INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The occurrence of recurrent episodes of wheezing in early childhood after respiratory syncytial virus bronchiolitis has been well documented (1). Actually, respiratory syncytial virus (RSV) bronchiolitis is followed by recurrent wheezing in 20 to 80% of the cases (6). It is thought that these wheezing episodes are triggered by viral upper respiratory tract infections, and they appear to be independent of atopy (9). In addition, follow-up studies have shown bronchial hyperresponsiveness 4 to 8 yr after hospitalization for RSV bronchiolitis (6, 7, 10).

Although RSV infection induces cytokine production by a number of cells in vivo and in vitro, the relation of these cytokine responses to recurrent wheezing is largely unknown. CD4+ T-cells can be functionally divided into Th1 and Th2 cells (11). This division is based on the profile of cytokine production. Th1 cells selectively secrete IFN- and promote cell-mediated immunity. In contrast, Th2 cells secrete IL-4, IL-5, and IL-13. These cytokines are involved in humoral immunity and are thought to contribute to allergic asthmatic inflammation. Recent studies have suggested that a Th2 cytokine profile during RSV bronchiolitis is associated with wheezing during follow-up (12, 13).

To our knowledge, the role of monocytes/macrophages in recurrent wheezing after RSV bronchiolitis is unknown. However, it is evident that alveolar macrophages participate in the immune response during RSV infections. Alveolar macrophages recovered from bronchoalveolar lavage (BAL) fluid from children with severe RSV bronchiolitis are quantitatively the most important cell type (14). In vitro data show that monocytes can readily be infected by RSV (15, 16). They have the ability to produce a spectrum of cytokines, including IL-12 and IL-10. Interleukin-12 is required for the initiation of the antiviral immune response (17, 18), whereas IL-10 has several properties, including downregulation of cytokine production by Th1-like T-cells and inhibition of antigen presentation by antigen-presenting cells (APC) (19, 20). Alveolar macrophages in BAL fluid from patients with RSV bronchiolitis show increased expression of proinflammatory cytokines, including IL-1 and TNF-, as compared with healthy control subjects (21). In addition, in vitro data show induction of IL-6, IL-8, and IL-10 production by macrophages infected with RSV (22).

The aim of this study was to determine whether cytokine responses during the acute and the convalescent phases of RSV bronchiolitis are associated with the subsequent development of recurrent wheezing. We therefore studied T-cell and monocyte cytokine responses in hospitalized children with RSV bronchiolitis and related these cytokine responses to recurrent episodes of wheezing during a 1-yr follow-up period.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Study Population

Fifty children were entered into the study during one winter epidemic in four hospitals in The Netherlands (Wilhelmina Children's Hospital, Utrecht; Beatrix Children's Hospital, Groningen; St. Elisabeth Hospital, Tilburg; Rijnstate Hospital, Arnhem). Inclusion criteria were: hospital admission, lower respiratory tract symptoms, age < 13 mo, and immunofluorescence for RSV infection of epithelial cells in nasopharyngeal secretions. Infants with wheezing illness prior to RSV bronchiolitis were not included. One investigator (LB) recorded the history of atopy in parents, grandparents, and siblings (asthma, documented food allergy, eczema, hay fever) and inquired whether either parent had smoked in the presence of the infant during the follow-up period. Twenty-seven control children < 13 mo of age without evidence of atopy or infection were selected for this study during the same winter season. Included were infants prior to minor surgery, prior to cardiac surgery in the absence of hemodynamic compromise, healthy prematurely born infants, healthy infants screened for congenital disorders, and infants with mild anemia. The study was approved by the Medical Ethical Committee in all participating centers. Parents of both patients and control subjects gave written informed consent.

Whole Blood Cultures

Heparinized venous or arterial blood was taken within 24 h after admission from subjects in the Wilhelmina Children's Hospital, Utrecht (n = 24). Three to four weeks later, during the convalescent phase, heparinized blood was taken from all subjects (n = 50). Freshly taken heparinized blood was diluted 1:10 in RPMI 1640 medium (Life Technologies, Grand Island, NY) and aliquoted (150 µl) into 96-well culture plates (Nunc International, Roskilde, Denmark).

The whole-blood culture stimulated with lipopolysaccharide (LPS) is a suitable ex vivo method to study monocyte cytokine production under conditions in which many of the physiologically relevant cellular interactions remain intact (23). To induce monocyte IL-10 and IL-12 production, LPS (100 ng/ml) + IFN- (20 ng/ml) (LPS + IFN-) were added and cultures were incubated for 48 h at 37° C in 5% CO₂. It has been shown that maximal monocyte IL-10 production is observed after 48 h, which is relatively late compared with that of monocyte proinflammatory cytokines (20). Also, monocyte IL-12 production is (sub)optimally induced after 48 h of stimulation (24). Furthermore, it has been established that monocytes are the main producers of IL-10 and IL-12 in LPS-stimulated, whole-blood cultures (25).

To induce lymphocyte cytokine production, phytohemagglutinin (PHA) (50 µg/ml) or anti-CD2,1 (1:12,000) + anti-CD2,2 (1:12,000) + anti-CD2,8 (1:3,000) monoclonal antibodies (anti-CD2/28 Moabs; CLB, Amsterdam, The Netherlands) were added and cultures were incubated for 48 h at 37° C in 5% CO₂.

All cultures were performed in quadruplicate. Pooled supernatants were kept at 70° C.

Cytokine Assays

In supernatants of LPS + IFN--stimulated blood cultures, IL-12 and IL-10 were measured. In supernatants of PHA-stimulated cultures IFN- and IL-4 were measured. Concentrations of IL-10, IFN-, and IL-4 were determined using ELISA kits supplied by the Dutch Laboratory for Blood Transfusion (CLB, Amsterdam, The Netherlands). The detection limit for IL-10 was 2.5 pg/ml, for IFN- it was 25 pg/ml, and for IL-4 it was 1.0 pg/ml. Concentrations of IL-12 were determined using ELISA kit from R&D (Oxon, UK), the detection limit was 7.8 pg/ml. This assay recognizes only the IL-12 heterodimer and not the individual subunits of the dimer. When cytokine values were not detectable, for statistical analysis the minimum detectable level was used. When IL-10 responses were above the maximum detectable level (300 pg/ml), this level was used.

Follow-up Evaluations

Follow-up data were collected for 1 yr after discharge. Follow-up was performed using diaries that were developed for this study. One investigator instructed parents how to use the diary. Starting 3 wk after discharge from the hospital, parents noted the presence of "coughing" and "wheezing" on a daily base. A disease episode was defined as the presence of symptoms for two or more consecutive days. At the end of the follow-up period, patients were classified as "recurrent wheezing" if more than one episode of wheezing was noted after discharge. In addition, at the end of the study period, one investigator contacted the general practitioners of the patients by telephone and inquired if "asthma" had been diagnosed.

Statistical Analysis

Cytokine production and IFN-/IL-4 ratios were analyzed after logarithmic transformation. Mean (geometric mean) and standard error of mean (SEM) of cytokine levels were calculated using logarithmically transformed values. Chi-square test was used to evaluate whether sex, the need for mechanical ventilation, or the presence of a positive family history of atopy were associated with recurrent wheezing. Differences in age at onset of disease, IL-10 and IL-12 responses and IFN-/ IL-4 ratios between infants with and without recurrent wheezing and infants with and without a family history of atopy were analyzed using Student's unpaired t test. The relation between cytokine response in the acute and convalescent phase were analyzed using Student's paired t test. Pearson's correlation coefficient was used to analyze the relation between cytokine levels and the number of reported wheezing episodes. All tests of significance were two-sided.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subject Characteristics

The investigated population consisted of 50 patients and 27 control subjects. Twenty-nine patients (58%) were boys, and median age was 3 mo. Eleven patients (22%), including three pairs of twins, were born prematurely (range: 29 4/7 to 36 5/7 wk). In the control group, 17 children were boys (63%), the median age was 4 mo, and two control subjects were born prematurely. Thirty-six patients (72%) and 18 control subjects (67%) had a positive family history of atopy. Respiratory distress was present in all children. Three infants had had apnea prior to admission. Fourteen subjects (28%) needed mechanical ventilation. In one child chronic lung disease was diagnosed, none of the patients had cardiac disease or an immunodeficiency. None of the patients had received ribavirin or systemic anti-inflammatory agents, including corticosteroids. Patients did not receive inhaled corticosteroids during RSV bronchiolitis. All patients survived.

Cytokine Responses

Cytokine responses in LPS + IFN--stimulated, whole-blood cultures from patients during the acute phase (n = 24) and convalescent phase (n = 50) are shown in Figures 1 and 2. During the acute phase, IL-12 production was significantly lower in patients than in control subjects (geometric mean 28 versus 66 pg/ml, p = 0.007). During the convalescent phase, IL-12 responses in patients increased to levels that were not significantly different from those of control subjects (44 pg/ml). In contrast, the amount of IL-10 produced during the acute phase was not significantly different from that in the control subjects (26 versus 38 pg/ml). During the convalescent phase, however, IL-10 production (112 pg/ml) was significantly higher than in the acute phase (p < 0.001) and higher compared with that in control subjects (p < 0.001).

View larger version (12K): [in this window] [in a new window]   Figure 1.   Ex vivo IL-12 production during the acute and convalescent phase of RSV bronchiolitis. Blood was obtained from patients with RSV bronchiolitis within 24 h after admission to the hospital (acute phase) and 3 to 4 wk later (convalescent phase) and controls. Whole-blood cultures were stimulated for 48 h with LPS (100 ng/ml) and IFN- (20 ng/ml). Data represent individual values and mean ± SEM. Data were analyzed after log transformation.

View larger version (14K): [in this window] [in a new window]   Figure 2.   Ex vivo IL-10 production during the acute and convalescent phase of RSV bronchiolitis. Samples and cultures are those described in Figure 1. Data represent individual values and mean ± SEM. Data were analyzed after log transformation.

In PHA-stimulated whole-blood cultures of patients during the acute phase of disease, both IL-4 and IFN- responses were low (n = 24). Actually, in 11 patients, levels of either cytokine were below detection limits. During the acute phase of disease, IL-4 production was lower than in control subjects (2.9 versus 9.6 pg/ml, p < 0.001), as were IFN- responses (89 versus 602 pg/ml, p = 0.001). In the convalescent phase both IL-4 and IFN- responses returned to values that were not significantly different from those of control subjects (n = 49). In one patient no supernatant of PHA-stimulated whole-blood cultures was collected for technical reasons.

In cultures stimulated with CD2 + CD28 the same response pattern of IL-4 and IFN- were seen. During the acute phase, in comparison with control subjects, we found decreased production of IL-4 (4.2 versus 8.1 pg/ml, p = 0.03) and IFN- (8.1 versus 186 pg/ml) (p < 0.001). In the convalescent phase IL-4 response (6.6 pg/ml) and IFN- response (170 pg/ml) were comparable to those in control subjects.

Follow-up Data

Forty-six patients in follow-up (92%) returned their diaries; the other four patients were lost to follow-up. Twenty-seven children (59%) had two or more episodes of wheezing (range, 2 to 11 wheezing episodes). In patients requiring mechanical ventilation, 43% had recurrent episodes of wheezing in the follow-up period, which was not significantly different from that in nonventilated infants. The number of episodes of wheezing correlated strongly with the number of episodes of coughing (r = 0.76, p < 0.001).

One or more episodes of coughing in follow-up was documented in 43 patients (93%). In four patients (9%) one episode of coughing was noted, in seven (15%) two episodes of coughing were noted and in 32 (70%) three or more episodes of coughing were noted. Asthma was diagnosed by a physician in 16 patients (35%). These patients all had more than one documented episode of wheezing.

A positive history of atopy was noted in 19 infants with recurrent wheezing (70%) and in 15 infants without recurrent wheezing (79%), which was not significantly different. Six infants were exposed to tobacco smoke by at least one parent during follow-up, three had recurrent episodes of wheezing. Age at onset of disease and sex were not associated with the occurrence of recurrent wheezing.

IL-10 responses during the convalescent phase were significantly higher in infants with recurrent wheezing than in those without recurrent wheezing (p = 0.006) (Figure 3). The difference was found between IL-10 responses in the convalescent phase in infants with and without physician diagnosed asthma (p = 0.004). Moreover, IL-10 levels during the convalescent phase correlated with the number of episodes of wheezing (r = 0.42, p = 0.004) (Figure 4).

View larger version (12K): [in this window] [in a new window]   Figure 3.   Ex vivo IL-10 production in patients with and without subsequent recurrent wheezing and physician-diagnosed asthma. Blood was obtained from patients with RSV bronchiolitis 3 to 4 wk after admission (convalescent phase) and from control subjects. Cultures are described in Figure 1. Data represent individual values and mean ± SEM. Data were analyzed after log transformation.

View larger version (10K): [in this window] [in a new window]   Figure 4.   Relation between ex vivo IL-10 production and number of wheezing episodes during a 1-yr follow-up period. Blood was obtained from patients with RSV bronchiolitis 3 to 4 wk after admission (convalescent phase). Cultures are those described in Figure 1. Data represent individual values. Linear regression line is drawn.

We considered the possibility that the association between IL-10 responses and recurrent wheezing could only be found during the first months of the follow-up period after RSV bronchiolitis. We therefore evaluated the association between IL-10 responses during the convalescent phase and wheezing during the last 3 mo of the study period (the winter season). During this period 20 patients had one or more wheezing episode, and again, a difference was found in IL-10 responses between infants with and without a wheezing episode (p = 0.02).

Interleukin-12 responses were not associated with recurrent wheezing during follow-up. In addition, no differences in IL-4 and IFN- responses in both PHA- and CD2 + CD28-stimulated blood cultures were found between wheezing and nonwheezing infants. Moreover, IFN-/IL-4 ratios in PHA- stimulated cultures (Figure 5A) during the convalescent phase were comparable for wheezing and nonwheezing infants. As expected, in infants with a positive family history of atopy, decreased IFN-/IL-4 ratios were found in PHA-stimulated cultures (p < 0.05) (Figure 5B). In CD2 + CD28-stimulated cultures the same association was seen, although this did not reach a significant level. Monocyte cytokine responses, including IL-10 responses during the convalescent phase, were not associated with family history of atopy.

View larger version (12K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   Figure 5.   (A) Relation between ex vivo IFN-/IL-4 ratios in PHA- stimulated whole-blood cultures and the development of recurrent wheezing after RSV bronchiolitis. Blood was obtained from patients with RSV bronchiolitis 3 to 4 wk after admission to the hospital (convalescent phase) and from control subjects. Whole-blood cultures were stimulated for 48 h with PHA (50 µg/ml). Data represent individual values and mean ± SEM. Data were analyzed after log transformation. (B) Ex vivo IFN- and IL-4 production and IFN-/IL-4 ratios in PHA-stimulated whole blood cultures in patients with a positive/negative family history of atopy. Samples and cultures are those described in A. Data represent individual values and mean ± SEM. Data were analyzed after log transformation.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

The results of this study demonstrate that increased monocyte IL-10 response in vitro upon stimulation with a specific stimuli during the convalescent phase of RSV bronchiolitis is associated with the development of subsequent recurrent episodes of wheezing during a 1-yr follow-up period. Moreover, there was a highly significant correlation between IL-10 production in the convalescent phase and the number of wheezing episodes. The same association was found between IL-10 response and physician-diagnosed asthma. Recurrent wheezing during the year after clinical bronchiolitis was not associated with a family history of atopy. In agreement with the latter finding, T-cell cytokine responses (IFN- and IL-4) in vitro upon stimulation with nonspecific stimuli were not associated with recurrent wheezing.

Clinical risk factors for the development of recurrent wheezing after RSV bronchiolitis were established by previous investigators (9, 26). These risk factors include male sex, low age at onset of disease, and disease severity. In our study, boys were not more likely to develop recurrent wheezing than were girls, consistent with findings of Sims and colleagues (6), but not with the findings of McConnochie and Roghmann (29), who found an increased risk for boys. Furthermore, we did not find a relation between either age at onset or disease severity during RSV bronchiolitis and subsequent recurrent wheezing, consistent with most other studies (3, 6, 7). Finally, some reports indicate an increased risk for post-bronchiolitis wheezing in children with a positive family history of atopy, but most recent reports, including ours do not support this suggestion (6, 8, 30).

Ex-vivo IL-10 production was significantly increased during the convalescent phase of RSV infection. Although IL-10 can be produced by different cell types, including monocytes, Th2 cells and B cells, it has been shown that IL-10 in LPS-stimulated whole blood cultures, IL-10 is most likely monocyte- derived (19, 20, 25). In the LPS + IFN--stimulated cultures in the present study, IL-8 was decreased significantly (data not shown), indicating that the increased IL-10 response is not explained by a general increase in monocyte cytokine responses.

Two mechanisms can explain changes in cytokine responses by monocytes in peripheral blood during RSV bronchiolitis. The presence of RSV ribonucleic acid (RNA) in the blood and the potential to cause productive infection in vitro in monocytes suggest that during RSV bronchiolitis, changes in monocyte function could result from direct infection (15, 31). Another explanation is that changes in monocyte cytokine responses are the systemic consequence of local production of cytokines and other mediators by epithelial cells and macrophages in the respiratory tract during RSV infection. Finally, we note that the immune response in respiratory tract and changes in cytokine production by local macrophages are potentially different from what is found in circulating monocytes. More research is required to evaluate whether cytokine responses by circulating monocytes reflect cytokine responses by macrophages in the respiratory tract.

Ex-vivo IL-12 production was significantly decreased during the acute phase of RSV bronchiolitis. Different viruses, including the measles virus, have been shown to inhibit IL-12 production in vitro by monocytes/macrophages (24, 32). Although the effect of RSV on IL-12 production by monocytes/ macrophages has not been investigated, it is conceivable that RSV itself effectively inhibits IL-12 production. More studies are needed to evaluate whether low IL-12 responses play a role in the pathogenesis of acute RSV bronchiolitis.

We propose two possible mechanisms by which increased production of monocyte IL-10 leads to recurrent wheezing. On the one hand, increased monocyte/macrophage IL-10 responses in vivo may result in suppression of Th1 cells and enhancement of Th2 cells by antagonizing IL-12 (33). As a result, this could then lead to allergic asthmatic airway inflammation. This latter possibility is not supported by our data, which show the absence of an association between IFN-/IL-4 ratios and recurrent wheezing. On the other hand, it is conceivable that in vivo increased IL-10 production leads to decreased antiviral immunity in the lower airways, as a result of suppression of antigen presentation by pulmonary macrophages. One could then speculate that viral infections in the upper respiratory tract more easily lead to infection and inflammation of the lower respiratory tract, leading to wheezing and bronchial hyperresponsiveness. This explanation is in line with the clinical picture of wheezing after RSV bronchiolitis, usually associated with upper respiratory symptoms (8).

In this study we found decreased IFN- and IL-4 responses in patients in the acute phase of RSV bronchiolitis. We recently described this finding and showed in addition that depressed lymphocyte proliferative responses and T-cell cytokine responses are markers of disease severity (36).

Overwhelming evidence is available that a Th2-like cytokine response pattern leads to allergic asthmatic airway inflammation. Therefore, one could hypothesize that a Th2-like cytokine response pattern also plays a role in recurrent wheezing after RSV bronchiolitis. For example, CD4+ T-cells, specific for the RSV attachment protein (protein G), secrete IL-4 and IL-5, but little IL-2 (i.e., a Th2-like pattern) when stimulated with antigen (37). This hypothesis is supported by the study of Renzi and colleagues (13) showing an association between a Th2 cytokine response after allergen (Dermatophagoides farinae) stimulation 5 mo after hospitalization for bronchiolitis, and subsequent wheezing in 26 infants. In contrast to our study, in the study of Renzi and colleagues, 43% of the patients were negative for RSV, and patients requiring mechanical ventilation or with radiographic evidence for bacterial infection were excluded. Interestingly, in our study we did not find evidence for an association between IFN-/IL-4 ratios and the subsequent development of recurrent wheezing. In the present study, IFN-/IL-4 ratios in both PHA- and CD2 + CD28-stimulated cultures were comparable for infants with and without recurrent wheezing in the follow-up period. We were capable of detecting a lower IFN-/IL-4 ratio in infants with a family history of atopy, which resulted mainly from differences in IFN- responses. These data demonstrate that our methods are suitable to detect biologically significant differences in IFN-/IL-4 ratios. We note, however, that other Th2-like cytokines, including IL-5 and IL-13, have not been measured. When other Th2-like cytokines are used to assess the Th1-Th2 cytokine balance or when other in vitro stimuli are used, it is conceivable that other results can be found with respect to role of the Th1-Th2 cytokine balance in recurrent wheezing after RSV bronchiolitis.

We conclude that monocyte IL-10 production increases during the course of RSV bronchiolitis and that increased IL-10 production is associated with the development of recurrent wheezing and physician-diagnosed asthma. We did not find support for a role of IFN-/IL-4 balances in the development of recurrent wheezing after RSV bronchiolitis. This study indicates that not only allergen-driven Th2 cytokine responses can result in asthmatic symptoms but also virus-induced changes in monocyte cytokine responses can lead to asthmatic symptoms.