Relationship among airborne pollen, sensitization, and pollen food allergy syndrome in Asian allergic children

Introduction

Airborne pollen from birch, alder, hazel, oak, hornbeam, chestnut, and beech are predominant in Northern and Central Europe and are a major cause of allergic rhinitis and possibly asthma symptoms (Biedermann et al., 2019). The broad cross-reactivity of birch pollen allergens extends to plant food allergens, resulting in pollen food allergy syndrome (PFAS). PFAS has different causes depending on the major pollen present in the region. The prevalence of PFAS is reportedly varied, ranging from 4.7% to >20% in children and 13%–58% in adults (Carlson & Coop, 2019). In Japan, the primary causative foods for initial food allergy are fruits among 4–6-year-olds and crustaceans (first) and fruits (second) among 7–19-year-olds (Ebisawa, Ito & Fujisawa, 2020). Few reports have assessed pediatric PFAS in East Asian countries such as Japan and South Korea (Takemura et al., 2020; Kim et al., 2018).

Busan (South Korea), Fukuoka and Tochigi (Japan) have similar average temperature, humidity, and types of fruits ingested; however, they differ in terms of tree vegetation (Fig. 1). In Japan, airborne pollen from Japanese cedar and other related trees are predominant and are major causes of seasonal allergic rhinitis (AR) (Kishikawa et al., 2017). In contrast, several pine pollen are scattered throughout South Korea. The airborne pollen counts may affect pollen sensitization, which may affect the prevalence of PFAS in the area. Therefore, the present study aimed to assess airborne pollen counts, Immunoglobulin E (IgE) sensitization rates, and PFAS prevalence among children with allergies in three regions of South Korea and Japan.

Materials & Methods

This multicenter prospective cross-sectional study was conducted at the Department of Pediatrics, Busan St. Mary’s Hospital, National Hospital Organization Fukuoka National Hospital and Dokkyo Medical University Hospital in Tochigi. Of the outpatients aged 5–17 years who visited three different institutions in 2017 for allergic diseases (bronchial asthma (BA), allergic rhinitis (AR), food allergy (FA), or atopic dermatitis (AD)), 57 out of 169 (34%) in Busan, 56 out of 384 (15%) in Fukuoka, and 20 out of 222 (9%) in Tochigi agreed to participate in the survey. A limiting factor for enrollment in this study would be if the patient had a recent blood test.

Results

In total, 133 patients (Busan: n = 57, Fukuoka: n = 56, and Tochigi: n = 20), were included in the study. There was no significant difference in terms of age, sex, PFAS incidence, seasonal AR, and AD in pediatric patients with allergies among the three sites (Table 1). However, the proportion of participants with BA and FA were higher in Fukuoka and Tochigi than in Busan (BA: 55%, 60%, 28%: p = 0.009, FA: 80%, 65%, 25%: p < 0.001).

According to the 2017 annual airborne pollen counts of each region (Table 2), in Busan, the pine pollen count was the highest (74%), followed by alder (14%), Japanese hop (7%), juniper (3%), grass (1%), and birch (1%), while airborne cypress pollen (68%) in Fukuoka and Japanese cedar pollen (79%) in Tochigi were predominant. The alder pollen counts in Fukuoka and Tochigi were not clearly evaluated, and grass and ragweed pollen counts did not differ among the three sites.

Children in Fukuoka and Tochigi had significantly higher sensitization rates to Japanese cedar (p < 0.001), cypress (p < 0.001), juniper (p < 0.001), orchard grass (p = 0.007), ragweed (p = 0.005), Japanese hop (p = 0.005), and tomato (p = 0.006), compared with children in Busan (Table 3). The sensitization rates to Dp, pine, birch, alder, and peach did not differ among children in the three sites. In Fukuoka and Tochigi, where Japanese cedar and cypress pollen were frequently scattered, a high sensitization rate among allergic children was observed. The sensitization rate was not affected by the pollen count in alder, grass, ragweed, and Japanese hop.

Table 3 shows the differences in sensitization rates between children with and without PFAS. In all three sites, the sensitization rates to birch and alder among patients with PFAS were higher than those among patients without PFAS. Additionally, there were regional differences in the sensitization rates between children with and without PFAS; in Busan, patients with PFAS had a greater sensitization rate to Japanese cedar and peach than those without PFAS, whereas in Fukuoka, sensitization rates to pine, ragweed, Japanese hop, and peach were higher in patients with PFAS than in those without PFAS.

Univariable logistic regression analysis showed that Japanese cedar, ragweed, alder, pine, and Japanese hop were statistically significant variables (Table 4). Adolescence was believed to increase PFAS incidence, but older age (≥10 years) had no significant effect on PFAS. The multivariable logistic regression analysis conducted using PFAS as the dependent variable and location and airborne pollen sensitization as independent variables, showed that only alder sensitization was significantly associated with PFAS in all patients (odds ratio: 6.62, 95% confidence interval: 1.63–26.87, p = 0.008)

Foods such as kiwi, peanuts, cashew nuts, banana, tomato, mango, pineapple, mandarin, melon, chestnuts, spinach, and eggplant were found to cause PFAS among children in Fukuoka (Table 5). Specifically, the number of children with PFAS caused by peach was higher in Busan and Tochigi than in Fukuoka, while kiwi and peanuts were the causative foods among children with PFAS at all sites.

Discussion

In the present study, we demonstrated that alder pollen sensitization was associated with PFAS among children with allergies in three regions of South Korea and Japan. Alder trees are distributed across the Korean Peninsula and all areas of Japan. In Busan, alder pollen was the most common pollen after pine pollen and it had a 71% sensitization rate among patients with PFAS. Similarly, in Fukuoka and Tochigi, although alder pollen was rare, patients with PFAS had a high alder pollen sensitization rate.

In Europe, birch pollen rather than alder pollen is a common cause of PFAS (Biedermann et al., 2019). Birch vegetation is prevalent in Busan but not in Fukuoka (Osawa et al., 2020). This corresponds to the annual airborne pollen counts that showed only a small proportion of birch pollen in Busan and Tochigi, while no birch pollen was found in Fukuoka. In South Korea and Japan, although there is some birch pollen dispersal, sensitization to pollen occurs from both the Betulaceae (alder and birch) and Fagaceae families (oak, chestnut, and walnut), which could comprise the birch homologous group causing PFAS (Masumoto et al., 2018), as in Europe (Pedrosa et al., 2020).

The birch allergen Bet v 1 (also known as PR-10), a component of apple and peanut, was also found to be associated with peach, which is a common causative food for PFAS among children in Busan and Tochigi. Unfortunately, Betv1-IgE measurements were not investigated in this study. In Fukuoka, however, peach was not the causative food item of PFAS, yet a high sensitization to birch, alder, and peach was observed. Previous studies have revealed that the average yearly consumption of peaches in Busan, Tochigi, and Fukuoka is 16, 1.9, and 0.8 per person, and that of apples is 27, 19, and 15 per person (Statista Research Department, 2022; Regional Container, 2022). Therefore, the lack of peach and apple-related PFAS detection in Fukuoka may be attributed to the low peach and apple consumption in the region.

Patients with PFAS in Fukuoka had a higher sensitization rate to ragweed and Japanese hop than those in Tochigi and Busan. The ragweed pollen allergens include Amb a 6 (known as LTP), which is a component of kiwi, melon, peanut, walnut, apple, and peach, and Amb a 8, which belongs to the Profilin family and is found in kiwi, melon, pineapple, peanut, apple, banana, and peach (Carlson & Coop, 2019). Japanese hop is a major allergenic airborne pollen in Korea, and it was not significantly associated with ragweed pollen sensitization (Park et al., 2001). Among children in Fukuoka, Japanese hop was an important allergenic airborne pollen causing PFAS (Jang et al., 2021).

Despite the high pollen counts in Busan, patients were not as sensitized to Japanese hop as in Fukuoka. Furthermore, in Busan and Fukuoka, the quantity of grass and weed pollen collected in each region was significantly smaller than the quantity of tree pollen, and there was no significant annual change in the pollen counts as in the case of Japanese cedar pollen (Kishikawa et al., 2019; Sung et al., 2014). Importantly, the Japanese and Korean methods of assessing airborne pollen counts differ; therefore, we could not compare the absolute value of the airborne pollen counts between the two countries. The pollen of Japanese hop is small and light; thus, it may have been underestimated by the Durham sampler, which is the gravity method in Fukuoka. However, pollen counts obtained by Rotorod and Durham sampler methods were previously reported to be correlated (Sun et al., 2017).

The airborne pollen sensitization rates may vary based on the amount of airborne pollen scattered and the characteristics of children with the allergy. Japanese cedar pollen was previously demonstrated to cross-react with tomato pollen (Kondo et al., 2002). In our study, the sensitization rate to tomato was significantly higher among individuals in Japan. This may explain the occurrence of PFAS associated with tomato among children in Fukuoka and Tochigi. Pine pollen has not yet been molecularly analyzed, but a previous study has shown immunological cross-reactivity between airborne pollen proteins and pine nuts, which are not consumed in Japan (Senna et al., 2000). In Fukuoka, pine pollen might have an allergenic component that causes PFAS. Therefore, further investigation on the allergenicity of pine and Japanese hop pollen should be performed to better understand their molecular backgrounds.

This study has several limitations. First, specific IgE tests for different types of airborne pollen after pollen dispersal were not conducted. If tests for various types of grass and weed pollen were performed after pollen dispersal, their association with PFAS could be further investigated. Second, only few patients from Tochigi were included in the analysis. Third, in this study, pollen count data were obtained from a single year. The amount of pollen dispersal varies yearly depending on the annual temperature and humidity, thus, limiting our evaluation of the relationship between pollen counts and sensitization. Finally, the proportion of patients with BA and FA were higher in Fukuoka and Tochigi than in Busan, which supports previous findings on the higher prevalence of BA and FA in Japanese children than in Korean children (Lai et al., 2009; Kim et al., 2015).

Conclusions

Despite a small proportion or none of birch pollen was observed, alder pollen sensitization was associated with PFAS among children at Pusan in South Korea and at Tochigi and Fukuoka in Japan. Ragweed and Japanese hop pollen sensitization were found to cause PFAS, particularly among children at Fukuoka in southern Japan. The characteristics of PFAS varied according to local vegetation, pollen size and weight, eating habits, and comorbid allergic diseases. Further investigation is required to assess the molecular background of local airborne pollen and food items causing PFAS.

Supplemental Information