Bidirectional interactions between viral respiratory illnesses and cytokine responses in the first year of life.

Abstract

Viral infections are the major cause of acute wheezing illnesses in childhood. Variations in immunologic responses at birth may be determinants of the risk of acquiring these illnesses.
To determine the immunologic risk factors for virus-induced wheezing in high-risk infants.
The study involves 285 children with a parental history of asthma and/or respiratory allergies. Mononuclear cells obtained at birth (umbilical cord blood) and at 1 year of age were incubated with phytohemagglutinin, respiratory syncytial virus, or rhinovirus, and supernatants were analyzed for IL-5, IL-10, IL-13, and IFN-gamma. Nasal secretions obtained at well child visits and during respiratory illnesses were analyzed for common respiratory viruses.
Respiratory syncytial virus-induced wheezing was associated with reduced phytohemagglutinin-induced IL-13 responses (medians, 213 vs 304 pg/mL; P = .026) from cord blood cells, and similar trends were found for wheezing in general. Furthermore, median IL-13 responses diminished by 28% in non-wheezing children by age 1 year, versus only 3% in wheezing children (P = .013). Children with > or =2 episodes of wheezing had lower phytohemagglutinin-induced IFN-gamma responses and were less likely to have rhinovirus-induced IFN-gamma responses at birth (P < .05). Finally, children with measurable cord blood IFN responses to respiratory syncytial virus were less likely to wheeze in their first year (odds ratio, 0.43 [0.23, 0.79]).
In children with a family history of allergies and/or asthma, mononuclear cell phytohemagglutinin-induced IL-13 and virus-induced IFN-gamma responses at birth are indicative of the risk for wheezing in the first year of life.

Full text

Page 1

Original articles
Bidirectional interactions between viral
respiratory illnesses and cytokine responses
in the first year of life
James E. Gern, MD,aG. Daniel Brooks, MD,bPatricia Meyer, MD,fAndy Chang, BS,a
Kunling Shen, MD,dMichael D. Evans, MS,cChris Tisler, MS,aDouglas DaSilva, BS,a
Kathy A. Roberg, MSN,aLance D. Mikus, MBA,aLouis A. Rosenthal, PhD,aCarole J.
Kirk, BS,ePeter A. Shult, PhD,eAbhik Bhattacharya, MS,cZhanhai Li, PhD,cRonald
Gangnon, PhD,cand Robert F. Lemanske, Jr, MDa,bMadison, Wis, Bejing, China, and
Buenos Aires, Argentina
Background: Viral infections are the major cause of acute
wheezing illnesses in childhood. Variations in immunologic
responses at birth may be determinants of the risk of acquiring
these illnesses.
Objectives: To determine the immunologic risk factors for
virus-induced wheezing in high-risk infants.
Methods: The study involves 285 children with a parental
history of asthma and/or respiratory allergies. Mononuclear
cells obtained at birth (umbilical cord blood) and at 1 year
of age were incubated with phytohemagglutinin, respiratory
syncytial virus, or rhinovirus, and supernatants were analyzed
for IL-5, IL-10, IL-13, and IFN-g. Nasal secretions obtained at
well child visits and during respiratory illnesses were analyzed
for common respiratory viruses.
Results: Respiratory syncytial virus–induced wheezing was
associated with reduced phytohemagglutinin-induced IL-13
responses (medians, 213 vs 304 pg/mL; P 5 .026) from cord
blood cells, and similar trends were found for wheezing in
general. Furthermore, median IL-13 responses diminished by
28% in nonwheezing children by age 1 year, versus only 3%
in wheezing children (P 5 .013). Children with $2 episodes
of wheezing had lower phytohemagglutinin-induced IFN-g
responses and were less likely to have rhinovirus-induced IFN-g
responses at birth (P <.05). Finally, children with measurable
cord blood IFN responses to respiratory syncytial virus were
less likely to wheeze in their first year (odds ratio, 0.43 [0.23,
0.79]).
Conclusion: In children with a family history of allergies and/or
asthma, mononuclear cell phytohemagglutinin-induced IL-13
and virus-induced IFN-g responses at birth are indicative of
the risk for wheezing in the first year of life. (J Allergy Clin
Immunol 2006;117:72-8.)
Key words: Respiratory syncytial virus, rhinoviruses, wheezing,
IL-13, cytokines, viral infections, bronchiolitis, IFN-g, atopy, birth
cohort
Acute episodes of wheezing in infancy are predomi-
nantly caused by viral infections. These illnesses are an
important source of morbidity, especially in the first year
of life, and also foreshadow an increased risk of recurrent
wheezing and potentially asthma. Although practically all
infants contract viral respiratory infections, most do not
wheeze, and this observation has led to the identification
of several predisposing factors, including reduced lung
function,1exposure to tobacco smoke,2prematurity and
other comorbid conditions,3genetic polymorphisms,4-8
and variations in host immune responses. As an example
of the last, quantitative differences in PBMC secretion
of IFN-g and IL-10 ex vivo during and/or after respiratory
syncytial virus (RSV) bronchiolitis may be related to the
subsequent development of asthma.9,10Furthermore, in
prospective studies, reduced IFN-g secretion from mito-
gen-stimulated cord blood cells has been linked to an
increased number of moderate to severe viral respiratory
infections,11andalsotoanincreasedriskofatopy,broadly
defined as including wheezing, atopic dermatitis, and/or
allergy.12Although these findings and others13,14suggest
that deficient TH1-like responses could predispose to at-
opy and asthma, high mitogen-induced IFN-g responses
by cord blood CD8 cells have been linked to subsequent
allergic sensitization.15Moreover, low cord blood IL-13
(TH2) responses to mitogen16and antigen17were found
to be associated with increased risk for atopy and wheez-
ing.Thesefindingssuggestthatcytokineresponseprofiles
in early life are biologically relevant to the risk of atopic
diseases, and indicate that additional studies are needed
to determine whether there are distinct immunologic risk
From the Departments of
Medical
Children’s Hospital;
fUniversidad de Buenos Aires.
Disclosure of potential conflict of interest: R. Lemanske has consultant ar-
rangements with AstraZeneca, Aventis, andNovartis,andis onthespeakers
bureau for Merck, GlaxoSmithKline, AstraZeneca, and Aventis. The rest of
the authors have no conflict of interest to disclose.
Supported by National Institutes of Health grants R01HL61879-01,
P01HL70831-01, and 5M01 RR03186-18.
ReceivedforpublicationApril24,2005;revisedSeptember27,2005;accepted
for publication October 3, 2005.
Available online November 29, 2005.
Reprint requests: James E. Gern, MD, K4/918 CSC, University of Wisconsin
Hospital, 600 Highland Avenue, Madison, WI 53792-9988.
0091-6749/$32.00
? 2005 American Academy of Allergy, Asthma and Immunology
doi:10.1016/j.jaci.2005.10.002
aPediatrics,
University
ethe Wisconsin State Laboratory of Hygiene; and
bMedicine, and
ofWisconsin-Madison;
cBiostatistics and
Informatics,
dBejing
72
Mechanisms of asthma and
allergic inflammation

Page 2

Abbreviations used
COAST: Childhood Origins of ASThma
RSV: Respiratory syncytial virus
factors for wheezing, as opposed to atopic dermatitis and
allergic sensitization.
To test the hypothesis that immune responses that are
present at birth influence the risk of wheezing with
respiratory viral infections, we conducted a prospective
birth cohort study involving children born to high-risk
families in which 1 or both parents had allergic rhinitis or
asthma. Mononuclear cells were collected at birth and at
1 year of age, and cytokine responses of these cells were
compared with the incidence of wheezing illnesses in the
first year of life. Viral diagnostic testing was used to
characterize the specific pathogens associated with these
illnesses. The results of the study indicate the presence
of a bidirectional relationship between early immune
responses and the onset of wheezing. That is, immune
responses contribute to the risk of wheezing, and viral
illnesses help to shape the development of specific
immune responses.
METHODS
Study population and experimental design
After obtaining informed consent, 289 subjects were enrolled in
the Childhood Origins of ASThma (COAST) Study14at birth, and
285 were followed prospectively for at least 1 year. To be eligible,
each of the children were required to have 1 or both parents with
allergic rhinitis (1 or more positive aeroallergen skin tests) and/or
asthma (by history), be delivered at ?37 weeks gestation, and be
otherwise healthy. Data collected by questionnaires included parent
and child health histories with a specific focus on atopic diseases,
behaviors affecting health, and environmental exposures. Details of
study population and design have been described previously.18
This study was approved by University of Wisconsin Human Sub-
jects Com-mittee.
Wheezing respiratory illnesses
A wheezing respiratory illness was defined as meeting 1 or more
of the following criteria: (1) physician-diagnosed wheezing at an
office visit; (2) an illness for which the child was prescribed short or
long-acting b-agonists, inhaled or oral corticosteroids, and/or long-
term controller medications such as cromolyn sodium or leukotriene
inhibitors; or (3) an illness given the following specific diagnoses:
bronchiolitis, wheezing illness, reactive airway disease, asthma, or
asthma exacerbation. Illnesses in which no wheezing was reported
but oral corticosteroids were administered for ‘‘croup’’ or a ‘‘barky
cough’’ reported by the parent were not considered to be wheezing
illnesses.
Immunologic studies
Specimens of cord blood and peripheral blood at age 1 year were
collected as previously described.19Blood mononuclear cells were
incubated with either phytohemagglutinin (5 mg/mL), RSV (104syn-
cytia forming units/mL), type 16 rhinovirus (RV-16, 107plaque
forming units/mL) or medium alone, and supernatant fluids collected
2dayslater(medium,phytohemagglutinin)and6dayslater(medium,
RSV, rhinovirus) were analyzed for cytokines by ELISA.11
Viral diagnostics
Nasopharyngeal mucus specimens were obtained under 2 circum-
stances: (1) scheduled protocol visits at 2, 4, 6, 9, and 12 months of
age, and (2) during acute respiratory illnesses. Parents contacted a
study coordinator when their child developed a respiratory tract
infection, and a respiratory symptom scorecard (maximum score,
31)11wascompleted.Ifthechildscored?5,anasopharyngealmucus
specimen was obtained within 48 hours. The collection and handling
methods of these samples have been described previously.20Nasal
samples were analyzed for RSV, influenza types A and B, parain-
fluenza types 1 to 4, rhinovirus, enteroviruses, and adenoviruses as
previously described.20In addition, ELISA was performed on serum
to detect anti-RSV IgG as previously described11to identify addi-
tional individuals who had RSV infections that were either mild or
asymptomatic and thus not detected by the sampling of nasal secre-
tions during periods of illness.
Statistical analysis
Cord blood and 1-year cytokine responses in subjects with
wheezing episodes in the first year were compared with subjects
who did not wheeze by using the Wilcoxon rank-sum test if at least
60% of the samples were above the detection limit; if not, rates of
detectable cytokine responses in the 2 groups were compared by
using Fisher exact test. Similar analyses compared cytokine produc-
tion between subjects with and without a first-year RSV (rhinovirus)
infection; among those with a first-year RSV (rhinovirus) infection,
cytokine production was compared between subjects who wheezed
with RSV (rhinovirus) and subjects who did not. The development
of cytokine responses during the first year, expressed as the ratio of
1 year to cord blood, was similarly analyzed when sufficient cord
blood production (?90% above detection limit) allowed this
determination.
RESULTS
Study subjects and wheezing episodes
Eighty-nine of the 285 children followed through
infancy (31%) had a total of 179 wheezing episodes (Fig 1).
Viruses were detected in 118 (66%) nasal lavage speci-
mens obtained during wheezing episodes, and the viruses
found most often were RSV (n 5 51) and rhinovirus (n 5
59). Smaller numbers of wheezing episodes were attrib-
uted to parainfluenza (n 5 16), influenza (n 5 4), echovi-
rus (n 5 1), and adenoviruses (n 5 1). There were a small
number of dual infections (n 5 14), most commonly with
RSVandrhinovirus(n59).Sixty-onewheezingepisodes
in 40 children were not associated with a detectable virus.
Finally, the virology of wheezing episodes changed with
the season: RSV (alone or in combination) accounted for
44 of 90 (49%) of wheezing episodes in the winter
(December to February), whereas rhinovirus accounted
for 16 of 43 (37%) of the episodes in the spring and 18
of 33 (55%) episodes in the fall (Fig 2).
Demographic information was compared for wheezing
and nonwheezing children (Table I). We previously re-
ported that, for this cohort of children, the presence of
other siblings in the home and day care attendance were
associated with wheezing illnesses, and particularly those
J ALLERGY CLIN IMMUNOL
VOLUME 117, NUMBER 1
Gern et al 73
Mechanisms of asthma and
allergic inflammation

Page 3

caused by rhinovirus.11Wheezing infants, including in-
fants who wheezed specifically with either rhinovirus or
RSV, were more likely to have mothers with asthma,
although these trends did not reach statistical significance
(Table I). In addition, infants who wheezed with RSV
were more likely to be sensitized to a common food aller-
gen (milk, egg, or peanut) at age 1 year. This was not true
for wheezing infants in general, or infants who wheezed
with rhinovirus. Other subject and family characteristics,
including allergic sensitization, breast-feeding, smoking,
and the presence of pets in the home, were similar among
the groups. No unique demographic features were found
to be associated with wheezing only in the absence of a
detectable virus.
Mitogen-induced cytokine responses at birth
As previously reported,19phytohemagglutinin-induced
IL-10 (median, 100 pg/mL), IL-13 (median, 282 pg/mL),
andIFN-g(median,57pg/mL)responsesfrom cordblood
cells were detectable in the majority of subjects, whereas
IL-5 responses (median, 2.1 pg/mL) were generally mini-
mal or undetectable. There was a statistically significant
reduction in IL-13 responses for children who wheezed
with RSV infections, and a similar trend was noted with
rhinovirus-induced wheeze (Fig 3; see this article’s
TableE1intheOnlineRepositoryatwww.jacionline.org).
In addition, having 2 or more episodes of wheezing
was associated with lower phytohemagglutinin-induced
IFN-g responses (median, 43 pg/mL; 95% CI, 17, 95)
compared with children who wheezed once (68 pg/mL
[37,157]; P 5 .020) or not at all (55 pg/mL [34, 126];
P 5 .030; Fig 4). Phytohemagglutinin-induced IL-5 and
IL-10 responses at the time of birth were not associated
with the risk of wheezing.
Relationship of viral infections and
wheezing to developmental changes
in phytohemagglutinin-induced
cytokine responses
We previously reported that phytohemagglutinin-
induced IL-5 responses increased, IL-10 responses did
not significantly change, and IFN-g and IL-13 responses
decreased in this cohort of infants over the period of the
first year of life.19Patterns of wheezing and symptomatic
viral infections were next compared with developmental
changes in cytokine responses, which were calculated as
the cytokine response at 1 year divided by the cord blood
response.Incontrastwiththedownwardtrendobservedin
thewholecohort,themedianIL-13responsesinwheezing
infants did not change (ratio, 0.97), whereas IL-13 re-
sponses of nonwheezing infants were reduced (0.72); the
difference between the groups was statistically significant
(P 5 .013; Fig 3). IL-13 responses also decreased in chil-
dren who had nonwheezing compared with wheezing
illnesses with either RSV (ratio, 0.72 vs 1.1; P 5 .0095)
or rhinovirus (ratio, 0.76 vs 1.2; P 5 .0079) over the
period of the first year. Changes in IL-13 were similar in
children who had illnesses without wheezing and those
with no documented viral illnesses (Fig 3). The net result
of these interval changes was that group median IL-13
responses at age 1 year were approximately equal.
The frequency of wheezing episodes was associated
with differences in the maturation of phytohemagglutinin-
induced IFN-g responses. One-year-old children with at
least2wheezingepisodeshadgreaterIFN-gresponses(35
pg/mL [22,55]) than children with either 1 (25 pg/mL
[15,42]; P 5 .063) or no wheezing episodes (25 pg/mL
[11, 51];P5.025).Hence,the developmentalchange(ra-
tioyear1:cordblood,medianvalues)forchildrenwith ?2
FIG 1. Flow chart of viral illnesses and specimen analysis. Because
some specimens (n 5 14) contained more than 1 type of virus, the
number of individual viruses detected (n 5 132) exceeds the
number of virus-positive specimens (n 5 118). RV, Rhinovirus;
PIV, parainfluenza virus.
FIG 2. Seasonality and etiology of wheezing episodes. The number
of wheezing episodes was plotted according to the month of onset
of the illness and the types of viruses that were detected. RV,
Rhinovirus.
J ALLERGY CLIN IMMUNOL
JANUARY 2006
74 Gern et al
Mechanisms of asthma and
allergic inflammation

Page 4

wheezing episodes (0.75) was significantly greater than
children with 1 episode (0.31; P 5 .0012) or no wheezing
(0.41; P 5 .0013).
When associations with specific infections were eval-
uated without regard to wheezing status, symptomatic rhi-
novirusinfectionswereassociatedwithasmallerreduction
in phytohemagglutinin-induced IFN-g responses (ratio,
0.50 vs 0.30; P 5 .019), and consequently, greater IFN-g
responses at age 1 year (31 vs 21 pg/mL; P 5 .0055;
see this article’s Table E1 in the Online Repository at
www.jacionline.org). Neither wheezing nor specific viral
infectionswererelatedtothedevelopmentofphytohemag-
glutinin-induced IL-5 or IL-10 responses.
Next, a multivariate analysis was conducted to deter-
minewhetherthepresenceofotherfactorsrelatedtoatopy
and wheezing (maternal and paternal asthma, atopic
dermatitis, and allergic sensitization) affected the rela-
tionship between wheezing, viral infections, and immu-
nologic outcomes. The observed associations were not
significantly altered after accounting for these potential
confounders.
Virus-induced cell cytokine responses at birth
Virus-induced IFN-g responses were generally low at
the time of birth, and only about a third of individuals
secreteddetectablelevelsinresponsetoeitherRSV(Fig5,
A) or rhinovirus (see this article’s Tables E2 and E3 in
the Online Repository at www.jacionline.org). Although
TABLE I. Demographics
WZ
n 5 89
2WZ
n 5 196P
RSV1WZ
n 5 51
RSV2WZ
n 5 88P
RV1WZ
n 5 43
RV2WZ
n 5 170P
Siblings (%)
Mom allergy (%)
Mom asthma (%)
Dad allergy (%)
Dad asthma (%)
RAST food (%)
RAST resp (%)
AD ever (%)
AD active (%)
Dog (%)
Cat (%)
Breast-fed (%)
IgE* (1 y, IU)
Eos* (cells/mL)
67
89
47
84
29
28
14
45
29
30
29
31
49
80
37
78
28
22
12
45
24
38
30
32
.004
.087
.11
.23
.79
.35
.67
.90
.38
.23
.95
.91
.40
.92
65
86
49
88
27
34
14
52
36
27
29
31
57
83
34
79
24
19
14
44
23
35
28
26
.36
.65
.083
.21
.69
.048
.98
.38
.093
.35
.90
.51
.93
.71
72
88
51
90
29
28
16
44
24
33
35
40
56
82
38
77
28
25
13
47
27
34
31
32
.062
.37
.12
.086
.83
.71
.52
.70
.74
.85
.59
.33
.65
.50
13 (5, 28)
198
(108, 337)
3.50
(3.15, 3.76)
27
14 (6, 32)
211
(104, 334)
3.58
(3.18, 3.89)
24
15 (7, 26)
197
(131, 389)
3.49
(3.15, 3.80)
24
12 (4, 34)
201
(104, 303)
3.58
(3.12, 3.86)
34
13 (5, 28)
232
(134, 395)
3.49
(3.15, 3.77)
30
14 (6, 36)
202
(103, 316)
3.55
(3.12, 3.92)
28
Birth weight* (kg) .68 .88.74
Passive smoke (%) .59.19 .80
WZ, wheeze; 2WZ, no wheeze; RSV1WZ, children who wheezed with an RSV infection; RSV2WZ, children who had an RSV infection without wheezing;
RV1WZ, children who had a rhinovirus infection with wheezing; RV2WZ, children who had rhinovirus infections without wheezing; resp, respiratory;
AD, atopic dermatitis; Eos, blood eosinophils at age 1 year.
*Median values with 25th and 75th percentiles in parentheses.
FIG 3. Wheezing and associated interval changes in IL-13 re-
sponses in the first year of life. Phytohemagglutinin-induced IL-
13 responses (median values) at birth and age 1 year were plotted
for nonwheezing infants (No Wz) and those who had at least
1 wheezing illness (Wz). A, Infants categorized according to any
wheezing illness whether or not a virus was isolated. B and C,
Wheezing and nonwheezing illnesses caused by RSV and rhinovi-
rus (RV) respectively. Infants with no known infections with either
RSV or RV are designated No RSV (B) or No RV (C).
FIG 4. The association between phytohemagglutinin-induced IFN-g
responses (median values) at birth and 1 year of age and the
frequency of wheezing episodes. The legend refers to groups of
childrenwith0,1,or?2episodesofwheezinginthefirstyearoflife.
J ALLERGY CLIN IMMUNOL
VOLUME 117, NUMBER 1
Gern et al 75
Mechanisms of asthma and
allergic inflammation

Page 5

detectableRSV-inducedIFN-gresponsesfromcordblood
cellscouldbefoundininfantsbornatanytimeoftheyear,
the month of birth was found to affect the likelihood of
a positive response to RSV, but not rhinovirus (overall
P values 5 .008 and .14, respectively). IFN-g responses
to RSV were more likely to occur in March (P 5 .01)
and were less likely to occur in May (P 5 .004) compared
with other months of the year (data not shown). Both rhi-
novirus and RSV were able to stimulate IL-10 secretion
from most individuals (see this article’s Tables E2 and
E3 in the Online Repository at www.jacionline.org).
Neither rhinovirus nor RSV induced significant secretion
of the TH2-like cytokines IL-5 and IL-13.
Virus-induced cytokine responses at birth were next
compared with wheezing episodes in the first year of life.
Individuals with detectable RSV-induced IFN-g re-
sponses at birth were less likely to wheeze (odds ratio,
0.43;95%CI,0.23,0.79;Fig5,B),andtherewasasimilar
trend towards a reduced risk of RSV-induced wheezing
(odds ratio, 0.45 [0.18, 1.04]; see this article’s Table E2
in the Online Repository at www.jacionline.org). In addi-
tion, rhinovirus-induced cytokine responses at birth were
less likely to be detected in individuals with ?2 wheezing
episodes (14% detectable IFN-g) compared with children
who wheezed once (39% detectable IFN-g; P 5 .016) or
not at all (34% detectable IFN-g; P 5 .020).
Relationship of viral infections and wheezing
to development of virus-induced cytokine
responses
Virus-induced cytokine responses evolved over the
period of the first year of life. For the group as a whole,
RSV-inducedIFN-gresponsesincreasedfrom<4.7pg/mL
to 41 pg/mL at age 1 year (Fig 5, A; P < .001), and a
smaller but significant increase was noted for median
RSV-induced IL-10 responses (73 pg/mL at birth and
103 pg/mL at age 1 year; P < .001). We next compared
episodes of viral infections and wheezing to the observed
changes in cytokine responses. Of the 285 study subjects,
44.5% developed an RSV infection. In comparing RSV-
infected versus uninfected subjects, RSV-induced IFN-g
responses tended to be greater (62 vs 35 pg/mL; P 5
.077) and IL-10 responses tended to be lower (97 vs 110
pg/mL; P 5 .10) at age 1 year. Developmental patterns
of IFN-g and IL-10 responses to rhinovirus were quite
similar to those induced by RSV (see this article’s Table
E3 in the Online Repository at www.jacionline.org), but
they did not vary with infection or wheezing status.
DISCUSSION
Altered immune responses have been measured after
severe respiratory infections in infancy, particularly those
caused by RSV, but understanding the interactions be-
tween viral infections and immune responses has been
limitedbyapaucityofinformationdescribingpreinfection
immune responses. In this study, we prospectively mea-
sured immune responses and identified specific viral
respiratory infections in early infancy. Our results dem-
onstrate that mitogen-induced and cytokine-induced
responses are immature at birth, and the quality of these
responses is related to the risk of wheezing. In particular,
vigorous IL-13 and IFN-g responses to phytohemagglu-
tinin and greater secretion of IFN-g in response to viruses
were associated with a reduced risk of developing wheez-
ing. Furthermore, infants who developed wheezing ill-
nesses in the first year of life had a distinct pattern of
immunologic development in regard to mitogen-induced
IL-13 and IFN-g, whereas symptomatic rhinovirus infec-
tions with or without wheezing were generally associ-
ated with enhanced IFN-g responses. When considered
together, these findings provide evidence of a bidirec-
tional interaction between immune responses and viral
infections.
Considering the broad-based evidence implicating the
TH2-like cytokine IL-13 in the pathogenesis of asthma,21
the relationship between low mitogen-induced IL-13 and
increased risk of wheeze seems paradoxical. Williams
et al16measured cord blood phytohemagglutinin-induced
IL-13 responses in 43 newborns on the basis of parental
history, and then compared cytokine responses to atopic
outcomes (skin tests, wheezing, and eczema) at age 3
years. High-risk infants who developed atopic diseases
had lower amounts of IL-13 compared with the low-risk
infants.IL-13responsesofhigh-riskinfantswithnoatopic
disease by age 3 years were intermediate, but not statisti-
cally different from the other groups. In the COAST
cohort, the larger sample size enabled separate statistical
analysis of associations between cytokine production
and specific atopic phenotypes: low cord blood IL-13
responses were associated with wheezing, but not atopic
dermatitis, peripheral blood eosinophilia, or total or aller-
gen-specific IgE levels.19Collectively, these findings
provide evidence that there are distinct immunologic
risk factors for wheezing in infancy compared with other
outcomes such as atopic dermatitis and allergy. In fact,
although several previous studies of immunologic risk
factors have broadly defined atopy to include recurrent
wheezing,12,16the great majority of wheezing illnesses
in infancy are caused by viral infections,22and most
wheezing infants do not go on to develop allergies and
asthma.23
FIG 5. RSV-induced IFN-g responses. Phytohemagglutinin-induced
IFN-g responses increased over the period of the first year of life
(A). Detectable IFN-g responses at birth were associated with a
reduced rate of wheezing in the first year of life (B).
J ALLERGY CLIN IMMUNOL
JANUARY 2006
76 Gern et al
Mechanisms of asthma and
allergic inflammation

Page 6

It is possible that low IL-13 and IFN-g production at
birth are indicators of an immature immune system with a
diminished antiviral response.16Our data also raise the
possibility that IL-13, or other factors related to the TH2
response, could actually have beneficial effects during
viral respiratory infections. Although we were unable to
demonstrate a direct antiviral effect of IL-13 (data not
shown), it is possible that the production of an appropriate
amount of IL-13 during an infection could moderate the
degree of virus-induced inflammation. In fact, IL-13 in-
hibits monocyte and macrophage production of proin-
flammatory factors including IL-1b, TNF-a, and nitric
oxide.24,25IL-13 also promotes the synthesis of mucus,26
which could be an important component of the innate an-
tiviral response. Data generated in a rat model of parain-
fluenza type 1 (Sendai) virus infection also suggest that
appropriate secretion of IL-13 (or a related factor) could
also serve to limit the extent of virus-induced inflamma-
tion and airway dysfunction during respiratory infections.
Wheninfectedatayoungage,BrownNorwayrats,butnot
F344 rats, develop chronic relapsing airway obstruction
and hyperresponsiveness.27Remarkably, during the acute
infection F344 rats generate greater amounts of IL-13 in
the lung, whereas the Brown Norway rats have a slow
onset of IL-13 production and then fail to downregulate
this cytokine as the acute illness subsides.27
Wheezing episodes were associated with a unique
developmental pattern of IL-13 and IFN-g cytokine
response profiles. Although median IL-13 responses
declined by about 25% in nonwheezers, they were un-
changedorgreaterinchildrenwhowheezed.Furthermore,
recurrent wheezing was associated with lower phytohem-
agglutinin-induced IFN-g responses at birth, but by age
1 year, this trend had reversed, and responses were
significantly higher compared with children who wheezed
once or not at all. In a previous prospective study reported
by Prescott et al,17atopic children had greater cord blood
IL-13 responses compared with nonatopic individuals,
but by 2 years of age, the pattern had reversed, and greater
IL-13 responses were found in atopic infants. We are con-
tinuingtotrackmitogen-inducedIL-13responsesyearlyto
testthehypothesisthatwheezinginfantswhodonotdown-
regulate IL-13 responses may be at greater risk for the
subsequent development of asthma.
Interactions between virus-specific immune responses
and wheezing were also observed. Notably, the subset of
children with detectable IFN-g responses to RSV at birth
were less likely to wheeze, whereas responses to rhinovi-
ruswere lesscommoninchildrenwithfrequentwheezing.
These findings raise questions about the mechanism of
virus-induced IFN-g responses in newborns, and at least 2
explanations have been proposed. First, RSV infections
occurring during the latter half of a pregnancy could lead
to sensitization of the fetus,28although this idea is contro-
versial.29In our study, IFN-g responses to either RSV or
rhinovirus were elicited in cord blood cells from children
born throughout the year, although RSV responses were
more likely to occur in samples obtained in March, just
after the peak of the RSV season in Madison. This finding
supports the concept that maternal exposures to RSV in
the weeks before delivery could modify the subsequent
response of the newborn. Alternatively, RSV can activate
cells through interactions with Toll-like receptor 4,30,31
providing a potential mechanism for innate immune stim-
ulation. Whether rhinovirus also interacts with innate sur-
facereceptorsisunknown.TheIFN-gresponsestoviruses
in our subjects were also developmentally regulated: they
wereeitherabsentorlowatbirth,andthenincreasedinthe
first year even in the absence of RSV infection as ascer-
tained by negative serology. Collectively, these findings
provide evidence of rapid development in antiviral re-
sponses during the first year oflife, andsuggest that accel-
eratedmaturationofresponsestoRSVcouldbebeneficial.
The COAST study was designed to evaluate how
interactions between immune responses and viral infec-
tions in early life affect the development of wheezing
illnesses and eventually asthma, and the experimental
design has several strengths, and some limitations, in this
regard. The strengths of the study include its prospective
nature, which enabled the measurement of immune
responses before and after viral infections and wheezing
episodes, documentation of specific viral pathogens, large
sample size, and excellent subject retention (98.6% at
1 year). One of the limitations of the study is that the
cohort includes only children from families in which at
least 1 parent has allergies or asthma. Some previous
studies have demonstrated immunologic differences
between healthy infants born to atopic versus nonatopic
parents.32,33Therefore, our data must be interpreted keep-
ing in mind that the healthy children in the COAST study,
who serve as our comparator group, could be immunolog-
ically different than healthy children of parents without
allergies or asthma. Nevertheless, determining risk factors
for childhood allergy and asthma is perhaps of greatest
interest to affected parents and high-risk families, and the
genetic and environmental relationships found in the
COAST study are relevant to these important questions.
One other potential limitation was the use of ex vivo
responses of peripheral blood cells as an indicator of
immune development. This approach was selected because
of the difficulty of obtaining specimens of airway cells,
and particularly lower airway samples, in this age group.
Nasalsecretions werecollectedatbothsickandwell-child
visits, and we are in the process of analyzing these speci-
mens to gain insight into patterns of immunologic matura-
tion in the airways.20
In summary, this study provides evidence of a bidirec-
tional interaction between immune responses in early life
and the development of wheezing illnesses. Associations
between specific cytokine responses and clinical wheez-
ing provide insight into the immunopathogenesis of viral
respiratory illnesses in early life, and complement a
growingbodyofinformationderivedfromgeneticstudies.
In addition, the suggestion that stronger responses to
viruses such as RSV and rhinovirus at the time of birth
could protect against wheezing provides a potential target
for the development of new approaches to the prevention
of wheezing illnesses in infancy. We are continuing to
J ALLERGY CLIN IMMUNOL
VOLUME 117, NUMBER 1
Gern et al 77
Mechanisms of asthma and
allergic inflammation

Page 7

track the relationship between immune responses and the
evolution of wheezing phenotypes in these children in an
effort to determine whether specific patterns of immune
responsesininfancymodifythe riskofdevelopingasthma
later in childhood.
REFERENCES
1. Martinez FD, Morgan WJ, Wright AL, Holberg CJ, Taussig LM. Dimin-
ished lung function as a predisposing factor for wheezing respiratory
illness in infants. N Engl J Med 1988;319:1112-7.
2. Gold DR, Burge HA, Carey V, Milton DK, Platts-Mills T, Weiss ST.
Predictors of repeated wheeze in the first year of life: the relative roles
of cockroach, birth weight, acute lower respiratory illness, and maternal
smoking. Am J Respir Crit Care Med 1999;160:227-36.
3. Shay DK, Holman RC, Roosevelt GE, Clarke MJ, Anderson LJ. Bron-
chiolitis-associated mortality and estimates of respiratory syncytial
virus-associated deaths among US children, 1979-1997. J Infect Dis
2001;183:16-22.
4. Hoffjan S, Ostrovnaja I, Nicolae D, Newman DL, Nicolae R, Gangnon
R, et al. Genetic variation in immunoregulatory pathways and atopic
phenotypes in infancy. J Allergy Clin Immunol 2003;113:511-8.
5. Choi EH, Lee HJ, Yoo T, Chanock SJ. A common haplotype of inter-
leukin-4 gene IL4 is associated with severe respiratory syncytial virus
disease in Korean children. J Infect Dis 2002;186:1207-11.
6. Hull J, Rowlands K, Lockhart E, Moore C, Sharland M, Kwiatkowski D.
Variants of the chemokine receptor CCR5 are associated with severe
bronchiolitis caused by respiratory syncytial virus. J Infect Dis 2003;
188:904-7.
7. Hull J, Thomson A, Kwiatkowski D. Association of respiratory syncytial
virus bronchiolitis with the interleukin 8 gene region in UK families.
Thorax 2000;55:1023-7.
8. Lahti M, Lofgren J, Marttila R, Renko M, Klaavuniemi T, Haataja R,
et al. Surfactant protein D gene polymorphism associated with severe
respiratory syncytial virus infection. Pediatr Res 2002;51:696-9.
9. Renzi PM, Turgeon JP, Marcotte JE, Drblik SP, Be ´rube ´ D, Gagnon MF,
et al. Reduced inteferon-g production in infants with bronchiolitis and
asthma. Am J Respir Crit Care Med 1999;159:1417-22.
10. Bont L, Heijnen CJ, Kavelaars A, Van Aalderen WM, Brus F, Draaisma
JT, et al. Monocyte IL-10 production during respiratory syncytial virus
bronchiolitis is associated with recurrent wheezing in a one-year fol-
low-up study. Am J Respir Crit Care Med 2000;161:1518-23.
11. Copenhaver CC, Gern JE, Li Z, Shult PA, Rosenthal LA, Mikus LD,
et al. Cytokine response patterns, exposure to viruses, and respiratory
infections in the first year of life. Am J Respir Crit Care Med 2004;
170:175-80.
12. Tang MLK, Kemp AS, Thorburn J, Hill DJ. Reduced interferon-g pro-
duction in neonates and subsequent atopy. Lancet 1994;344:983-5.
13. Upham JW, Lee PT, Holt BJ, Heaton T, Prescott SL, Sharp MJ, et al.
Development of interleukin-12-producing capacity throughout childhood.
Infect Immun 2002;70:6583-8.
14. Holt PG, Sly PD. Interactions between respiratory tract infections and
atopy in the aetiology of asthma. Eur Respir J 2002;19:538-45.
15. Rowe J, Heaton T, Kusel M, Suriyaarachchi D, Serralha M, Holt BJ,
et al. High IFN-gamma production by CD81 T cells and early sensitiza-
tion among infants at high risk of atopy. J Allergy Clin Immunol 2004;
113:710-6.
16. Williams TJ, Jones CA, Miles EA, Warner JO, Warner JA. Fetal and
neonatal IL-13 production during pregnancy and at birth and subsequent
development of atopic symptoms. J Allergy Clin Immunol 2000;105:
951-9.
17. Prescott SL, Macaubas C, Smallacombe T, Holt BJ, Sly PD, Holt PG.
Development of allergen-specific T-cell memory in atopic and normal
children. Lancet 1999;353:196-200.
18. Lemanske RF Jr. The childhood origins of asthma (COAST) study.
Pediatr Allergy Immunol 2002;13(suppl 15):38-43.
19. Neaville WA, Tisler C, Anklam K, Gilbertson-White S, Hamilton R,
Adler K, et al. Developmental cytokine response profiles and the clinical
and immunologic expression of atopy in infancy. J Allergy Clin Immunol
2003;112:740-6.
20. Gern JE, Martin MS, Anklam KA, Shen K, Roberg KA, Carlson-Dakes
KT, et al. Relationships among specific viral pathogens, virus-induced
interleukin- 8, and respiratory symptoms in infancy. Pediatr Allergy
Immunol 2002;13:386-93.
21. Wills-Karp M. Interleukin-13 in asthma pathogenesis. Immunol Rev
2004;202:175-90.
22. Heymann PW, Carper HT, Murphy DD, Platts-Mills TA, Patrie J,
McLaughlin AP, et al. Viral infections in relation to age, atopy, and sea-
son of admission among children hospitalized for wheezing. J Allergy
Clin Immunol 2004;114:239-47.
23. Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M, Morgan
WJ, et al. Asthma and wheezing in the first six years of life. N Engl J
Med 1995;332:133-8.
24. Minty A, Chalon P, Derocq JM, Dumont X, Guillemot JC, Kaghad M,
et al. Interleukin-13 is a new human lymphokine regulating inflammatory
and immune responses. Nature 1993;362:248-50.
25. Bogdan C, Thuring H, Dlaska M, Rollinghoff M, Weiss G. Mechanism
of suppression of macrophage nitric oxide release by IL-13: influence of
the macrophage population. J Immunol 1997;159:4506-13.
26. Wynn TA. IL-13 effector functions. Annu Rev Immunol 2003;21:
425-56.
27. Sorkness RL, Gern JE, Grindle KA, Mosser AG, Rosenthal LA, Mikus
LD, et al. Persistence of viral RNA in 2 rat strains differing in suscepti-
bility to postbronchiolitis airway dysfunction. J Allergy Clin Immunol
2002;110:607-9.
28. Legg JP, Jones CA, Warner JA, Johnston SL, Warner JO. A hypothesis:
antenatal sensitisation to respiratory syncytial virus in viral bronchiolitis.
Arch Dis Child 2002;86:431-3.
29. Krishnan S, Halonen M, Welliver RC. Innate immune responses in
respiratory syncytial virus infections. Viral Immunol 2004;17:220-33.
30. Kurt-Jones EA, Popova L, Kwinn L, Haynes LM, Jones LP, Tripp RA,
et al. Pattern recognition receptors TLR4 and CD14 mediate response to
respiratory syncytial virus. Nat Immunol 2000;1:398-401.
31. Haeberle HA, Takizawa R, Casola A, Brasier AR, Dieterich HJ,
Van Rooijen N, et al. Respiratory syncytial virus-induced activation
of nuclear factor-kappaB in the lung involves alveolar macrophages
and toll-like receptor 4-dependent pathways. J Infect Dis 2002;186:
1199-206.
32. Miles EA, Warner JA, Jones AC, Colwell BM, Bryant TN, Warner JO.
Peripheral blood mononuclear cell proliferative responses in the first
year of life in babies born to allergic parents. Clin Exp Allergy 1996;
26:780-8.
33. Miles EA, Warner JA, Lane AC, Jones AC, Colwell BM, Warner JO.
Altered T lymphocyte phenotype at birth in babies born to atopic parents.
Pediatr Allergy Immunol 1994;5:202-8.
J ALLERGY CLIN IMMUNOL
JANUARY 2006
78 Gern et al
Mechanisms of asthma and
allergic inflammation