C.-M. Chen and S. Weidinger contributed equally to this work.
Common variants in FCER1A influence total serum IgE levels from cord blood up to six years of life
Article first published online: 24 FEB 2009
© 2009 John Wiley & Sons A/S
Volume 64, Issue 9, pages 1327–1332, September 2009
How to Cite
Chen, C.-M., Weidinger, S., Klopp, N., Sausenthaler, S., Bischof, W., Herbarth, O., Bauer, M., Borte, M., Schaaf, B., Lehmann, I., Behrendt, H., Krämer, U., Berdel, D., Von Berg, A., Bauer, C. P., Koletzko, S., Illig, T., Wichmann, H.-E., Heinrich, J. and the LISA and GINI Study Group (2009), Common variants in FCER1A influence total serum IgE levels from cord blood up to six years of life. Allergy, 64: 1327–1332. doi: 10.1111/j.1398-9995.2009.02005.x
- Issue published online: 14 SEP 2009
- Article first published online: 24 FEB 2009
- Accepted for publication 22 January 2009
- cord blood;
- total IgE
Background: In a recent genome wide scan, a functional promoter variant (rs2251746) in the gene encoding the alpha chain of the high affinity receptor for immunoglobulin E (IgE) (FCER1A) was identified as major determinant of serum IgE levels.
Objective: The aim of this study was to investigate the role of rs2251746 on total IgE levels measured at different stages of life from birth (cord blood) up to the age of 6 and to evaluate its interaction with the environmental influences in two German birth cohorts.
Method: Data from two German birth cohorts were analysed (n = 1043 for the LISA cohort and n = 1842 for the GINI cohort). In the studies, total serum IgE was measured from cord blood, and blood samples taken at the age of 2/3 and 6 years. In a subgroup of the LISA study, house dust samples were collected at age of 3 months and the amount of endotoxin was determined. Random effect models were used to analyse the longitudinal health outcomes.
Results: In the two cohorts, the heterozygote and the rare homozygote of rs2251746 was consistently associated with lower total IgE levels from birth up to the age of 6 years with an allele-dose effect (P < 0.02 for blood samples taken at each time point in both cohorts). No interaction between the two FCER1A encoding gene and environmental exposures including endotoxin, worm infestation and day care centre attendance during early childhood were observed.
Conclusion: Common variants in FCER1A strongly influence basal IgE production independently from environmental stimuli. These effects can be observed already in cord blood pointing to altered gene expression in foetus.
Atopic diseases such as asthma, rhinitis and eczema are among the most common chronic diseases in industrialized countries (1, 2). Immunoglobulin E (IgE) is a central player in the allergic response, and raised total IgE levels are considered as an indicator for atopy or potential development of atopy. In addition, high levels of IgE are associated with parasite and helminth infections to which they are thought to confer resistance (3). Furthermore, IgE may be involved in certain autoimmune diseases and anti-tumoral defence (4, 5).
There is a large body of evidence that the regulation of serum IgE levels is under a strong genetic control with heritability estimates of 50–80% (6–8), and it has been suggested that genetic regulation of basal IgE production is independent of specific responses to allergens (9, 10).
However, so far only few studies investigated total IgE as primary trait in unselected populations and there are only very few genes that have been associated consistently and robustly, most notable STAT6 and IL13 (11). Very recently, a genome wide scan in a German population-based cohort of adults identified the gene encoding the alpha chain of the high affinity receptor for IgE (FCER1A) on chromosome 1 as novel susceptibility locus influencing total serum IgE levels. A functional promoter variant affecting FCER1A expression (rs2251746) and a tightly linked variant (rs2427827) were found to be highly associated with total serum IgE levels, a finding that was replicated in more than 10 000 individuals from four independent population-based cohorts (12).
We sought to further investigate the effect of the functional polymorphism rs2251746, which was shown to affect gene expression, on the value of total IgE measured at birth from cord blood and at the age of 2, 3 and 6 years in two German birth cohorts to determine at which period of life the single-nucleotide polymorphism (SNP) affects total IgE levels and to evaluate the role of environmental stimulations during specific periods of life.
The study population was composed of two ongoing birth cohorts conducted in Germany. The influences of lifestyle-related factors on the immune system and the development of allergies in childhood study (LISA) and the German Infant Nutritional Intervention Program (GINI). Detailed description of screening and recruitment has been provided elsewhere (13–16).
Briefly, in the GINI study, 5991 newborns from maternity wards in Munich and Wesel, Germany fulfilled the inclusion criteria were recruited in the study between September 1995 and June 1998. There are two subgroups in the GINI study: interventional study arm and noninterventional study arm (observational cohort group). A number of 2252 infants with at least one parent or sibling having a history of allergic disease agreed to participate in the randomized trial which aimed to compare the effect of hydrolyzed formulas and conventional cow’s milk formula on the prevention of the development of allergic disease in high risk children. The infants with no family history of allergic disease and those whose parents denied participating in the trial were allocated in the noninterventional study arm (n = 3739). Children from both groups were followed-up at the age of 1, 2, 3, 4 and 6 years.
In the LISA study, parents of neonates admitted to maternity hospitals in Munich, Leipzig, Wesel, and Bad Honnef, Germany were contacted. A number of 3097 neonates were recruited in the study between December 1997 and January 1999. LISA is designed as a population-based study and the participants were not preselected based on family history of allergic disease. The cohort has been followed up at the age of 6, 12 and 18 months and 2, 4 and 6 years. House dust samples were only collected at the Munich and the Leipzig populations. Both studies were approved by the ethics committees of Bavarian General Medical Council and the Ethics Commission of the University of Leipzig.
Information on potential environmental stimulus of total IgE including day care centre attendance and doctor diagnosed worm infestations collected using self-administered questionnaires after birth and during follow-ups.
Collection and analysis of blood samples
In the LISA study, cohort’s cord blood samples were collected from 1983 (64%) children at birth and the blood samples were collected from 2176(82%) and 1193 (54%) children when the children were 2 and 6 years old, respectively. Total IgE antibodies were measured using RAST FEIA CAP system (Pharmacia, Freiburg, Germany) with a detection limit 0.32 kU/l. Children with total IgE level under the detection limit were assigned with 0.18 kU/l. Cord blood IgA in the samples was also determined to exclude cord blood contaminated from maternal blood(17). The analysis showed that none of the samples contains IgA level exceeded the cut-off point of 32 μg/l.
In the GINI Study, cord blood samples were collected from 1441 (64%) of the interventional arm with a detection limit of 0.18 kU/l. Children with total IgE level under the detection limit were assigned with 0.18 kU/l. Contamination of cord blood by maternal blood was also excluded by measurement of cord blood IgA levels in samples with detectable IgE levels. At age of 3 years, blood samples were again collected from 1027 (59%) of the interventional arm and when the children were 6 years old, blood samples of 1962 (51%) children from the whole GINI study population (n = 1017 from the interventional arm) were collected.
In the 6-year follow-up, the parents from a total number of 1043(47%) children from the LISA study and 1842(47%) children from the GINI study have agreed to genotype their children’s blood samples. Two FCER1A variants were typed: the functional polymorphism rs2251746 and rs2427827, which was previously shown to be in strong linkage disequilibrium (LD) with rs2251746. Genotyping of SNPs was realized with the iPLEX (Sequenom, San Diego, CA, USA) method by means of matrix assisted laser desorption ionization-time of flight mass spectrometry method (MALDI-TOF MS, Mass Array; Sequenom) according to the manufacturers instructions. Genotyping was carried out by laboratory personnel blinded to IgE status. Standard genotyping quality control included 10% duplicate and negative samples. Genotyping discordance rate was below 0.3%. All SNPs were in Hardy–Weinberg Equilibrium (HWE).
Collection and analysis of house dust
In the LISA study, the house dust samples were collected when the children were 3 months old. A total of 2166 (89%) families recruited from the Munich and Leipzig centre participated. Trained field workers carried out dust sampling based on a standardized operating procedure (18, 19). Dust samples were taken from parents’ and child’s mattresses by vacuuming 1 m2 of the mattress surface for 2 min using vacuum cleaners equipped with special nozzles (Abelló allergen mouthpiece; ALK, Hørsholm, Denmark). The samples were stored at −20°C until extraction to prevent bacteria growth. The dust samples plus filter paper were extracted with 0.125 mol/l NH4HCO3 + 0.05% Tween-20 (v/v) for 2 h at room temperature under constant shaking, with an extraction ratio of 1 : 10–1 : 100 (w/v), depending on the amount of sampled dust. House dust endotoxin levels from the dust extracts were quantified using the kinetic Limulus Amebocyte Lysate assay with 50 EU/g dust detection limit. The endotoxin levels were expressed as surface load, the amount of endotoxin/m2 of sample surface.
Effects on total IgE were analysed using longitudinal analysis with random effects regression models. The longitudinal analysis approach takes into account the dependence of repeated outcome measures within each subject. The total IgE levels were log transformed for the analysis due to their right-skewed distribution. The analyses were performed with proc glimmix procedure, sas version 9.1 (SAS Institute Inc., Cary, NC, USA).
Phenotype and genotype information
Total IgE levels of the study population and baseline information are shown in Table 1. The genotyping success rate was 99.5% for rs2251746 and 98.0% for rs2427897 in the LISA cohort and 97.0% for rs2251746 and 99.8% for rs2427897 in the GINI cohort. As statistical analysis confirmed the previously observed strong LD between the two SNPs (D′ = 0.98, R2 = 0.97 for the LISA study population and D′ = 0.997, R2 = 0.99 for the GINI study population), we report only effects of the functional variant rs2251746. Analysis showed strong association of this polymorphism on total IgE levels in cord blood and in the blood samples collected when the children were 2/3 and 6 years old, and in addition pointed to an allele-dose effect (Table 2). Figure 1 shows the percentage of children of each RS2251746 genotypes who have Total IgE level above the 75th percentile of each study cohort. Longitudinal analysis approaches showed that in both cohorts, the heterozygote and the rare homozygote genotype of rs2251746 were constantly associated with a lower total IgE level from birth up to age of 6 years (Table 3). Analysis of interaction with time showed that the effect of polymorphism did not vary with the outcomes measured at different time points during childhood.
|Serum total IgE level (kU/l) (median (25th and 75th percentile)||Cord blood||0.18 (0.18, 0.45)||0.18 (0.18, 0.37)|
|At age 2||19.60 (8.20, 48.20)||25.50 (10.50, 61.10)|
|At age 6||40.40 (14.70, 109.00)||43.10 (16.30, 115.00)|
|n/N (%)||n/N (%)|
|rs2251746||CC||71/1043 (6.8)||132/1842 (7.2)|
|TC||399/1043 (38.3)||706/1842 (38.3)|
|TT||568/1043 (54.5)||948/1842 (51.5)|
|Day care attendance||At age 1||44/975 (4.5)||15/1744 (0.9)|
|At age 2||203/1035 (19.6)||85/1707 (5.0)|
|Doctor diagnosed worm infection||At age 2||1/1032 (0.1)||9/1752 (0.5)|
|At age 4||22/1007 (2.2)||39/1725 (2.3)|
|At age 6||27/1035 (2.6)||45/1805 (2.5)|
|Allele-dose effect*† (P-value)|
|LISA||rs2251746||Total cord blood IgE||<0.0001‡|
|Total IgE at age 2||0.0201|
|Total IgE at age 6||<0.0001|
|GINI||rs2251746||Total cord blood IgE||0.0067§|
|Total IgE at age 3||0.0065|
|Total IgE at age 6||0.0188|
|Longitudinal analysis (kU/l)|
|TC||−3.02 (−4.37, −1.4)|
|CC||−4.24 (−6.24, −1.34)|
|TC||−3.91 (−6.11, −1.29)|
|CC||−7.88 (−10.59, −4.12)|
Effect of FCER1A variants on other clinical outcomes
Further analysis showed no significant association of FCER1A variants on other clinical outcomes such as atopic eczema or specific sensitization against common inhalant and food allergens (data not shown). However, the number of cases for these traits was low in our cohorts thus limiting statistical power [sensitization to common inhalant allergen at age 6, n = 260 (26%) in the LISA cohort, n = 554 (30%) in the GINI cohort].
Interaction with environmental exposure
Dust samples from 2118 (87%) parent mattresses and 2098 (86%) children mattresses were available from the Munich and Leipzig sub study population of the LISA study. All the dust samples contained endotoxin above the detection limit of 50 EU/g. The median (25th–75th percentile) of the load and concentration of endotoxin from mothers’ mattress dust were 2071 (595–6919) EU/m2 and 3008 (1046–7913) EU/g, respectively, and in dust samples of children’s mattress, they were 1015 (330–3022) EU/m2 and 5866 (2336–14669) EU/g, respectively. The correlation between the amount of endotoxin in mothers’ and children’s mattress dust samples were 0.4 for both concentration and surface load. In those families who have agreed the settled house dust collection, 25% of them dropped out the study during follow-ups up to age 6. However, exclusion of the lost-of-follow-ups did not substantial change the distribution of the endotoxin exposure (data not shown).
No interaction between rs2251746 and environmental exposures including endotoxin exposure, worm infestation during childhood and day care centre attendance during early childhood were observed (data not shown).
In our study on two independent birth cohorts, we could confirm the recently reported association of a functional FCER1A promoter polymorphism with total IgE levels (12). Interestingly, the variant was consistently and strongly associated with IgE levels throughout the observation period, i.e. with total IgE value measured in cord blood and blood samples taken at the age of 2, 3 and 6 years. In addition, we observed an additive allele-dose effect. No interaction between the SNP and environmental exposures including indoor endotoxin (available only in a sub group of the LISA study), day care centre attendance and worm infestation was observed. IgE is usually present at low concentration in serum, and increased levels are observed after stimulation, e.g. by infection (20) or allergens. The high-affinity receptor for IgE (FcɛRI) plays a key role in the induction and maintenance of allergic responses, and its α-chain is responsible for binding of IgE. The previous studies indicated that polymorphisms in the gene encoding the β-chain of the high affinity IgE receptor, which functions as an amplifier of FCER1 expression and signalling, is associated with IgE responsiveness, but so far lack robust replication (21).
It is well established that IgE synthesis can be observed in foetal liver and lung already at 11 weeks of gestation (22). However, so far it is unclear whether genetic factors influence IgE production already in utero, and whether these factors are the same as those operating later in life. Based on the low longitudinal correlation of IgE levels and on results from twin studies, it has been suggested that different genetic factors regulate serum IgE levels at different ages (7, 23–25). The results of our present study indicate that FCER1A is a major gene, with common variants influencing IgE levels at different stages of life and operating as early as during the foetus period. The fact that none of the environmental exposures after birth included in the present analysis showed a substantial modifying effect of total serum IgE level measured after birth may indicate that some pattern of the IgE response is already developed during the foetal period and is possibly influenced by environmental exposures in utero. The origin of cord blood IgE is still a matter of debate (26, 27). Although measurement of IgA antibodies was performed in CB samples to exclude contamination with maternal blood (28), we cannot totally rule out the possibility of transmission of maternal IgE during pregnancy or contamination with maternal IgE during delivery in our samples. However, it is well known that total serum IgE levels and receptor bound IgE are closely correlated, and that the expression level of FCER1A is crucial for serum IgE levels (29). Thus, the observed association of rs2251746 with cord blood IgE levels could also be due to its influence on in utero FCER1A expression, which in turn determines the levels of free serum IgE. However, our results may also be interpreted as a lack of dependence of the FCER1A variants on environmental stimuli or that we have not yet found the relevant environmental factors. Further analysis of a potential interaction with environmental factors such as allergens and pet exposures need to be carried out in larger cohorts with increased statistical power to clarify that issue. Furthermore, the number of children who have been exposed to the environmental factors which were analysed in this study such as attending the day care centre or having worm infestation is very small. The negative results of the gene environmental interaction may also be due to the lack of power.
In our study, the further sub-analyses showed no significant associations between the FCER1A variants and specific sensitization or atopic eczema. In the original report by Weidinger et al. (12), a smaller yet significant effect of FCER1A variants on specific IgE levels was shown, whereas no association with atopic eczema was detected. However, the number of cases for these traits was low in our children cohorts and thus we cannot rule out that the lack of association is due to the lack of power. In addition, it has to be considered that in complex diseases such as atopic eczema many genes may act through several intermediate phenotypes to increase disease risk, and therefore absence of association between a single intermediate-trait-associated gene and a related complex disease does not exclude the possibility that the gene plays a role in the pathogenesis of the disease. However, it is important to note that based on the current study, FCER1A variant alone does not show a significant association on the development of clinical diseases. Further studies including gene–gene and gene–environment interaction in atopic disease cohorts are needed to clarify the impact of FCER1A in atopic disease.
In this study, <50% of the children have agreed to provide their blood samples for genotyping, we therefore performed sensitivity analysis to ensure the representative of the children with the genotype information. The results showed that there is no significant difference of the prevalence of day-care attendance and reported doctor diagnosed worm infestation between the children who have agreed the genotyping and those who did not consent. The distributions of the total IgE level of the children with and without the genotype information were also compared and no significant difference was observed.
In summary, in our large-scale study on two independent birth cohorts, we could confirm the association of a functional FCER1A variant previously reported in adults. In addition, we could demonstrate that the effect is independent from age and environmental influences indicating a basal mechanism.
FCER1A variants exhibit strong effects on total serum IgE levels independently from age and environmental stimuli, which appear to operate already in utero.
Declaration of Funding
The study was funded by grants of the Federal Ministry of Environment, (BMU) (for IUF, FKZ 20462296), and Federal Ministry for Education, Science, Research and Technology (No. 01 EG 9705/2 and 01EG9732).
- 21Maternal inheritance of atopic IgE responsiveness on chromosome 11q. Lancet 1992;15:340., , , , , et al.