The association between maternal psychological stress and inflammatory cytokines in allergic young children

View article


The incidence of childhood allergic disease has increased recently throughout the world, particularly in westernized countries. Psychological stress, if sustained, can adversely affect critical functions, such as immune surveillance (Glaser, 2005), gastrointestinal integrity (Meddings & Swain, 2000) and compromise host defenses against viral infections (Steelman et al., 2009). Previous studies have documented a relationship between psychological stress and allergic diseases at all ages. Sandberg et al. (2000) prospectively examined the temporal relationship between stressful experiences and asthma in children, and showed that stressful life events increase the risk of a new asthma attack in the coming few weeks. Severe stress is associated with higher risk of self-reported asthma incidence in adults (Rod et al., 2012). Furthermore, psychological stress affects additional allergic diseases, including allergic rhinitis and food allergies (Rod et al., 2012; Heyman, 2005; Lebovidge et al., 2009).

Because young children have close ties with their parents, the psychological condition of parents may strongly influence their children. Wolf et al. (2008) reported that higher levels of perceived stress and depression in parents are associated with increases in IL-4 production and eosinophil cationic protein release in children. This pattern was observed in both asthmatic children and healthy children (Wolf, Miller & Chen, 2008). However, there have been no studies reporting a relationship between parental psychological stress and other biomarkers related to allergic conditions in children.

For children, inhalation and oral exposures are the main exposure routes of allergens, and this study focuses on them. Inhalation allergens, especially house dust mite (HDM), are mainly related to asthma and allergic rhinitis (Gandhi et al., 2013), and recent studies have reported that IL-22, predominantly secreted by T helper type 17 (Th17) cells, is highly expressed in various infectious and inflammatory diseases, especially childhood allergic rhinitis and asthma (Farfariello et al., 2011; Tsuji et al., 2012a). Polycyclic aromatic hydrocarbons (PAHs) are one of the causes of asthma related to air pollution, and Ple et al. (2015) recently found that PAH-induced IL-22 level in asthmatic patients is significantly higher than that in healthy subjects. IL-6 and IL-8 are pro-inflammatory cytokines, and many studies have confirmed increased IL-6 and IL-8 expressions in patients with asthma and allergic rhinitis (Shi et al., 2010; Tsuji et al., 2012b).

On the other hand, oral exposure is important, and a disruption of the intestinal barrier leads to food allergy (Jeon et al., 2013). Intestinal epithelial toll-like receptor (TLR) activation prevents food allergic responses, and the activated TLR increases IL-10/IL-13 expressions and suppresses IL-6 and IL-8 expressions (De Kivit et al., 2014). The deletion/anergy of reactive T cells against specific antigens are associated with the expansion of regulatory T cell population, and IL-10 production is mainly involved in oral tolerance (Crittenden & Bennett, 2005). Thus, in consideration of the main exposure routes of allergens among children, IL-6, IL-8, IL-10 and IL-22 were considered as proper biomarkers in this study.

For the evaluation of maternal psychological status, the Japanese Perceived Stress Scale (JPSS) and the Centre for Epidemiological Studies Depression Scale (CES-D) were used. The JPSS is a standardized self-reported questionnaire of globally perceived stress (Cohen, Kamarck & Mermelstein, 1983). The questions address how stressful, overwhelming, and uncontrollable a person has found his/her life during the last month. The CES-D is a short, self-reporting scale designed to measure depressive symptoms in the general population and clinical populations. The 20 symptom-related items of the CES-D examine how often one experienced depressive cognitions, affect, and behaviors during the past one week (Radloff, 1977).

The primary objective of this study is to test the usefulness of specific inflammatory biomarkers in the analysis of the potential relationship between diverse childhood allergies, including asthma, allergic rhinitis, and specific IgE reactions to HDM and food, and maternal psychological stress, as assessed using the JPSS and CES-D.

Material and Methods

Study subjects

In this study, children who were under the age of 4 and either of their parents were recruited at one clinic between January 2009 and February 2010 (Tsuji et al., 2012b; Tsuji et al., 2012a). We invited 203 pairs of mothers and children to participate in this study, and 158 of them (78%) agreed to participate. All participants were asked to return to the clinic when the child had been stable for more than one month without any clinical symptoms of allergic diseases. When they returned to the clinic, three questionnaires were administered to the mother. An initial questionnaire was given at the maternal interview, including questions about duration (weeks) of pregnancy, breastfeeding, maternal smoking habits, and allergy histories. Regarding maternal smoking habits, mothers who never smoked near their children were considered as non-smokers (Tsuji et al., 2012b; Tsuji et al., 2012a). After that, maternal psychological status was evaluated using two scales, the JPSS and CES-D, and 157 of them answered both the JPSS and CES-D. The remaining mother answered only the JPSS. Non-fasting venous blood was collected from children who were not infected with any diseases during the month preceding the interview. Children’s sera collected were subjected to the CAP radio-allergosorbent test (CAP-RAST) to examine responses to egg, milk, wheat and HDM. IL-6, IL-8, IL-10 and IL-22 mRNA were analyzed by quantitative reverse-transcriptase PCR (qRT-PCR) to estimate gene expression levels using the children’s blood samples. A blood sample for qRT-PCR was not collected from all of the 158 children. Pediatricians discontinued blood collection because children cried and/or their blood vessels were very thin and it was difficult to obtain sufficient blood volumes. Thus, the total number of study subjects was 152 in this study. The asthma group was defined by children who met at least two out of the three following criteria: (1) diagnosed by their primary physician as having asthma; (2) having experienced shortness of breath and/or wheezing; and (3) having received treatment for asthma or asthmatic bronchitis (Tsuji et al., 2012b; Tsuji et al., 2012a). The allergic rhinitis group was defined by children who met both of the following criteria: (1) diagnosed by an otolaryngologist as having allergic rhinitis; and (2) having experienced paroxysmal iterative sneezing, watery rhinorrhea and stuffy nose.

This study, including the manner in which written informed consent was obtained from the subjects’ guardians, was approved by the review boards of Kumamoto University (genome No 118) and the University of Occupational and Environmental Health (H23-03).

Quantitative real-time PCR (q-RT-PCR) for cytokine mRNA expression

Blood samples were collected into heparin-coated vacuum tubes and kept at 4 °C until aliquots were transferred into RNA preparation solution within 6 h of collection. Total RNA was isolated from 185 µl of the whole blood of children using the QIAamp RNA Blood Mini kit (Qiagen, Hilden, Germany). RNA was reverse transcribed into cDNA using QuantiTect Reverse Transcription (Qiagen, Hilden, Germany) according to the manufacturer’s instructions and stored at −80 °C for qRT-PCR analysis. Quantitative detection of glyceraldehyde-3-phosphate dehydrogenase (GAPDH), IL-6, IL-8, IL-10 and IL-22 was performed using the StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA) and the Fast SYBR Green Master Mix (Applied Biosystems, Foster City, CA, USA) according to the manufacturer’s protocol. GAPDH is one of the most commonly used housekeeping genes in comparisons of gene expression data. PCR amplification was carried out in a total volume of 20 µL containing 2 µL cDNA 10 µL and 0.2 µM of each primer. The PCR cycling conditions were 95°C for 5 min, followed by 55 cycles of 95°C for 10 s and 60°C for 30 s. To confirm the amplification specificity, we subjected all of the PCR products to melting curve analysis. The basic methodology of qRT-PCR was as described previously (Tsuji et al., 2012b; Vogel et al., 2005).

The primers for each gene were designed on the basis of the respective cDNA or mRNA sequences using OLIGO primer analysis software provided by Steve Rozen and the Whitehead Institute/MIT Center for Genome Research. Primers were as follows: for GAPDH: 5′-GAGTCAACGGATTTGGTCGT-3′ (forward), 5′-TTGATTTTGGAGGGATCTCG-3′ (reverse); for IL-6: 5′-GAACTCCTTCTCCACAAGCG-3′ (forward), 5′-TTTTCTGCCAGT GCCTCTTT-3′ (reverse); for IL-8: 5′-CTGCGCCAACACAGAAATTA-3′ (forward), 5′-ATTGCATCTGGCAACCCTAC-3′ (reverse); for IL-10: 5′-TGGGGGAGAACCTGAAGAC-3′ (forward), 5′-CCTTGCTCTTGTTTTCACAGG-3′ (reverse); and for IL-22: 5′-ACAGCAAATCCAGTTCTCCAA-3′ (forward), 5′-TCCAGAGGAATGTGCAAAAG-3′ (reverse).

Allergen-specific IgE assays

Egg-, milk-, wheat-, and HDM-specific immunoglobulin E (IgE) was determined by the CAP-RAST using 0.3 ml of serum per allergen at FALCO Biosystems Ltd., in Japan. A result of 0.35 UA/ml was taken as a sign of sensitization, so 0.35 UA/ml or higher subjects were categorized as positive and lower than 0.35 UA/ml subjects were categorized as negative (Tsuji et al., 2012b).

Statistical methods

The comparisons of CES-D and JPSS scores by the presence of allergic diseases and potential determinants were conducted by the Mann–Whitney U test or Kruskal–Wallis test. Each mRNA level of cytokine was applied in multivariable regression models as a dependent variable after log transformation. In multivariable analysis, age and sex of the study subjects were always included in the models because they are basic characteristics. In addition, the number of siblings was also included in the model based on the results of the association with parental psychological stress. All analyses were performed by STATA Version 10 (Stata Corporation, USA), and all P values presented are two-sided (α = 0.05).


Characteristics of the study subjects are summarized in Table 1. Median ages of children and their mothers were 13 months and 31 years old, respectively. Thirty-three mothers did not provide their age and one mother did not report gestational age at birth in the questionnaires. Three parents could not remember their child’s birth weight. HDM-specific IgE of 5 children could not be measured because of insufficient blood volume.

Table 1:
Characteristics of the study population.
Number/Median (%/25th and 75th percentiles)
ALL 152 (100%)
Boy 83 (55%)
Girl 69 (45%)
Child (month) 13 (7, 26)
Mother (year)a 31 (29, 34)
Gestational age (week) a 39 (38, 40)
Birth weight (kg) a 3.0 (2.7, 3.3)
Breast 66 (43%)
Mix (breast + milk) 68 (45%)
Milk 18 (12%)
Number of siblings
None 83 (55%)
1 54 (36%)
≥2 15 (10%)
Smoking habits of parents
Smokers 86 (57%)
Non-smokers 66 (43%)
No 117 (77%)
Yes 35 (23%)
Allergic rhinitis
No 134 (88%)
Yes 18 (12%)
Food-specific IgE (egg and/or milk and/or wheat)
Negative 60 (39%)
Positive 92 (61%)
HDM-specific IgE a
Negative 118 (81%)
Positive 28 (19%)
DOI: 10.7717/peerj.1585/table-1


The information of mother’s age, gestational age, birth weight, and HDM-specific IgE was unavailable in 33, 1, 3, and 5 subjects, respectively.

The distributions of maternal CES-D and JPSS scores were compared by the following variables: history of allergic diseases, the presence of specific IgE antibodies against foods (egg, milk or wheat) and HDM, age and sex of the children, mother’s age, the number of siblings, and smoking habits of parents. The maternal JPSS scores of the IgE-positive group for HDM were significantly higher than those of the negative group. A similar tendency was also observed in CES-D scores. The maternal CES-D scores of boys were significantly higher than those of girls (P = 0.012). There was a similar trend in JPSS scores and the difference was marginally significant (P = 0.083). Mothers with more than one child showed significantly higher JPSS scores than mothers with only one child (P = 0.011). A similar tendency was also observed in CES-D scores (Table 2).

Table 2:
Comparison of CES-D and JPSS scores by potential determinants for allergic disease.
N Median 25th, 75th P valuea N Median 25th, 75th P valuea
All 150 8 3, 12 151 23 17, 27
No 116 7 3, 12 0.355 117 23 17, 26 0.187
Yes 34 9 6, 12 34 25 19, 29
Allergic rhinitis
No 132 8 3 ,12 0.285 133 23 17, 27 0.691
Yes 18 10 6, 12 18 23 19, 27
Food-specific IgE
Negative 59 7 3, 12 0.739 59 23 18, 28 0.913
Positiveb 91 8 4, 12 92 23 17, 27
HDM-specific IgE
Negative 117 7 3, 12 0.148 117 23 17, 26 0.034
Positive 28 9 5, 14 28 26 19, 31
Boy 83 9 5, 13 0.012 83 24 19, 28 0.083
Girl 67 6 1, 10 68 22 16, 26
Age of children (month)
<13 72 7 2, 12 0.365 73 23 17, 27 0.570
≥13 78 8 5, 12 78 24 18, 27
Age of mother (year)
<31 57 8 3, 12 0.512 59 23 16, 27 0.127
≥31 60 7 3, 11 59 22 17, 26
Unknown 33 8 5, 12 33 26 20, 31
Number of siblings
0 81 7 3, 11 0.106 82 22 17, 26 0.011
≥1 69 9 4, 14 69 25 20, 30
Smoking habits of parents
Smokers 84 9 4, 14 0.100 86 23 19, 28 0.416
Non-smokers 66 7 3, 10 65 23 17, 26
DOI: 10.7717/peerj.1585/table-2


P value was obtained by Mann–Whitney U test except for age of mother. Kruskal–Wallis test was applied for age of mother.
Egg and/or milk and/or wheat-specific IgE were positive.

The full scores of CES-D and JPSS were 60 and 56, respectively.

Table 3:
Results of multivariable analysis in the relationships between CES-D and cytokine expressions by allergic disease.
IL-6 IL-8 IL-10 IL-22
CES-D Coefficient SE P valuea Coefficient SE P valuea Coefficient SE P valuea Coefficient SE P valuea
All (N = 150) −0.011 0.013 0.393 −0.005 0.014 0.737 0.008 0.016 0.615 −0.016 0.024 0.503
No (N = 116) −0.032 0.016 0.043 −0.024 0.017 0.168 −0.006 0.020 0.778 −0.044 0.029 0.134
Yes (N = 34) 0.045 0.022 0.048 0.037 0.024 0.143 0.043 0.030 0.153 0.043 0.039 0.276
Allergic rhinitis
No (N = 132) −0.018 0.014 0.182 −0.012 0.015 0.437 0.002 0.018 0.892 −0.021 0.026 0.421
Yes (N = 18) 0.072 0.055 0.217 0.045 0.041 0.274 0.065 0.044 0.169 −0.044 0.064 0.504
Asthma and/or allergic rhinitis
No (N = 107) −0.032 0.016 0.046 −0.021 0.018 0.236 −0.001 0.021 0.949 −0.045 0.030 0.137
Yes (N = 43) 0.043 0.025 0.092 0.033 0.024 0.170 0.031 0.027 0.244 0.043 0.038 0.258
Food-specific IgE
Negative (N = 59) −0.014 0.026 0.591 −0.013 0.023 0.583 −0.006 0.037 0.877 −0.011 0.047 0.824
Positiveb (N = 91) −0.011 0.015 0.477 −0.003 0.186 0.859 0.011 0.017 0.527 −0.020 0.028 0.489
HDM-specific IgE
Negative (N = 117) −0.002 0.143 0.881 −0.022 0.017 0.192 0.003 0.020 0.900 −0.018 0.029 0.531
Positive (N = 28) 0.003 0.027 0.899 0.020 0.033 0.550 0.033 0.031 0.300 −0.055 0.048 0.265
DOI: 10.7717/peerj.1585/table-3



regression coefficient


standard error

P value was obtained by multivariate regression models adjusted for sex, age, and number of siblings.
Egg and/or milk and/or wheat specific IgE were positive.
Table 4:
Results of multivariable analysis in the relationships between JPSS and cytokine expressions by allergic disease.
IL-6 IL-8 IL-10 IL-22
JPSS Coefficient SE P valuea Coefficient SE P valuea Coefficient SE P valuea Coefficient SE P valuea
All (N = 151) −0.001 0.011 0.951 0.007 0.012 0.556 0.002 0.014 0.883 −0.005 0.021 0.794
No (N = 117) −0.013 0.014 0.325 −0.011 0.015 0.476 −0.010 0.017 0.563 −0.024 0.025 0.347
Yes (N = 34) 0.029 0.019 0.141 0.052 0.022 0.023 0.026 0.026 0.320 0.029 0.033 0.391
Allergic rhinitis
No (N = 133) −0.009 0.012 0.441 0.001 0.013 0.945 −0.003 0.015 0.832 −0.012 0.022 0.598
Yes (N = 18) 0.115 0.045 0.023 0.063 0.036 0.104 0.055 0.042 0.216 0.013 0.061 0.837
Asthma and/or allergic rhinitis
No (N = 108) −0.018 0.013 0.195 −0.010 0.015 0.531 −0.010 0.018 0.564 −0.026 0.026 0.310
Yes (N = 43) 0.042 0.021 0.050 0.050 0.021 0.022 0.028 0.023 0.232 0.038 0.032 0.241
Food-specific IgE
Negative (N = 58) 0.0028 0.024 0.908 0.005 0.022 0.825 0.002 0.035 0.948 −0.019 0.045 0.673
Positiveb (N = 91) 0.0006 0.013 0.964 0.005 0.015 0.733 −0.005 0.014 0.696 −0.001 0.023 0.960
HDM-specific IgE
Negative (N = 116) 0.005 0.012 0.654 −0.013 0.014 0.379 −0.008 0.017 0.655 −0.012 0.025 0.638
Positive (N = 28) 0.017 0.025 0.484 0.033 0.030 0.275 0.033 0.028 0.247 −0.028 0.045 0.535
DOI: 10.7717/peerj.1585/table-4



regression coefficient


standard error

P value was obtained by multivariate regression models adjusted for sex, age, and number of siblings.
Egg and/or milk and/or wheat specific IgE were positive.

To clarify the effects of maternal psychological stress on their children, we examined the associations between CES-D/JPSS scores and the expression levels of four inflammatory biomarkers. The results of multivariable regression analyses are shown in Tables 3 and 4. The maternal CES-D scores were positively associated with IL-6 expression in children with asthma but not in those without allergic diseases. This interaction was statistically significant for IL-6 expression (P = 0.011 by likelihood ratio test). A similar tendency was also observed between subjects positive and negative for HDM-specific IgE, but there was no statistical significance. On the other hand, association or interaction with cytokine expressions was not observed for food-specific IgE (Table 3).

Regarding the JPSS, the score was positively associated with IL-8 expression among subjects with asthma and/or allergic rhinitis and there was marginally significant association with IL-6. However, subjects without these allergic diseases showed negative tendencies (Table 4). Additionally, there were significant interactions between JPSS score and the presence of asthma and allergic rhinitis regarding IL-6 (P = 0.017) and IL-8 (P = 0.034) expressions. The subjects positive for HDM-specific IgE were more likely to show positive associations between JPSS score and cytokine expressions than those negative for HDM-specific IgE. However, there was no statistical significance between them.


In this study, maternal CES-D score was positively associated with IL-6 expression in children with asthma, and maternal JPSS score was positively related to IL-6 and IL-8 expressions among children with asthma and/or allergic rhinitis. They were not so strongly correlated, but the results are interesting. Recently, Shi et al. (2010) found that the protease-activated receptor (PAR)/phosphatidylinositol 3 kinase (PI3K)/NFκB signaling pathway is involved in the induction of IL-6 and IL-8 by Dermatophagoides pteronyssinus 1 (Der p1) in airway epithelial cells. Der p1 is one of the major protease allergens of HDM, and is a crucial aeroallergen and an independent risk factor for asthma and/or allergic rhinitis (Huss et al., 2001). The PAR/PI3K/NFκB signaling pathway by Der p1 might be activated by the stress of caregivers. As a result, IL-6 and IL-8 might have increased in our study. Certainly, if this pathway is really related to our results, the relationship of the stress of caregivers with IL-8 and IL-6 would have been stronger in the HDM-specific IgE-positive group. However, the subjects were young children. Generally, HDM-specific IgE increases after the age of three (Ng, Holt & Prescott, 2002). The age of our subjects might have influenced our results.

Since the psychological status of young children cannot be evaluated, the question is whether there is a correlation between the psychological stress of a mother and her child. A recent cohort study, Growing Up in Ireland, National Longitudinal Study of Children, reported that babies’ development could be affected by stress or depression among their parents (Nixon, Swords & Murray, 2013). This finding suggested the possibility that maternal stress could influence the immune system of children. Since the immune responses of allergic children are more sensitive than those of healthy children, the increases of IL-6 and -8 expressions could be detected in their study. In order to support our research, further experiments in vitro and in vivo are required to clarify the relationship between psychological stress, allergy, and PAR/PI3K/NFκB signaling pathway.

In this study, both maternal perceived stress and depression showed similar tendencies in the association with cytokine expression patterns by the presence of allergic diseases (Tables 3 and 4). Although there was a significant correlation between CES-D and JPSS scores, Spearman’s rho: 0.78, P = 0.001 (data not shown), the associations between JPSS and cytokine expressions were relatively evident. Perceived stress is slightly different from depression psychologically. Since there is a strong relationship between high stress level and depression (Wiegner et al., 2015), it is difficult to distinguish between them (Van Praag, 2004). Furthermore, the difference in the score distribution between CES-D and JPSS could explain the evident results in JPSS. The JPSS score distribution was close to normal distribution with a median of 23 (56 points = full score). However, the CES-D score distribution was skewed to the right with a median of 8 (60 points = full score), and this might be one of the reasons for the weak association with CES-D. Tomljenovic, Pinter & Kalogjera (2014) reported that chronic psychological stress might be one of the factors that modifies disease severity and may lead to uncontrolled diseases in chronic rhinosinusitis patients. Patterson et al. investigated whether participants’ reporting of allergy flares correlated with perceived emotional stress, depression, mood, and a biomarker of stress. They reported that the allergy flare group had higher perceived stress scores than those of the group without allergy symptoms. Perceived stress, but not depressive symptoms, is positively correlated with allergy flares (Patterson et al., 2014). Considering the findings of these studies, allergic diseases might be more related to perceived stress than depression.

Our results were interesting in that IL-6 and IL-8 were related to physiological stress and childhood allergies. IL-6 is a T helper type 2 (Th2) cytokine. Th2 cells produce other various cytokines, such as IL-4, IL-5 and IL-13 (Fallon et al., 2002). IL-1β potently stimulates IL-8 expression (Liu et al., 2014). We should measure such cytokines in the future to confirm our results.

In this study, the asthmatic group included subjects with a combination of criteria: diagnosis by a doctor, symptoms and receipt of treatment. However, diagnosing asthma in pre-school children is difficult. Spirometry and lung function parameters could support the clinical diagnosis of asthma, but most techniques depend on the effort and cooperation of the patient and thus not applicable in pre-school children (Kooi et al., 2006). Therefore, there is the possibility that children without asthma might have been included in the asthma group in this study.

In conclusion, the most important finding from this study is that there are significant relationships between maternal psychological stress and IL-6 and IL-8 mRNA expressions in children with asthma and allergic rhinitis.

Supplemental Information

5 Citations   Views   Downloads