TCID)so/ml; 250 IIIwere slowly instilled into each nostril of the seven susceptible subjects. The four control subjects received the same volume of uninfected egg allantoic fluid diluted in an identical manner.
Nasal Lavage and Specimen Handling Nasal lavages were performed by instilling 15 to 20 ml of lactated Ringer's solution (Travenol Laboratories, Chicago, IL) warmed to room temperature into the nostrils. The subject expelled the lavage fluid into a plastic receptacle. The volume of lavage fluid recovered ranged from 10 to 15 ml and was recorded. After collection, the lavage fluids were held on ice for no longer than 30 min. The fluids were vigorously mixed for 15 s, and aliquots were removed for viral assays. Aliquots for C3a and C5a determinations werecentrifuged at 1,400 g for 10 min, and the supernatants were removed and supplemented with a final concentration of 40 mM ethylenediamine tetraacetic acid at pH 7.4 (8). These specimens were refrigerated for as long as 16 h and then frozen in small aliquots at -70 0 C until assay. Aliquots for viral assays were processed as described under assessment of infection.
Study Design Subjects were admitted to the isolation facility between 6:00 A.M. and 7:00 A.M. on the first day of the study and remained in the facility for the following 8 days. Virus or placebo was administered on Day 2 between 10:00 A.M. and 1:00 P.M. Vital signs were obtained daily at 8:00 A.M., 12:00noon, 4:00 P.M., 8:00 P.M., and 12:00 midnight. Symptoms were monitored twice daily at 8:00 A.M. and 4:00 P.M. Nasal lavage fluids for viral assays and measurements of C3a and C5a were collected after symptom assessment at 8:00 A.M., 4:00 P.M., and 12:00 midnight. Sera for measurement of HAl antibody response to the challenge virus were collected immediately prior to administration of the challenge virus or placebo and 28 days later.
(Received in original form July 23, 1990 and in revised form December 11, 1990)
Administration of Virus and Placebo Influenza A/Bethesda/1I85 (H3N2) was obtained from Dr. Carole Heilman (National Institutes of Health, Bethesda, MD). The viral pool was produced and safety-tested in accordance with standards set by the National Institutes of Health. An aliquot of challenge pool was diluted in Leibovitz's L-15 medium (Gibco Laboratories, Grand Island, NY) to
1 From the Divisions of Immunology and Clinical Virology, James N. Gamble Institute of Medical Research, Cincinnati, Ohio. 2 Supported by Contract RPOIA from the Procter & Gamble Company, Cincinnati, Ohio. 3 Correspondence and requests for reprints should be addressed to Dr. Ann B. Bjornson, Division of Immunology, James N. Gamble Institute of Medical Research, 2141 Auburn Avenue, Cincinnati, OH 45219.
1063
COMPLEMENT ACTIVATION IN INFLUENZA INFECTION
Assessment of Infection
Measurement of C3a and C5a
Infection was documented by the occurrence of a fourfold or greater serum HAl antibody response to the challenge virus and/or recovery of influenza virus from nasal lavage fluids. For viral assay, 6 ml of lavage fluid were homogenized by repeated passage through an I8-gauge needle, mixed with 1 ml of 7 x veal infusion broth (Difco Laboratories, Detroit, MI) supplemented with 700 units/ml penicillin, 700 ug/ml streptomycin, 17.5 ug/rnl amphotericin B, and 3.5070 bovine serum albumin, divided into aliquots, and frozen at -70 0 C. Within 8 wk of collection, the lavage fluids were assayed for the presence of influenza virus using a minor modification of the method of Knight and coworkers (9). Two hundred microliters of undiluted lavage fluid were inoculated onto Madin-Darby canine kidney (MDCK) cells in 24-well (diameter, 1.5 em) plastic tissue culture plates (Flow Laboratories, McLean, VA). After 1 h, the inoculum was removed by aspiration, and Eagle minimum essential medium (Flow Laboratories) containing 2 ug/ml of trypsin and antibiotics was added. Plates were incubated at 330 C, and hemadsorption with 0.1070 guinea pig erythrocytes was determined after 3 to 4 days of incubation. Negative wellswererinsed, and fresh medium was added for repeat tests after 10 and 14days of incubation. Virus titration of hemadsorption-positive specimens was performed using tenfold dilutions of the lavage fluids in Hanks' balanced salt solution containing 0.5 0J0 gelatin. End points were determined by hemadsorption after 6 to 7 and 13to 14days of incubation as described above. One hemadsorption-positive specimen from each subject was confirmed as influenza A by indirect immunofluorescence (10). Briefly, duplicate tubes of MDCK cells were inoculated with 200 ul of undiluted lavage fluid. After incubation for 3 days at 33 0 C, one tube was tested for the presence of virus by hemadsorption. Infected cellsfrom the other tube were removed, washed twice with sterilephosphate-buffered saline and spotted onto glass microscope slides. Indirect immunofluorescence was performed using a 1:100 dilution of separate pools of mouse monoclonal antibodies to influenza A and B (Centers for Disease Control, Atlanta, OA) and fluorescein isothiocyanate-conjugated sheep antiserum to mouse IgO (Sigma Chemical Co., St. Louis, MO). Serum HAl antibody titers were measured by a standard method using the challenge strain of influenza as the specific antigen (11).
C3a and C5a were measured in duplicate in nasal lavage fluids by radioimmunoassay using kits from Amersham Corp. (Arlington Heights, IL). The method outlined by the manufacturer was followed with the addition of sample blanks. According to the manufacturer, the antiserum to C3a des Arg in the kits was less than IOJo cross-reactive with C5a des Arg and C4a des Arg, and the antiserum to C5a des Arg was less than 0.5 0J0 crossreactive with C3a des Arg and C4a des Arg. The lower limit of detection in the assays was 40 ng/ml for C3a and 4 ng/ml for C5a. Specimens containing undetectable C3a or C5a werereported as 40 ng/ml for C3a and 4 ng/ml for C5a.
Statistical Analysis Significant differences were determined by analysis of variance (13). TCIDsolml was transformed to log., before analysis of variance. Correlations wereassessed by regression analysis (14).
Results
Occurrence of Infection Influenza virus was recovered from nasal lavage fluids in all seven subjects challenged with virus during 4 to 186 h after challenge, whereas none of the four subjects who received the placebo shed virus (table 1). Five of the seven subjects who shed virus had a fourfold or greater rise in serum HAl titers to the challenge virus, whereas HAl titers did not increase in the control subjects (table 1). These results documented that subjects challenged with virus were infected and the control subjects were not infected.
Presence of Complement Cleavage Fragments in Nasal Lavage Fluids
Occurrence of Illness Four of the seven infected subjects (Subjects 6, 8, 11, and 12) developed clinical
C3a and C5a increased in nasal lavage
TABLE 1 VIRAL SHEDDING IN NASAL LAVAGE FLUID AND THE SERUM ANTIBODY RESPONSE TO THE CHALLENGE VIRUS IN THE STUDY SUBJECTS·
Group Challenged with virus
Assessment of Illness The Jackson scoring system (12) was adapted for monitoring symptoms of influenza. At each evaluation, the followingsymptoms were scored by a physician as 0 (absent) or 1 (present): fever (~ 38 0 C), myalgia, headache, chills, rhinitis, pharyngitis, and cough. Scores for each symptom were totaled, with a score of seven being the maximal possible score obtainable.
evidence of influenza illness as reflected by maximal symptom scores of 3 to 7 and fever of 39° C at one or more evaluations during a period of 28 to 116h after challenge. The three remaining infected subjects (Subjects 1, 4, and 10) and the control subjects (Subjects 2, 3, 5, and 7) had maximal symptom scores of ~ 2 and temperatures of < 37° C during the same time period. These subjects were judged to be asymptomatic. Symptom scores were significantly higher in ill subjects than in uninfected control subjects at 44, 68, 76, and 100 h after challenge (p < 0.05) (figure 1). Differences between the symptom scores in ill and asymptomatic infected subjects were significant at 68, 76, 100, and 124h after challenge (p < 0.05). The relationship between viral shedding and illness was also investigated. Viral titers were significantly higher in nasal lavage fluids from ill subjects than from asymptomatic infected subjects at 12, 28 to 44, and 76 h after challenge (p < 0.05) (figure 2). In one of the subjects (Subject 11)who developed illness in response to viral challenge, influenza A was isolated from two separate aliquots of prechallenge nasal lavage fluid (1.4 TCIDso/ml [loglOD. This subject did not have a history of recent upper respiratory infection before the study and therefore was presumed to have a subclinical influenza A infection at the time of challenge. Data from this subject will be presented separately and compared with data from the other ill subjects.
Control
SUbject No.
Maximal TCIDsolml (Iog,o)
Fold Rise in HAl Titer
1 4 6 8 10 11 12
1.4 3.7 5.7 6.2 6.0 6.2 3.4
0 32 2 8 4 16 8
2 3 5 7
NI NI NI NI
0 0 0 0
Definition of abbreviations: HAl = hemagglutination inhibition; NI = not isolated. • Viral titers in lavage fluids from subjects challenged with virus were maximal at 28 to 68 h after challenge. Influenza virus was not isolated from control subjects.
1064
~
BJORNSON, MELLENCAMP, AND SCHIFF
Fig. 1. Symptom scores in ill and asymptomatic subjects infected with influenza virus and in subjects who received placebo. Data are expressed as mean ± SEM. Closed circles indicate infected ill group (Subjects 6,8,11, and 12); open triangles indicate infected asymptomatic group (Subjects 1,4, and 10);open squares indicate control group (Subjects 2, 3, 5, and 7).
5
o o
4
(fJ
E o
a. 3 E
J)2
o...........~.........~-,--.~..-.-""""""""T~~~-,--.~..-.-.....--r~..--, -28
-4
20
44
68
92
116140164188212
Hours After Challenge 7 6
........0 0\
5
-=-
4
o
E
Fig. 2. Viral titers in nasal lavage fluids from ill and asymptomatic subjects infected with influenza virus. Data are expressed as mean ± SEM. Closed circles indicate infected ill group (Subjects 6, 8, 11,and 12);open triangles indicate infected asymptomatic group (Subjects 1, 4, and 10).
'03 o
Lf)
U .....
2
20
44
68
92
116
140
164
188
212
Hours After Challenge
Discussion
1400 1200
........
E
ANN B. BJORNSON, MARTHA A. MELLENCAMP, and GILBERT M. SCHIFF
Introduction
Influenza virus infection produces a local inflammatory response that is characterized by edema and infiltration of inflammatory cells into the airway (1-5). This response occurs in association with desquamation of ciliated epithelial cells induced by replication of the influenza virus. The cells recruited into the airway in response to influenza infection consist of polymorphonuclear leukocytes and mononuclear cells (1, 2, 4, 6, 7). Recruitment of these cells continues throughout the course of infection into the recovery phase. Complement is a central mediator of inflammation and as such may play an important role in the local inflammatory response associated with influenza virus infection. As an essential first step in determining the involvement of complement in this response, we have asked the question: Does complement activation occur in the upper respiratory tract during experimental influenza virus infection in human volunteers? The results of our study document the presence of C3a and C5a in nasal lavage fluids from subjects who develop influenza illness. Methods Subjects Twelve subjects 18 to 40 yr of age were enrolled in this pilot study (nine women and three men with a mean age of 29.3 yr). These subjects were in good health as assessed by physical examination and routine laboratory tests. None of the subjects had a history of significant disease of the respiratory tract, kidney, liver, or heart; excessive alcohol use; hematologic disorders; hypertension; allergies; recent upper respiratory infections; or other diseases or conditions that might put the subject at undue risk or affect the outcome of the study. Pregnancy was ruled out by history and laboratory tests. One subject withdrew from the study prior to administration of the challenge virus. Seven of the remaining subjects were susceptible to the challenge virus as determined by screening serum hemagglutination inhibition (HAl) antibody titers of ~ 1:8 and were assigned to receive 1062
SUMMARY The purpose of this pilot study was to determine whether complement Is activated In the upper respiratory tract during experimental Influenza virus Infection In human volunteers. Seven subjects were challenged with influenza AlBethesda/1/85 (H3N2), and four subjects received placebo. C3a and C5a concentrations were measured by radioimmunoassay In nasal lavage fluids before challenge and for 8 days after challenge. A significant increase (p < 0.05) In C3a and C5a concentrations was demonstrated In lavage fluids from subjects who developed Influenza Illness as compared with unlnfected control subjects and infected subjects who remained asymptomatic. Maximal levels of C3a and C5a were detected during the recovery phase of illness. These results suggest that complement Is activated In the airway In response to influenza Illness. AM REV RESPIR DIS 1991; 143:1062-1066
viral inoculation. The remaining four subjects had screening HAl antibody titers to the challenge virus of 1:32or 1:64and were assigned to the control group. Use of any prescription or nonprescription medication other than contraceptives or vitamins was prohibited during the 48 h prior to the initiation of the study and during the course of the study. Informed consent was obtained from the study subjects, and guidelines for human experimentation of the James N. Gamble Institute of Medical Research were followed in the conduct of the clinical research.
contain 107 .4 tissue culture infectious dose (TCID)so/ml; 250 IIIwere slowly instilled into each nostril of the seven susceptible subjects. The four control subjects received the same volume of uninfected egg allantoic fluid diluted in an identical manner.
Nasal Lavage and Specimen Handling Nasal lavages were performed by instilling 15 to 20 ml of lactated Ringer's solution (Travenol Laboratories, Chicago, IL) warmed to room temperature into the nostrils. The subject expelled the lavage fluid into a plastic receptacle. The volume of lavage fluid recovered ranged from 10 to 15 ml and was recorded. After collection, the lavage fluids were held on ice for no longer than 30 min. The fluids were vigorously mixed for 15 s, and aliquots were removed for viral assays. Aliquots for C3a and C5a determinations werecentrifuged at 1,400 g for 10 min, and the supernatants were removed and supplemented with a final concentration of 40 mM ethylenediamine tetraacetic acid at pH 7.4 (8). These specimens were refrigerated for as long as 16 h and then frozen in small aliquots at -70 0 C until assay. Aliquots for viral assays were processed as described under assessment of infection.
Study Design Subjects were admitted to the isolation facility between 6:00 A.M. and 7:00 A.M. on the first day of the study and remained in the facility for the following 8 days. Virus or placebo was administered on Day 2 between 10:00 A.M. and 1:00 P.M. Vital signs were obtained daily at 8:00 A.M., 12:00noon, 4:00 P.M., 8:00 P.M., and 12:00 midnight. Symptoms were monitored twice daily at 8:00 A.M. and 4:00 P.M. Nasal lavage fluids for viral assays and measurements of C3a and C5a were collected after symptom assessment at 8:00 A.M., 4:00 P.M., and 12:00 midnight. Sera for measurement of HAl antibody response to the challenge virus were collected immediately prior to administration of the challenge virus or placebo and 28 days later.
(Received in original form July 23, 1990 and in revised form December 11, 1990)
Administration of Virus and Placebo Influenza A/Bethesda/1I85 (H3N2) was obtained from Dr. Carole Heilman (National Institutes of Health, Bethesda, MD). The viral pool was produced and safety-tested in accordance with standards set by the National Institutes of Health. An aliquot of challenge pool was diluted in Leibovitz's L-15 medium (Gibco Laboratories, Grand Island, NY) to
1 From the Divisions of Immunology and Clinical Virology, James N. Gamble Institute of Medical Research, Cincinnati, Ohio. 2 Supported by Contract RPOIA from the Procter & Gamble Company, Cincinnati, Ohio. 3 Correspondence and requests for reprints should be addressed to Dr. Ann B. Bjornson, Division of Immunology, James N. Gamble Institute of Medical Research, 2141 Auburn Avenue, Cincinnati, OH 45219.
1063
COMPLEMENT ACTIVATION IN INFLUENZA INFECTION
Assessment of Infection
Measurement of C3a and C5a
Infection was documented by the occurrence of a fourfold or greater serum HAl antibody response to the challenge virus and/or recovery of influenza virus from nasal lavage fluids. For viral assay, 6 ml of lavage fluid were homogenized by repeated passage through an I8-gauge needle, mixed with 1 ml of 7 x veal infusion broth (Difco Laboratories, Detroit, MI) supplemented with 700 units/ml penicillin, 700 ug/ml streptomycin, 17.5 ug/rnl amphotericin B, and 3.5070 bovine serum albumin, divided into aliquots, and frozen at -70 0 C. Within 8 wk of collection, the lavage fluids were assayed for the presence of influenza virus using a minor modification of the method of Knight and coworkers (9). Two hundred microliters of undiluted lavage fluid were inoculated onto Madin-Darby canine kidney (MDCK) cells in 24-well (diameter, 1.5 em) plastic tissue culture plates (Flow Laboratories, McLean, VA). After 1 h, the inoculum was removed by aspiration, and Eagle minimum essential medium (Flow Laboratories) containing 2 ug/ml of trypsin and antibiotics was added. Plates were incubated at 330 C, and hemadsorption with 0.1070 guinea pig erythrocytes was determined after 3 to 4 days of incubation. Negative wellswererinsed, and fresh medium was added for repeat tests after 10 and 14days of incubation. Virus titration of hemadsorption-positive specimens was performed using tenfold dilutions of the lavage fluids in Hanks' balanced salt solution containing 0.5 0J0 gelatin. End points were determined by hemadsorption after 6 to 7 and 13to 14days of incubation as described above. One hemadsorption-positive specimen from each subject was confirmed as influenza A by indirect immunofluorescence (10). Briefly, duplicate tubes of MDCK cells were inoculated with 200 ul of undiluted lavage fluid. After incubation for 3 days at 33 0 C, one tube was tested for the presence of virus by hemadsorption. Infected cellsfrom the other tube were removed, washed twice with sterilephosphate-buffered saline and spotted onto glass microscope slides. Indirect immunofluorescence was performed using a 1:100 dilution of separate pools of mouse monoclonal antibodies to influenza A and B (Centers for Disease Control, Atlanta, OA) and fluorescein isothiocyanate-conjugated sheep antiserum to mouse IgO (Sigma Chemical Co., St. Louis, MO). Serum HAl antibody titers were measured by a standard method using the challenge strain of influenza as the specific antigen (11).
C3a and C5a were measured in duplicate in nasal lavage fluids by radioimmunoassay using kits from Amersham Corp. (Arlington Heights, IL). The method outlined by the manufacturer was followed with the addition of sample blanks. According to the manufacturer, the antiserum to C3a des Arg in the kits was less than IOJo cross-reactive with C5a des Arg and C4a des Arg, and the antiserum to C5a des Arg was less than 0.5 0J0 crossreactive with C3a des Arg and C4a des Arg. The lower limit of detection in the assays was 40 ng/ml for C3a and 4 ng/ml for C5a. Specimens containing undetectable C3a or C5a werereported as 40 ng/ml for C3a and 4 ng/ml for C5a.
Statistical Analysis Significant differences were determined by analysis of variance (13). TCIDsolml was transformed to log., before analysis of variance. Correlations wereassessed by regression analysis (14).
Results
Occurrence of Infection Influenza virus was recovered from nasal lavage fluids in all seven subjects challenged with virus during 4 to 186 h after challenge, whereas none of the four subjects who received the placebo shed virus (table 1). Five of the seven subjects who shed virus had a fourfold or greater rise in serum HAl titers to the challenge virus, whereas HAl titers did not increase in the control subjects (table 1). These results documented that subjects challenged with virus were infected and the control subjects were not infected.
Presence of Complement Cleavage Fragments in Nasal Lavage Fluids
Occurrence of Illness Four of the seven infected subjects (Subjects 6, 8, 11, and 12) developed clinical
C3a and C5a increased in nasal lavage
TABLE 1 VIRAL SHEDDING IN NASAL LAVAGE FLUID AND THE SERUM ANTIBODY RESPONSE TO THE CHALLENGE VIRUS IN THE STUDY SUBJECTS·
Group Challenged with virus
Assessment of Illness The Jackson scoring system (12) was adapted for monitoring symptoms of influenza. At each evaluation, the followingsymptoms were scored by a physician as 0 (absent) or 1 (present): fever (~ 38 0 C), myalgia, headache, chills, rhinitis, pharyngitis, and cough. Scores for each symptom were totaled, with a score of seven being the maximal possible score obtainable.
evidence of influenza illness as reflected by maximal symptom scores of 3 to 7 and fever of 39° C at one or more evaluations during a period of 28 to 116h after challenge. The three remaining infected subjects (Subjects 1, 4, and 10) and the control subjects (Subjects 2, 3, 5, and 7) had maximal symptom scores of ~ 2 and temperatures of < 37° C during the same time period. These subjects were judged to be asymptomatic. Symptom scores were significantly higher in ill subjects than in uninfected control subjects at 44, 68, 76, and 100 h after challenge (p < 0.05) (figure 1). Differences between the symptom scores in ill and asymptomatic infected subjects were significant at 68, 76, 100, and 124h after challenge (p < 0.05). The relationship between viral shedding and illness was also investigated. Viral titers were significantly higher in nasal lavage fluids from ill subjects than from asymptomatic infected subjects at 12, 28 to 44, and 76 h after challenge (p < 0.05) (figure 2). In one of the subjects (Subject 11)who developed illness in response to viral challenge, influenza A was isolated from two separate aliquots of prechallenge nasal lavage fluid (1.4 TCIDso/ml [loglOD. This subject did not have a history of recent upper respiratory infection before the study and therefore was presumed to have a subclinical influenza A infection at the time of challenge. Data from this subject will be presented separately and compared with data from the other ill subjects.
Control
SUbject No.
Maximal TCIDsolml (Iog,o)
Fold Rise in HAl Titer
1 4 6 8 10 11 12
1.4 3.7 5.7 6.2 6.0 6.2 3.4
0 32 2 8 4 16 8
2 3 5 7
NI NI NI NI
0 0 0 0
Definition of abbreviations: HAl = hemagglutination inhibition; NI = not isolated. • Viral titers in lavage fluids from subjects challenged with virus were maximal at 28 to 68 h after challenge. Influenza virus was not isolated from control subjects.
1064
~
BJORNSON, MELLENCAMP, AND SCHIFF
Fig. 1. Symptom scores in ill and asymptomatic subjects infected with influenza virus and in subjects who received placebo. Data are expressed as mean ± SEM. Closed circles indicate infected ill group (Subjects 6,8,11, and 12); open triangles indicate infected asymptomatic group (Subjects 1,4, and 10);open squares indicate control group (Subjects 2, 3, 5, and 7).
5
o o
4
(fJ
E o
a. 3 E
J)2
o...........~.........~-,--.~..-.-""""""""T~~~-,--.~..-.-.....--r~..--, -28
-4
20
44
68
92
116140164188212
Hours After Challenge 7 6
........0 0\
5
-=-
4
o
E
Fig. 2. Viral titers in nasal lavage fluids from ill and asymptomatic subjects infected with influenza virus. Data are expressed as mean ± SEM. Closed circles indicate infected ill group (Subjects 6, 8, 11,and 12);open triangles indicate infected asymptomatic group (Subjects 1, 4, and 10).
'03 o
Lf)
U .....
2
20
44
68
92
116
140
164
188
212
Hours After Challenge
Discussion
1400 1200
........
E
