Allergic Sensitization Is Crucial for the Suppressive Role of Streptococcus pneumoniae in the Acute Exacerbation of Asthma

Asthma, which is the most common chronic lung disease in children, is characterized by respiratory problems and obstructed airflow induced by airway swelling and narrowing and the production of mucus. Classically, allergic sensitization is one of the most important factors associated with the development, severity, and exacerbation of asthma [1, 2]. Allergic sensitization is often described as immunoglobulin E (IgE)-mediated sensitization to at least one aeroallergen in the environment (e.g., pollen, fungal spores, house dust mites, or animal allergens) that can be recognized by skin prick testing and/or the detection of serum antigen-specific IgE antibodies [3]. It is widely known that the characteristics of inflammation in the airways of asthma patients with allergic sensitization involve the production of T helper 2 (Th2)-associated cytokines such as interleukin (IL)-4, IL-5, and IL-13, affecting the infiltration and activation of eosinophils and the production of IgE, which is called type 2 (T2) inflammation [4]. The recent management of childhood asthma includes a phenotyping step for the stratification of allergic, eosinophilic, and non-T2 phenotypes due to measurement of biomarkers (e.g., eosinophils in blood, fractional exhaled nitric oxide, and serum IgE) [5]. For example, patients with allergic asthma respond better to corticosteroids, while patients with severe asthma respond better to biologics such as omalizumab [6]. Thus, the consideration of the allergic phenotypes of patients with asthma is a critical issue.

To date, several lines of evidence are available to support the suppressive effects of respiratory bacterial pathogens on airway T2 inflammation. Bacterial lysate, inactivated bacterial extracts from respiratory bacteria, is a representative agent that demonstrates the suppressive effect of bacteria. A previous study demonstrated that the use of bacterial lysates reduced both wheezing episodes and asthma exacerbations in children [7]. The immune-modulating effects of bacterial lysate seem to be due to Th1 cell activation and Th2 cell suppression [8].

In this context, our recent study showed that pharyngeal Streptococcus pneumoniae colonization in the acute exacerbation of childhood asthma was associated with a decreased duration of wheezing [9], suggesting the suppressive role of S. pneumoniae in childhood asthma. Therefore, the purpose of this study is to clarify the role of allergic inflammation upon bacterial colonization in the acute exacerbation of childhood asthma in terms of allergic sensitization, pharyngeal bacterial colonization, biomarkers (e.g., serum eosinophil cationic protein (ECP) and cytokines/chemokines) and symptoms in the acute exacerbation of asthma.

This retrospective study enrolled children (n = 53, 32 boys, 21 girls; mean/median age 2.7/2.5 years; range 10 months to 6 years) with an acute exacerbation of asthma who were treated as outpatients and/or inpatients at Tokai University Hospital, Tokai University Oiso Hospital, Hadano Red Cross Hospital, and Gunma Children’s Medical Center between January 1, 2008, and December 31, 2016. Written informed consent was obtained from the parents of the patients. All patients were diagnosed with bronchial asthma according to Japanese guidelines; asthma was confirmed based on the presence of recurrent wheezing and dyspnea on three or more independent occasions and the identification of reversible bronchoconstriction [10].

We collected clinical data on the severity of the exacerbation, duration of wheezing, and complication with pneumonia from the medical chart written by the doctor during the periods from the onset of the exacerbation to the day of improvement of the exacerbation. Information about the general assessment from caregivers (usually parents) and/or patients on the first day of the emergency department visit/hospitalization was recorded in each patient’s medical chart by their pediatrician. The doctor performed a general asthma examination at the patient’s first hospital visit/hospitalization (i.e., confirming acute asthmatic symptoms, physical examination, blood examination, chest radiography, and other examinations). In hospitalized patients, the doctor performed a physical examination (e.g., examined wheezing by chest auscultation) and recorded information in the patient’s medical chart every day. In patients who were managed as outpatients, the doctor confirmed the same details at each hospital visit. Wheezing was defined as the detection of obvious wheezing on chest auscultation by the pediatrician, and the duration of wheezing was defined as the period from the confirmation of wheezing at the first hospital visit/hospitalization to the disappearance of wheezing confirmed by the same doctor. Acute asthma was diagnosed by physicians in the emergency department based on evidence of wheezing and dyspnea. A mild exacerbation was defined as mild wheezing, stable disease, no dyspnea, and a percutaneous oxygen saturation (S_pO₂) of ≥96%. Moderate exacerbation was defined as wheezing and dyspnea, apparent retraction, and S_pO₂ of 92–95%. Severe exacerbation was defined as severe wheezing and dyspnea with S_pO₂ of ≤91% [10].

The exclusion criteria were congenital heart disease, chronic lung disease, or evidence of a foreign body. These criteria were set because these conditions can affect asthma-related outcome measures.

Pharyngeal samples were collected via swabs from the throats of 53 patients during an acute exacerbation of asthma who were not under antibiotic treatment at the time of the examination. The samples were cultured immediately at the hospitals, and the microorganisms were identified by examining the morphology of the colonies using Gram staining techniques. Bacterial isolates, such as S. pneumoniae, Moraxella catarrhalis, and Haemophilus influenza, were identified by standard microbiological protocols [11]. Nasal aspirates were also obtained from the patients and analyzed via antigen detection kits for respiratory syncytial (RS) virus (Becton Dickinson, Fukushima, Japan), influenza virus types A and B (Denka-Seiken, Gosen, Japan), adenovirus (Tauns, Izunokuni, Japan), and human metapneumovirus (Meiji Seika Pharma, Uji, Japan). The remaining secretions were frozen at −80°C until reverse transcription-polymerase chain reaction (RT-PCR) and direct DNA sequencing were performed, as described previously [12]. PCR products were analyzed by automatic electrophoresis (MCE-202 MultiNA; Shimadzu, Kyoto, Japan) [13].

Peripheral white blood cell (WBC) counts, including neutrophils and eosinophils, and C-reactive protein (CRP) levels were measured in samples from patients with acute asthma who were not under systemic corticosteroid treatment at the time of the examination. An automated fluoroenzyme immunoassay (FEIAUnicap® 100; Phadia AB, Uppsala, Sweden) was used to measure serum levels of total and allergen-specific IgE against inhalant allergens, including Dermatophagoides farinae/pteronyssinus, house dust, Japanese cedar, orchard grass, ragweed, cat and dog dander, Alternaria, and Cladosporium, as described previously [14]. Allergen-specific IgE values ≥0.35 U_A/mL (Class I) were considered positive, and the sensitivity in the detection of total IgE was 2.00 IU/mL. Allergic sensitization was defined as having at least one positive value for aeroallergen-specific IgE.

Serum ECP and 17 cytokines/chemokines (IL-1β, IL-2, IL-4, IL-5, IL-6, IL-7, IL-8, IL-10, IL-12, IL-13, IL-17, interferon [IFN]-γ, tumor necrosis factor [TNF]-α, granulocyte-macrophage colony-stimulating factor [GM-CSF], granulocyte colony-stimulating factor [G-CSF], monocyte chemoattractant protein [MCP]-1, and macrophage inflammatory protein [MIP]-1β) were measured in samples from asthma patients who were not under systemic corticosteroid treatment at the time of examination. A multi-cytokine detection system, Bio-Plex (Bio-Rad, Hercules, CA), was used to quantify serum cytokines and chemokines, according to the manufacturer’s instructions, with a Luminex System (Luminex Corporation, Austin, TX), with calculations performed using the Bio-Plex software program (Bio-Rad), as described previously [12]. The sample values of serum ECP and cytokines/chemokines were excluded from the analysis if those values fell outside the standard curve and/or the detection limit.

The characteristics of the patients were evaluated using the Pearson χ² test or Fisher’s exact test for categorical variables. Unpaired data were analyzed by the Mann-Whitney U test. Correlation coefficients were calculated by Spearman’s rank correlation coefficient analysis. p values of <0.05 (two-sided) were considered statistically significant. All analyses were performed using the SPSS software program for Windows (version 26.0; SPSS Japan Inc., Tokyo, Japan).

The patient characteristics are shown in Table 1. The severity of the exacerbation of asthma was moderate in most cases (88.6%). Allergic sensitization was recognized in 33 (62.2%) patients with an acute exacerbation of asthma. When the clinical characteristics were compared between patients with allergic sensitization and those without it, the patients with allergic sensitization were significantly older than those without it (3.4 ± 1.5 vs. 1.7 ± 0.6 years, p < 0.001). No significant differences in the exacerbation severity or complication of pneumonia were noted among patients with or without allergic sensitization.

Among the 53 pharyngeal samples from patients, the following bacteria were detected: S. pneumoniae (n = 25; 47.1%), M. catarrhalis (n = 6; 11.3%), H. influenzae (n = 6; 11.3%), and other bacteria (n = 16; 30.1%) (Fig. 1a). No significant differences were found in the detection of these 3 bacteria among patients with or without allergic sensitization (Fig. 1b, c), while no significant differences were observed in allergic sensitization among patients with or without these 3 bacteria. In addition, the detection of these bacterial pathogens was not altered regardless of the severity of the exacerbation (data not shown). Among patients with an acute exacerbation, two major viruses were detected: rhinovirus (n = 9; 16.9%) and RS virus (n = 10; 18.8%). There were no significant associations in the detection rates of bacteria or viruses.

The duration of wheezing was compared among patients with and without S. pneumoniae, M. catarrhalis, H. influenzae, rhinovirus, and RS virus. No significant differences were found in the duration of wheezing, with the exception of patients with S. pneumoniae (Fig. 2). The patients with S. pneumoniae had a significantly shorter duration of wheezing than those without it (4.7 ± 3.6 vs. 7.1 ± 3.5 days, p = 0.024) (Fig. 2a). In addition, limited to patients with allergic sensitization, the patients with S. pneumoniae had a significantly shorter duration of wheezing than those without it (4.0 ± 3.6 vs. 7.7 ± 4.0 days, p = 0.003) (Fig. 2f), while there were no significant differences in the duration of wheezing in patients with M. catarrhalis or H. influenzae (Fig. 2f, g). In contrast, limited to patients without allergic sensitization, no significant differences were found in patients with or without S. pneumoniae or M. catarrhalis (Fig. 2k, l). Because of the small number of patients with H. influenzae, this could not be included in the analysis. The detection of rhinovirus or RS virus did not affect the duration of wheezing, regardless of allergic sensitization (Fig. 2d, e, i, j, m, n).

We next analyzed the correlation between the duration of wheezing and peripheral WBC counts, neutrophils, eosinophils, and CRP levels. In all patients, the duration of wheezing showed a significant inverse correlation with the WBC count, neutrophil count, and CRP levels (Fig. 3). Conversely, in patients with or without allergic sensitization, no significant correlations with WBC counts, neutrophils, eosinophils, or CRP levels were recognized, while the duration of wheezing tended to be inversely correlated with these parameters (Fig. 3).

We evaluated the differences in allergic sensitization-specific values for serum total IgE, ECP, and 17 cytokines/chemokines. Serum total IgE and ECP was significantly higher in patients with allergic sensitization than in those without it (346 [63.9–1,840] vs. 32.7 [4.4–212] IU/mL, p < 0.001, 33.1 [2–109] vs. 7.8 [3–35] ng/mL, p = 0.042, respectively) (Table 2). Among 17 cytokines/chemokines, only IFN-γ was significantly lower in patients with allergic sensitization than in those without it (5.6 [4–10] vs. 16.4 [7–28] pg/mL, p = 0.032) (Table 2).

Considering the reduced duration of wheezing in allergic patients with S. pneumoniae colonization, we performed an analysis of the association between serum total IgE, ECP, IFN-γ, and S. pneumoniae colonization. No significant differences were found in these parameters among patients with or without S. pneumoniae colonization (data not shown), except for lower serum total IgE in patients with S. pneumoniae than those without (81.9 [7.8–894] vs. 287 [4.4–1,840] IU/mL, p = 0.014).

Finally, we evaluated the correlation between serum ECP or IFN-γ and the eosinophil count or serum total IgE as biomarkers of T2 inflammation. In all patients, serum ECP was significantly correlated with serum total IgE (Fig. 4b), whereas IFN-γ showed a significant inverse correlation with serum total IgE (Fig. 4d). Conversely, neither serum ECP nor IFN-γ were correlated with the eosinophil count (Fig. 4a,c).

In this study, we found that among patients with an acute exacerbation of childhood asthma, pharyngeal S. pneumoniae colonization induced a decreased wheezing duration in patients with allergic sensitization but not in patients without allergic sensitization. In addition, serum total IgE was significantly lower in patients with S. pneumoniae than in those without. This result indicates that allergic sensitization is a critical factor for the suppressive role of pharyngeal S. pneumoniae colonization, which we have previously reported [9]. Furthermore, higher serum ECP and lower IFN-γ were significantly associated with allergic sensitization in acute asthmatic children. However, the levels of serum ECP and IFN-γ between the patients with or without S. pneumoniae had no significant differences. There are a couple of possible explanations for these observations. First, the suppressive effects of S. pneumoniae may be observed in the airway samples rather than serum samples. Second, total IgE may also be associated with the mechanisms of suppressive roles of S. pneumoniae. In fact, the past study in a mouse model revealed that infection with S. pneumoniae decreased serum IgE levels and reduced eosinophil counts in bronchoalveolar fluid [15]. Collectively, total IgE rather than type 2 immune responses and/or eosinophilic inflammation may have an important role in the suppressive effects of S. pneumoniae colonization on acute exacerbation of asthma. To the best of our knowledge, this is the first study to examine the association between allergic sensitization and bacterial colonization and potential mechanisms in children with the acute exacerbation of asthma.

Although several studies have reported that S. pneumoniae plays a role in the suppression of asthma, few studies have reported the potential mechanism. Studies in animal models have underscored that the intratracheal administration of whole killed S. pneumoniae during or after sensitization to ovalbumin significantly reduces the development of eosinophilia in the blood and the subsequent recruitment of eosinophils into bronchoalveolar fluid (BALF), resulting in the improvement of dynamic lung compliance and peripheral airflow [16, 17]. Further studies demonstrated that S. pneumoniae components (e.g., type-3-polysaccharide, pneumolysoid, and aminopeptidase N) suppress OVA-induced T2 airway inflammation, including eosinophilia, mucus hypersecretion, Th2 cytokine production, and IgE production [18, 19]. It has also been shown that the pneumococcal conjugate vaccine suppressed the development of allergic airway diseases [20]. These suppressive effects on asthma have been shown to involve natural killer T cells, regulatory T cells, Toll-like receptors, and the adapter protein myeloid differentiation primary response gene 88 [17, 18, 20‒22], while elucidation of the mechanism in greater has remained a challenge. Our results, which demonstrated that patients with S. pneumoniae colonization and allergic sensitization had a shorter duration of wheezing, were consistent with the past observation that S. pneumoniae especially suppresses T2 inflammation.

In the present study, we showed that the production of ECP in serum was significantly higher in patients with allergic sensitization than in those without. ECP is one of the granular proteins derived from eosinophils and released from activated eosinophils. ECP has been widely investigated as a potential marker of asthma because the secretory activity of eosinophil granulocytes is a crucial component in the pathophysiology of asthma. Several studies demonstrated that higher plasma ECP levels were associated with higher blood eosinophils, aeroallergen sensitization, IgE, and exhaled nitric oxide concentrations and were further associated with poorer asthma control, greater airflow limitation, bronchial hyper-responsiveness, and more frequent exacerbations in asthma patients, suggesting that ECP may be a useful marker of T2 inflammation [12, 23‒25]. Moreover, patients with higher ECP who received systemic corticosteroids showed a greater improvement in asthma symptoms [23]. Based on these findings, higher serum ECP levels – as marker of T2 inflammation – in patients with allergic sensitization may indicate an association between the effect of S. pneumoniae colonization and T2 inflammation.

We also found that levels of serum IFN-γ were significantly lower in patients with allergic sensitization than in those without it. It is well known that Th1/Th2 imbalance describes a simplistic model of the involvement of T cells in asthma, and decreased Th1 cytokine IFN-γ has been suspected to play a prominent role in asthma pathophysiology. A past study revealed that serum levels of IFN-γ in asthmatic children were significantly lower than those in healthy controls, irrespective of corticosteroid treatment [26]. Another study showed that there was a significant relationship between the serum levels of IFN-γ and the severity of atopic asthma (79.8 ± 3.4 in mild atopic asthma, 70.1 ± 2.3 in moderate, and 59.1 ± 3.7 in severe, pg/mL, p < 0.05) [27]. In addition, several longitudinal studies involving in vitro experiments investigated the production of IFN-γ in children with asthma [28] or allergic sensitization [29]. Asthmatic children without inhaled or systemic corticosteroid and allergen-specific immunotherapy showed lower production of IFN-γ by phytohemagglutinin (PHA)-stimulated peripheral blood mononuclear cells (PBMCs), which was correlated with the deterioration of the pulmonary function (e.g., forced expiratory volume in 1 s) [28]. Moreover, a 2-year observational study revealed that children who remained in a state of allergic monosensitization showed significantly higher production of IFN-γ by PHA-stimulated PBMCs than those who developed additional allergic sensitization, that is, polysensitization [29]. Taken together, decreased Th1 responses, including IFN-γ production, may promote the typical allergic shift to Th2 responses, resulting in asthma-related airway changes and the development of allergic sensitization.

Of particular interest in the present study is that the association among lower serum IFN-γ (as a decreased Th1 response), higher serum ECP (as an increased Th2 response), and the suppressive effect of pharyngeal S. pneumoniae colonization on wheezing and serum total IgE may indicate potential mechanisms through which Th1/Th2 balance is involved in asthma and wherein S. pneumoniae colonization exerts suppressive role in the acute exacerbation of childhood asthma. It is also noteworthy that serum ECP and IFN-γ were significantly correlated with total IgE, but not peripheral blood eosinophils, which might indicate that the ECP/IFN-γ/total IgE axis is a potential marker for childhood asthma patients with a predominant Th2 response.

This study was associated with some limitations, including its retrospective observational design. Prospective studies are needed to clarify the precise mechanism through which pharyngeal S. pneumoniae colonization brings about a shorter duration of wheezing in patients with the acute exacerbation of allergic asthma. Further studies are warranted to examine our findings in a larger population and to elucidate the molecular mechanism through which S. pneumoniae colonization influences T2 inflammation.

The generalizability of this study is limited by the patient characteristics. Because the present study population consisted of patients of 10 months to 6 years of age with an acute exacerbation of asthma, it may be possible to generalize our findings to other children of the same age group. However, the small number of patients with a mild or severe exacerbation of asthma resulted in limiting the findings of this study to exacerbations of all degrees of severity. Moreover, the present study only enrolled patients with an acute exacerbation of asthma and not those with stable asthma. Based on these findings, our results may not apply to older patients, patients with mild or severe exacerbations, or those with stable asthma.

In conclusion, our data raise the important possibility that the suppressive effects of S. pneumoniae colonization is observed only in patients with allergic sensitization, wherein serum total IgE, ECP, and IFN-γ may have an important role on acute exacerbation of asthma. This novel insight may be useful for clinical practice to stratify the phenotypes of childhood asthma, resulting in precise medicine according to the state of T2 inflammation and bacterial colonization.

We thank Taeko Miyake of the Gunma Children’s Medical Center for their excellent technical assistance.

This study was approved by the Ethics Committee of Tokai University Hospital (14R-128), Tokai University Oiso Hospital (14R-205), Hadano Red Cross Hospital (28-01), and Gunma Children’s Medical Center (1,105). Written informed consent was obtained from the parents/legal guardians for all participants aged under 18 years old and assent was obtained from children considered old enough (generally >9 years old).

The authors declare no conflicts of interest in association with the present study.

This work did not receive any specific grants.

Yuichi Kama and Masahiko Kato designed the study, collected samples, performed the data analysis, wrote the first draft, and finalized the manuscript. Yoshiyuki Yamada, Takashi Koike, Mayumi Enseki, Kota Hirai, and Hiroyuki Mochizuki collected samples and performed the data analysis. Yuichi Kama and Masahiko Kato interpreted the results. All the authors read and approved the final manuscript.

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.

Edited by: H.-U. Simon, Bern.