Background
The incidence of childhood allergic diseases has exhibited a marked increase in recent years, with the incidence in China being greater than 3.3 % [
1]. A family history of allergic diseases, especially maternal allergies, is associated with a high likelihood of childhood allergy development [
2,
3]. Some prospective cohort studies of Western countries have shown that susceptibility to allergic diseases in children may occur early in life [
4‐
6], and during this time period regulatory T-cells (Tregs) could be involved. However, analogous follow-up studies in China have not been performed.
Because Tregs influence the early steps of allergy initiation [
7], it is important to measure Treg cell abundance. New research has found that CD4
+CD25
+ FOXP3
+ T cells were found to be excellent markers for Tregs [
8]. Stable expression of FOXP3 has been observed in mature Tregs, and this expression profile was found to be highly specific for Tregs [
9]. Other Treg specific genes include glucocorticoid-induced TNFR-related protein (GITR) and lymphocyte activation gene 3 (LAG-3) [
10,
11]. Furthermore, interleukin (IL)-10 is a relatively specific cytokine of Tregs, and secretion of this cytokine may act as an additional marker for Tregs [
12]. The suppressive capacity of Tregs is required to maintain the balance of T helper (Th) cells, including Th1, Th2, and Th17 cells. Unchecked proliferation of these Th cell populations has been shown to play a key role in the pathogenesis of allergic diseases. Toll-like-receptor (TLR)-induced innate immune responses affect the early differentiation of T cells. In particular, TLR2 and TLR4 are induced by peptidoglycans (PPG) and lipid A (LPA), respectively, and have been shown to be important mediators of T cell differentiation. Immune system stimulation may be critical to early immune modulation, and several studies have reported that young children growing up in environments rich in microbial stimuli have a lower risk of allergy [
13,
14]. At the molecular level, activation of the TLR2/4 pathways has been shown to upregulate Tregs and induce cytokine secretion in both children and animal models with allergy development [
13‐
15].
We hypothesize that the extent of the T cell response to TLR2/4 stimulation may be influenced by a maternal background of atopy. Therefore, the aim of the present study was to examine whether TLR2/4 stimulation influences immunity in addition to the activity of T cell subsets such as Tregs. Secretion of cytokines IFN-γ (Th1 cells), IL-13 (Th2 cells), and IL-17 (Th17 cells) were also examined. Furthermore, follow-up studies were conducted to assess whether maternal allergic status affected the susceptibility of developing allergic diseases in the first three years of childhood.
Methods
Study populations
A total of 332 mothers of healthy newborns in the city of Changchun, China were recruited to this study. Of these mothers, 234 had no allergies, while 98 had been diagnosed with allergies. We defined maternal allergy history as a diagnosis of asthma [
16], allergic rhinitis [
17], and/or allergic eczema [
18] with at least one positive specific IgE (sIgE). Enrollment occurred from January 2012 through December 2012 with the approval of the ethics committee of the Second Hospital of Jilin University (2012–003). Pregnant mothers were approached at least one month before delivery for consent and they completed a detailed questionnaire that assessed maternal and infant data. In addition, cord blood was collected from each neonate (
n = 332). The study inclusion criteria consisted of healthy neonates and mothers with uncomplicated pregnancies. Exclusion criteria were preterm deliveries, parents with autoimmune diseases, perinatal infections, maternal antibiotic or systemic glucocorticoid use in the last trimester, rural newborns, unhealthy newborns, or paternal allergic diseases.
Follow-up studies were conducted from 2012 to 2015 by questionnaire when the children were 12, 24, and 36 months old. Follow-up questionnaires were only completed for 273 children. The population characteristics that were examined for the children included: vaccination status, pet exposure, air conditioner exposure, breast feeding, diet, frequency of viral respiratory infections, medications, household income, parental education, and cohabitation with a smoker. Allergic wheezing (as a proxy for asthma in children ≤ 5 years old [
19]), allergic rhinitis, and allergic eczema were diagnosed by paediatricians based on clinical manifestations, the presence of at least one sIgE+, and parent-described patient symptoms. The cohort belonged to a low economic status group (≤20,000 Renminbi income/year) based on available data from the Statistics Department (
http://www.stats.gov.cn/) on the Per Capita Disposable Income (PCDI) for the Jilin Province collected from 2012–2015.
Dust collection and analysis
To evaluate the presence of environmental allergens, dust samples were collected from the living rooms of the pregnant women one month before delivery. Dust samples were subsequently collected from the children’s bedrooms at 12 ± 1, 24 ± 1, and 36 ± 1 months post delivery. All dust samples were collected on pre-weighed glass fibre filters. Collection was performed using vacuum cleaners with sampling nozzles (ALK, Horsholm, Denmark) according to a standardized protocol with photo and video instructions. Collected samples were stored at −20 °C until extraction. Dust-containing filters were weighed and extracted in a volume of 5–40 ml, as determined by the net dust weight (<0.5 g, 5 ml; 0.5–1.0 g, 10 ml; 1.0–2.0 g, 20 ml; > 2.0 g, 40 ml). Endotoxin levels were measured with a kinetic chromogenic Limulus amebocyte lysate (LAL) test and glucan levels were detected in an inhibition enzyme immunoassay an inhibition EIA [
20].
IgE detection
Total IgE and sIgE were measured in peripheral blood serum samples collected from both mothers and children using the RAST method. Levels of sIgE in serum against six foods (cow’s milk, hen’s egg, hazelnut, carrot, peanut, and wheat flour) and 13 inhalant allergens (Dermatophagoides farinae; Dermatophagoides pteronyssius; cat fur; dog fur; horse fur; Alternaria species; mugwort, plantain, birch, alder, hazel, and rye pollen; and grass pollen mix) were assayed. A positive sIgE was defined as one or more positive reactions ≥0.35 IU/mL to a panel of 19 common allergens.
Lymphocyte stimulation and cytokine production
Cord blood (20 ml) and peripheral blood (5–10 ml) samples were collected from healthy neonates within 24 h of birth. Cord blood mononuclear cells (CBMCs) were isolated from cord blood by density gradient centrifugation with Ficoll-Paque PLUS (GE Healthcare, Piscataway, NJ, USA) after dilution in PBS (Gibco, Carlsbad, CA, USA). Cells were washed in RPMI 1640 (Gibco) and diluted in 10 % human serum (Gibco) to a concentration of 5 × 106 cells/ml. CBMCs were cultured for three days in the presence of 10 μg/ml PPG, 0.1 μg/ml LPA (Sigma-Aldrich, St. Louis, MO, USA), or without a stimulation factor. Concentrations of IL-10, IL-13, IL-17, and interferon-γ (IFN-γ) in the CBMC supernatants were measured with a Human Cytokine Multiplex Assay kit (Bio-Rad, Munich, Germany) with Luminex detection. The lower limits of detection for the assay (pg/mL) were: 0.9 (IL-10), 2.1 (IL-13), 0.2 (IL-17), and 1.3 (IFN-γ).
Flow cytometry
CBMCs were analysed by flow cytometry (BD Biosciences, San Jose, CA, USA) after three days of cultivation. For surface staining, 2 μl of anti-human CD4-fluorescein isothiocyanate (FITC), 1 μl of CD25-RPE-Cy5, 1 μl of IgG1-FITC (DakoCytomation, Glostrup, Denmark), and 0.5 μl of IgG2a RPE-Cy5 (BD Biosciences) were added. For CD4/CD25/FOXP3 co-staining, 8 μl of anti-human CD4-FITC and 4 μl of anti-human CD25-RPE-Cy5 antibodies were added to 1 × 106 cells in 100 ml PBS. Cells were then permeabilized and FOXP3-PE or the corresponding isotype control antibodies were added. Data were analysed with CellQuest software (BD Biosciences).
Real-time quantitative RT-PCR
CBMCs were collected after three days of cultivation and total RNA was isolated immediately with the TriPure Isolation Reagent (Roche, Mannheim, Germany). RNA concentration was determined with a NanoDrop2000 (Thermo Scientific, Waltham, MA, USA), and samples were stored at −20 °C. Reverse transcriptase were performed (Invitrogen, Karlsruhe, Germany) for cDNA synthesis. Specific primer pairs were designed with Primer Express software (Vector NTI Advance10). Quantitative real-time PCR was performed with an iCycleriQ-multicolor Real-Time PCR Detection System. SYBR Green (Applied Biosystems, Darmstadt, Germany) was used to detect double-stranded PCR products. The CT means the number of PCR cycles which are required for the fluorescence signal to exceed the detection threshold value. The threshold cycle (CT) of each target product is associated with the amplification plot of 18S. The level of mRNA from individual gene is described as gene expression. The fold difference in gene expression was calculated based on the difference in normalized CT values between the stimulation and non-stimulation groups. The relative quantitative results showed changes of gene expressions in stimulated samples when compared with those in unstimulated ones.
Functional analysis of tregs
CD3− T cells were isolated (CD3− Cell Isolation kit, MiltenyiBiotec, Köln, Germany) and irradiated as antigen-presenting cells. CD3− T cell purity reached 98 %. CD4+CD25+ T cells were isolated from CBMCs by tandem positive selection (MiltenyiBiotec). CD4+CD25− T cells were collected from the flow through during CD25 selection. The purity of both the CD4+CD25− and CD4+CD25+ T cell populations was > 95 %. CD4+CD25− T cells (effector T cells) were labelled with 5 μmol/L CFSE (CellTrace™ CFSE Cell Proliferation Kit, Invitrogen), and were subsequently incubated for three days with irradiated CD3− cells in co-culture with or without CD4+CD25+ T cells (Tregs). The co-cultures were stimulated with phytohaemagglutinin (0.8 μg/ml). CFSE-stained CD4+CD25− T cells were isolated by flow cytometry and 3H-thymidine incorporation was used to measure their proliferation. Concentrations of cytokines IL-13, IL-17, and IFN-γ were measured in the cell culture supernatants using a Human Cytokine Multiplex Assay Kit. The resulting values were compared between the effector T cells only group and the effector T cells plus Tregs group. The suppressive capacity of the Tregs on the effector T cells was further compared between the maternal allergy and non-allergy groups.
Statistical analysis
Data were analyzed using SigmaStat1.0 software. Differences in cell counts, gene expression, suppressive function, and cytokine production for the Tregs present in the neonate cord blood samples were compared between maternal allergy and non-allergy groups, and between children’s allergy and non-allergy groups. Correlations between newborn Treg function and the development of childhood allergy were analysed. Metric data were evaluated by either a t-test or the Mann–Whitney rank test with or without normal distribution. Data are reported as the mean ± standard error of the mean (SEM) or the median with a 25–75 % interquartile range (IQR), depending on their distribution. Attributes data were analyzed with a X
2
test or Fisher’s exact test. Multiple logistic regression was performed for the correlation analyses. Odds ratio (OR) values were reported when applicable. Statistical significance was defined as a P-value < 0.05.
Discussion
In this birth cohort study, Tregs in the cord blood samples obtained from neonates with allergic mothers were fewer in number. The Tregs also exhibited compromised suppressive function, lower levels of expression for associated genes, and decreased IL-10 production. Moreover, the PPG-TLR2 innate immune response pathway was particularly affected. As a result, an increased susceptibility to allergies in early childhood (0–3 years old) was observed. Maternal allergic disease history and neonatal impairment of Tregs were both identified as risk factors for susceptibility to allergies in the children examined, independent of social and environmental factors for the children and their families. These results are consistent with those of previous studies of allergies in European populations [
4,
5].
Many studies have reinforced the importance of in utero exposure for fetal immune development and the programming of susceptibility to asthma and allergic diseases [
21]. In the present study, these factors were adequately considered to avoid potential interference. In addition, based on the results of a study by Schaub et al. where a higher percentage of allergic neonates with allergic mothers were more likely to have allergic fathers [
5], paternal allergies were established as an exclusion criterion for the present study in order to focus on the effect of maternal allergic disease history on offspring. The study by Schaub et al. also reported increased birth weights and lengths for the newborns with allergic mothers [
5]. However, no differences in birth weight or length between the groups of the present study were observed. This may be due to the similarity in the diets and medication use of the mothers included in the present cohort. Over the course of child development, many social and environmental factors should be considered when assessing influences on allergy development. Therefore, factors which have been shown to influence allergic disease development in children were considered in the present study, including: environmental exposure, diet, frequency of viral respiratory infections, household income, parental education, and co-habitation with a smoker [
22]. It was observed that the endotoxin content in the living environments of the children with allergies was lower than in the living environments of the children without allergies. These results are consistent with those from Karvonen et al. where endotoxin exposure in early life was found to be closely related to the development of allergic diseases in children [
23]. Consequently, microbial content was adjusted for in the analyses of risk factors for children’s allergic status.
As shown in Fig.
1a,
b and
c, FOXP3
+ T cells represented > 90 % of the unstimulated CD4
+CD25
+ T cells. Therefore, CD4
+CD25
+ T cells were isolated from the unstimulated CBMCs to serve as a Treg population that would have its suppressive function examined in vitro. There were fewer CD4
+CD25
+FOXP3
+ T cells present in both the neonates with allergic mothers and the neonates who developed allergies in early childhood. While the expression levels of GITR, LAG3, and FOXP3 were previously shown to positively correlate with each other [
5], the specificity and sensitivity of their expression slightly differed in the TLR-mediated innate immune responses that were induced. It has been shown that Tregs secreting IL-10 play an important role in allergic diseases, independent of CD25
+ Tregs [
12]. Therefore, IL-10 production was assayed as a readout of function for this Treg subgroup. However, IL-10 is also secreted from Th2 cells. In the present data, PPG-stimulation reduced IL-10 secretion in the newborns with allergic mothers, while secretion of the Th2-specific cytokine, IL-13, increased (Table
3). These results suggest that IL-10 production is characteristic of this Treg subgroup in the TLR-mediated innate immune response. Overall, the risk factors for allergies in children of the present cohort included a reduced percentage of CD4
+CD25
+FOXP3
+ T cells, reduced
FOXP3 expression, and PPG-stimulated IL-10 secretion (Tables
6,
7 and
8). However, the other cytokines assayed, IL-13, INF-γ, and IL-17, were not identified as risk factors (data not shown). To date, the traditional Th1/Th2 imbalance theory has not been sufficient to explain the complex pathogenesis for allergic diseases. In addition, Tregs, Th17 cells, Th9 cells, and other subgroups of T cells have all been implicated in allergic disease pathogenesis. Based on the importance of Tregs maintaining their suppressive function, we hypothesize that impairment of Tregs in neonates with allergic mothers leads to dysregulation of the differentiation of Th0 cells into Th1, Th2, and Th17 cells, and this is the driving force for allergy development in children.
When the children of this study were divided into those with or without allergies, regardless of maternal allergy status, down-regulation of Tregs according to the TLR2/4 pathway at the newborn stage partly predicted the children’s allergic status. However, it was not clear whether the down-regulation of Tregs derived exclusively from maternal allergy history. By using multiple logistic regression, exclusion of maternal allergic disease history and environmental factors led to the identification of a reduced number of CD4
+CD25
+FOXP3
+ T cells, reduced
FOXP3 gene expression, and reduced IL-10 secretion in neonates as independent risk factors for the allergic status of children (Tables
6,
7, and
8).
It was previously demonstrated that impairment of Tregs in neonates with allergic mothers was more profound in the TLR2/4-mediated innate immune response pathways, due in part to genetic variations in the TLRs present in cord blood [
24]. Genetic-immunological interactions of the TLR pathway influence Tregs early in life and are modulated by maternal allergic disease history. Moreover, it is likely these interactions are relevant for immune maturation in the development of allergic diseases in childhood. In addition to a genetic component, the time, frequency, and intensity of activation of TLR-mediated immune pathways are important for the maturation and differentiation of Tregs and other Th cells. For instance, increased microorganism exposure in early life in children and in mouse models has been shown to regulate the immune system by enhancing the Th1 pathway and weakening the Th2 pathway through TLR2/4 immune responses [
25‐
27].
The results of the present study have revealed that maternal allergy status has a significant impact on Treg differentiation in the TLR2/4 immune pathway, yet not in the unstimulated pathway. The present results also confirm that the occurrence of allergic disease is the result of interactions between genetic and environmental factors. In particular, the neonatal period appears to be a key immune maturation period, and environmental endotoxin levels during this period can influence the extent of differentiation that takes place to generate different T cell subgroups in children via the TLR2/4 pathway. The number of Tregs present in fetal umbilical cord blood was observed to increase following PPG and LPA stimulation, thereby facilitating the differentiation of naive T cells into new Tregs. Maternal allergy history can influence modulation of differentiation and inhibit the differentiation rate of Tregs through the interactions of other T cell subgroups.
Thorburn et al. [
28] showed that high fibre or acetate diets during pregnancy, which influence gut microbiota, can lead to marked suppression of allergic airway diseases in offspring by enhancing Treg numbers and function. These results are consistent with the observation that the intake of probiotics during pregnancy and early life have the potential to prevent childhood eczema [
29] and IgE-associated allergies to some extent [
30]. However, avoiding activation of the TLR-mediated immune pathway appears safer than a desensitization approach for preventing allergic disease development in the offspring of allergic pregnant women since the latter can up-regulate CD4
+CD25
+ T cells and IL-10 production in offspring [
31,
32]. In the risk factor analysis performed in the present study, a decrease in PPG-stimulated Tregs was identified as a risk factor for children’s allergic status, allergic rhinitis, and allergic eczema. Meanwhile, a decrease in unstimulated Tregs was only a risk factor for allergic wheezing. Thus, micro-stimulation of the TLR2 and TLR4 pathways with adjustment of diet and environmental exposure may be an effective early intervention for preventing allergic disease in neonates that have lower numbers of Tregs or that have a maternal allergic disease history by establishing a more beneficial Th1/Treg balance.
Acknowledgements
Not applicable.