Abstract
The interaction of genetic and environmental contributions to immunological traits and their association with atopic disease remain unclear. Flow cytometry and in vitro cytokine responses were used to characterize immune profiles from 93 school-aged twin pairs. Using an established twin pair analytical strategy, the genetic and environmental influences on immunological traits were evaluated, along with their association with atopy. Our findings suggest strong genetic influence on several traits, particularly B cell abundance. In contrast, cytokine responses from in vitro stimulations appeared mainly shaped by environmental exposures. Regarding associations with atopy, greater abundance of both B cells and basophils were observed in atopic individuals as well as increased expression of the immunoglobulin E (IgE) receptor. Genetic influence appeared central to regulating IgE receptor expression on basophils, whereas expression on dendritic cells instead appeared sensitive to environmental exposures. Identifying environmentally regulated immune traits may facilitate the development of targeted therapies to limit the impact of atopic disease in the future.
Genetic and environmental influence on immunological traits is related to allergic disease in young twins.
INTRODUCTION
Atopic sensitization is characterized by induction of T helper 2 (TH2) immune memory, reduced function of regulatory CD4+ T cells (Treg cells) and increased levels of circulating immunoglobulin E (IgE). Further, sensitization involves increased IgE receptor (FcεR1) expression (1–3), heightened sensitivity to allergen exposure and inflammatory mediator release by basophils and mast cells, while also modulating antiviral responses in plasmacytoid dendritic cells (pDCs) (4, 5), potentially linking two major risk patterns that typify asthma development (6).
Atopic sensitization generally arises during the first years of life and is driven by complex interactions between genetic and environmental factors. Genetic factors include variants in ORMDL3/GSMDB, genes regulating innate immune pathways (e.g., CD14, IL13, IL33, TLRs, and STAT6), adaptive immunity (e.g., HLA-DQ), and barrier function (7, 8). Environmental factors include exposure to respiratory viral infections, reduced microbial diversity, and exposure to pollutants and toxins, as reviewed (9–11). The collective impact of these factors on circulating immune cell populations and the extent to which risk of sensitization is influenced by genetic or environmental exposures are not well understood.
Twin pairs provide a unique opportunity to unravel the genetic and environmental determinants of complex conditions, including atopic disease (12). Monozygotic (MZ) twins have largely identical genetics, while dizygotic (DZ) twins share, on average, 50% of their genetic makeup. Thus, a reduction in trait variability between MZ and DZ twin pairs suggests a genetic influence on the trait in question. Twin pair studies have successfully identified the (non)heritability of a diverse range of immune-related disease factors (13, 14). However, many studies focus on adult twins whose local environment and infection histories have diverged. Evaluating young twins, who share a home environment and potential exposures, represents a rare opportunity to evaluate determinants of specific traits while minimizing nonshared environmental influences. Therefore, the relative contribution of genetic and environmental determinants to a range of immunological traits was initially evaluated, followed by assessment of the association between the traits and atopy. The findings may assist in identifying immunological processes sensitive to environmental exposure as targets for future interventions.
RESULTS
Several immune cell subsets show a high degree of genetic influence
In this study, we used a unique cohort of school-aged twin pairs (median age, 10.6; range, 7.2 to 12.9 years) to examine the relative contributions of genetic and environmental factors to immune and atopy-related phenotypes (Fig. 1A). A lower level of twin pair variation in MZ, relative to DZ or unrelated pairs, provides prima facie evidence for a role of genetic variation in immune-related phenotypes, while variation between MZ twins within a pair generally reflects nonshared environmental factors.
Fig. 1. Genetic and environmental influences on immune traits.
(A) Overview of study design. (B) Uniform Manifold Approximation and Projection (UMAP) projection of relevant immune cell subsets including conventional dendritic cells (cDCs) and pDCs. (C) Genetic, environmental, and individual contributions to cell subset abundance variation evaluated using an ACE model. (D) Genetic versus environmental influences on immune cell subsets. (E) Within-pair variation for B cell abundance for MZ, DZ, and unrelated pairs. (F and G) Normalized expression of phenotype markers for relevant cell subset (F) and genetic influence on respective markers (G). UMAP is based on data from all individuals (n = 186). Data are displayed as mean variance contribution in (C), (D), and (G), as individual values in (E), and as normalized expression for each marker in (F). Significance of differences in variation was calculated using a Wilcoxon nonparametric test for differences in variation of cell subset abundance and indicated as ***P < 0.001 either using lines or colored symbols for significant difference compared to unrelated pairs. (A) was rendered using images from National Institutes of Health BioArt.
Abundance of immune cell subsets was determined using flow cytometry. Dimensionality reduction and projection of data using Uniform Manifold Approximation and Projection (UMAP) were used to visualize all cell populations simultaneously and to guide the manual gating strategy on which subsequent analysis is based (Fig. 1B and figs. S1 and S2). Genetic influence was evaluated using the ACE model (15) that evaluates the influence on genetic (A), shared environment (C), and individual variance (E) on a given trait. Adjusting for sex, age, and experimental batch, we identified high (>0.5) genetic influence for abundance of CD4+ and CD4− (presumably CD8+) T cell subsets, B cells, cDC2, and basophils (Fig. 1, C and D). In contrast, the abundance of CD4+ Treg cells, monocytes, and cDC1 appeared to be influenced by the shared environment. To visualize the high genetic influence on B cell abundance, we compared the within-pair variation in MZ and DZ twins to that of randomly reassigned twins into unrelated pairs, showing significantly different variation between MZ and DZ pairs (Fig. 1E). In addition, there was no difference between males and females in DZ pairs that were discordant for sex, confirming limited impact of sex on heritability in this context (fig. S3A).
In evaluating the genetic contribution to cellular phenotype, the expression levels of relevant markers for each cell subset were assessed (Fig. 1F). Low heritability was observed for markers expressed on T cells, such as CD25, FoxP3, and human leukocyte antigen–DR (HLA-DR), whereas the expression of FcεR1 showed considerable heritability on monocytes, cDC2, and basophils. Strong heritability was also observed for expression of HLA-DR on monocytes and pDCs, potentially reflecting baseline activation (Fig. 1G). Comparing level of expression between males and females in DZ pairs discordant for sex suggested that the expression of FcεR1 was elevated on cDC2, pDCs, and basophils in the male compared to the female twin; however, after adjusting for multiple comparisons, the difference was nonsignificant (fig. S3B).
Functional immune responses likely affected by environmental exposures
We next assessed the relative genetic versus environmental contribution to immune function of peripheral blood mononuclear cells (PBMCs) following in vitro stimulation with the representative mitogen phytohemagglutinin (PHA), bacterial lipopolysaccharide (LPS), and house dust mite (HDM) allergen, respectively. Mean cytokine responses from the total cohort are displayed in Fig. 2A, highlighting a strong interleukin-8 (IL-8) response from all three stimulants. Genetic, environmental, and individual influences were evaluated as above using an adjusted ACE model and factor variance displayed for each component (Fig. 2B) as well as on a principal component plot (Fig. 2C). Because of low number of samples positive for LPS-induced IL-5 and HDM-induced IL-18, these were excluded in this analysis. On this basis, cytokine responses appeared largely driven by environmental exposures with genetic influence mainly observed for IFN-γ and IL-1β induced by PHA and HDM, respectively (Fig. 2, B and C). To further evaluate the genetic versus environmental balance, we assessed the number of cytokines with a genetic, environmental, or individual dominance (>0.5 of variance) per stimulant, revealing that while cytokine responses to PHA appeared balanced, LPS responses appeared mostly influenced by shared environmental exposures, whereas HDM responses were dominated by individual exposures (Fig. 2D).
Fig. 2. Environmental impact on functional immune traits.
(A) Mean cytokine levels in cell culture supernatant following PHA, LPS, and HDM stimulation in all individuals (n = 186). (B) Genetic, environmental, and individual influences to cytokine responses following PHA, LPS, or HDM stimulation. NA, not applicable. (C) Principal components analysis (PCA) of factor variance for each cytokine and stimulant. (D) Number of cytokines with a dominant (>0.5 variance explained) by genetic, environmental, or individual factors following PHA, LPS, or HDM stimulation. Data in (A) are displayed as mean of baseline corrected (δ) cytokine levels. Data in (B) are displayed as mean contribution for respective factor.
Genetic influence on immunological traits associated with atopy
We next aimed to assess the contribution of identified genetic and environmental influences on factors associated with atopic sensitization. Across the whole cohort, there were 139 atopic individuals (74.7%). Individuals with atopy were more likely to be male, displayed elevated exhaled nitric oxide, and elevated eosinophil count in whole blood (table S1). Skin prick testing (SPT) responses were used to define atopy where sensitization to allergens such as HDM, grass, cockroach, cat dander, and Alternaria were observed across the cohort (table S1). A total of 41 pairs were discordant for atopy (17 MZ and 24 DZ, of which 58% of the DZ pairs were concordant for sex) (Fig. 3A). The comparison of clinical variables between MZ and DZ twin pairs are displayed in table S2. Of note is the elevated proportion of males as the atopic twin in DZ pairs discordant for atopy. To identify ex vivo immune traits associated with atopy, all atopic and nonatopic individuals were compared. This identified that atopic individuals displayed an elevated proportion of basophils and a trend (adjusted P = 0.08) toward elevated B cell abundance (Fig. 3B). Evaluating surface markers associated with atopy revealed greater expression of FcεR1 and lower expression of HLA-DR across several cell subsets (Fig. 3C). As body mass index (BMI) has been associated with atopy (16), a regression analysis was used to adjust for BMI (fig. S4, A and B). Although BMI was associated with some of the identified traits, it did not affect their association with atopy. To evaluate potential environmental impact on B cell and basophil abundance, we compared MZ and DZ pairs discordant for atopy. Significantly elevated levels of both basophils and B cells were observed in the atopic compared to the nonatopic MZ twins, whereas no difference was observed for DZ pairs (Fig. 3D). No difference in atopic and nonatopic ratios was observed between sex-discordant and -concordant DZ pairs (fig. S5A). Within-pair correlation was also evaluated (fig. S5, B and C), identifying a strong correlation for both B cells and basophils among MZ pairs with no significant correlation in DZ pairs. To assess whether adjusting for atopy status affected the ACE loading for cell abundance across basophils and B cells, we generated two ACE models where atopy was included in the second model as a covariant (Fig. 3E). A limited impact of atopy was observed, confirming the high genetic influence on these two cell subsets. Further, atopy also had limited impact on variance components for cell subsets not associated with atopy (fig. S5D).
Fig. 3. Atopy-related immune traits.
(A) Overview of twin pairs who were concordant versus discordant for atopy. (B) Abundance of immune cell subsets in atopic (n = 139) and nonatopic (n = 47) individuals from the total cohort. (C) Scaled difference in expression of cellular markers in atopic versus nonatopic individuals from total cohort ranked on strength of association. MFI, mean fluorescence intensity. (D) Paired comparison of abundance of basophils and B cells in MZ and DZ pairs discordant for atopy. (E) Genetic, environmental, and individual influences of basophil and B cell abundance in ACE model with or without including atopy as covariate. (F) Significance of paired comparisons of expression of FcεR1α and HLA-DR on relevant cell subsets in MZ versus DZ pairs discordant for atopy. (G) Factor contribution to serum IgE levels. (H) Factor contribution to expression of FcεR1α across relevant cell subsets using ACE models with or without including IgE levels as covariate. Data in (B) are displayed as mean and SD, and those in (C) are displayed as scaled difference between atopic/nonatopic individuals. In (D), individual values are displayed, and in (E), (G), and (H), mean factor contribution is displayed. Significance of difference between groups was calculated using a Mann-Whitney U test for (B) and (C) or Wilcoxon signed-rank test for paired analysis in (D) and (F). Where relevant, P values were adjusted for multiple comparisons. Significance of difference is indicated as P < 0.001; ***P < 0.01; **P < 0.05; * indicates nonsignificant (n.s.).
Assessing the expression of FcεR1 and HLA-DR among pairs who were discordant for atopy in MZ and DZ pairs identified significant differences for FcεR1 expression between atopic and nonatopic twins for both MZ and DZ pairs. The expression of FcεR1 on monocytes was only significantly greater in DZ pairs. No difference in HLA-DR expression was observed within the twin pairs (Fig. 3F). No difference in atopic and nonatopic ratios was observed between sex-discordant and -concordant DZ pairs (fig. S5E). FcεR1 expression on several subsets is closely linked to serum IgE levels (1). Using an ACE model, significant genetic contribution to IgE levels was observed (Fig. 3G). These were also visualized by significant correlation within MZ pairs but not in DZ pairs (fig. S5F). To assess whether adjusting for serum IgE levels affected the ACE loading for FcεR1 expression, we generated two models where IgE was included as a covariant in the second model. A limited impact of IgE was observed for basophils and monocytes. However, including IgE decreased the genetic loading for both cDC2 and pDCs (Fig. 3H). These findings suggest that the expression of FcεR1 on basophils is influenced by genetic factors, whereas the expression on cDC2 and pDCs is mainly affected by environmental exposures.
Atopic effector responses are influenced by environmental and individual variation
In addition to cellular ex vivo phenotypes, we also evaluated the genetic and environmental impact on in vitro allergic effector responses. For this, HDM was used as a model allergen, as 88% of atopic individuals returned a positive SPT for HDM. Our analysis included comparisons of 117 HDM-atopic and 44 nonatopic individuals or 14 MZ and 23 DZ HDM-atopy discordant pairs (Fig. 4A). As allergic effector responses are complex and interconnected, the response to ex vivo exposure to HDM was evaluated using response clusters. These clusters were derived from postculture expression of cellular markers and cytokine production in each individual. In total, nine response clusters were identified on the basis of biologically relevant functions (Fig. 4B and table S3). Clusters 3 and 9 demonstrated significantly greater expression in HDM-atopic versus nonatopic individuals following HDM exposure. These clusters were characterized by TH2 cytokine responses such as IL-13 and IL-5 (cluster 3) as well as tumor necrosis factor (TNF) and IL-10 (cluster 9) (Fig. 4C and table S3). Comparing cluster activity in HDM-atopy concordant and discordant MZ and DZ pairs revealed that cluster 3 was elevated in the HDM-atopic twin compared to the nonatopic twin for both MZ and DZ pairs. No significant difference between discordant pairs was observed for cluster 9 (Fig. 4D). No difference in HDM-atopic and nonatopic ratios was observed between sex-discordant and -concordant DZ pairs (fig. S6A). Within-pair correlation was evaluated for both clusters (fig. S6, B and C), identifying a significant correlation for cluster 3 in DZ pairs, whereas cluster 9 correlated significantly among both MZ and DZ twin pairs. To assess whether adjusting for HDM atopy affected the ACE loading for each cluster, we generated two ACE models as above (Fig. 4E). A limited difference between the two models was observed, and the loadings confirmed high environmental and individual influences on HDM response clusters.
Fig. 4. Environmental impact on atopic effector responses.
(A) Proportion of atopic individuals sensitized to HDM and their distribution in MZ and DZ discordant pairs. (B) Factor correlation of expression of phenotype markers and cytokine production following HDM stimulation in HDM-atopic (n = 117) and nonatopic (n = 44) individuals. (C and D) Comparison of cluster expression in HDM-atopic and nonatopic individuals following HDM stimulation for all clusters (C) or significantly different clusters 3 and 9 (D) and displayed as paired comparisons across the MZ and DZ pairs discordant for HDM atopy. AU, arbitrary units. (E) Factor contribution to cluster expression for clusters 3 and 9 using ACE models with and without including HDM-sensitization as a covariate. Data in (C) are displayed as mean and SD, in (D) as individual values, and in (E) as mean factor contribution. Significance of difference was calculated using the Mann-Whitney U test or the Wilcoxon signed-rank test in (D). Note that comparisons in (D) were not possible in pairs who were concordant for atopy. Where relevant, P values were adjusted for multiple comparisons and indicated as P < 0.001; ***P < 0.01; **P < 0.05; n.s. indicates nonsignificant.
DISCUSSION
The current study leveraged a unique pediatric cohort comprising MZ and DZ twin pairs, selected on the basis of atopic sensitization within the wider Western Australian Twin study of Child Health (WATCH) cohort (17). This enabled analysis with a closely matched control group and the evaluation of the genetic and environmental contributions to pediatric immune profiles. Using this approach, we found that the abundance of several immune cell subsets, including B cells, appeared strongly influenced by genetic background, independent of atopic status. In contrast, cytokine responses to generic stimuli such as LPS appeared mainly shaped by environmental exposures. Immune profiles associated with atopy were influenced by both genetic and environmental factors acting on B cells, basophils, and the expression of the FcεR1. Last, atopic effector responses appeared to have limited influence from genetic factors but were rather shaped by environmental or individual exposures.
Several twin studies have previously described genetic and environmental influences on immune cell subsets (14, 18–20). Many of these were conducted on adult twins likely exposed to separate environments for an extended period of time. In contrast, this study focused on childhood when nonshared environmental exposures are minimal. One of the major findings was that the significant genetic influence on the abundance of B cells and the remarkably similar B cell abundance across MZ but not DZ twin pairs compared to unrelated pairs. This suggests that environmental exposures have a limited impact on B cell abundance during the first 10 years of life, whereas genetic makeup plays a significant role, aligning with previous findings (20). Similar results were also observed for the abundance of cDC2, CD4− T cells (presumably CD8+ T cells), and basophils, although some environmental influence could be attributed to CD4− T cell proportions.
As cytokine responses are shaped by both environmental exposures and host genetics (21–23), we assessed cytokine production patterns in response to three classical immune stimuli (PHA, LPS, and HDM) to evaluate the relative contribution of genetic versus environmental factors in our twin study. Our data suggest that without considering atopy status, cytokine response profiles following LPS stimulation are predominantly influenced by environmental factors, whereas individual factors appear to influence responses to HDM. Mixed influence was observed for PHA responses. However, different relative contribution of gene versus environment influence was apparent when evaluating individual cytokine responses. For example, PHA-induced production of proinflammatory IFN-γ or HDM-induced IL-1β showed evidence of greater genetic influence compared to other stimulus-cytokine combinations, highlighting the context specificity of factors that shape cytokine responses.
Previous studies in twins have suggested a significant heritability in atopic disease (24, 25). In addition, reports on immune profiles in atopic individuals, including studies from our group, demonstrate elevated abundance of CD4+ T cells, B cells, and basophils (26, 27). Elevated abundance of circulating B cells and basophils were also detected in the current study with significantly higher abundance in atopic compared to nonatopic individuals. Comparing discordant MZ pairs also revealed significant differences, which were not observed in DZ pairs. The strong genetic influence and lack of environmental influence on the abundance of B cells and basophils suggest that the difference observed between MZ pairs discordant for atopy may be driven by unique individual exposures that are only evident when most genes are shared.
Evaluating cellular phenotypes that were associated with atopy identified up-regulation of FcεR1 and down-regulation of HLA-DR. While up-regulation of FcεR1 has been demonstrated previously (1, 28, 29), few studies have reported on expression levels of HLA-DR related to genetic versus environmental control in atopic disease. Down-regulation of HLA-DR may suggest reduced baseline antigen presentation in atopic versus nonatopic children in this cohort. Evaluating differences in HLA-DR expression within MZ and DZ pairs did not detect any differences in atopy-discordant twin pairs.
Comparing within-pair differences identified an association between atopic status and FcεR1 expression on pDCs, basophils, and cDC2 in both MZ and DZ pairs, whereas the association for monocytes was only evident in discordant DZ pairs. Notably, the expression of FcεR1 showed high correlation between MZ twin pair members across all cell subsets, and this was largely also true for DZ pairs, suggesting that expression of FcεR1 is sensitive to genetic and environmental exposures. However, the expression of FcεR1 on basophils did not correlate in DZ pairs, suggesting that FcεR1 expression on basophils is largely mediated through genetic control, which was also confirmed by component variance analysis, where serum IgE levels had limited impact. Previously, twin (30) and genome-wide association studies (31) have demonstrated strong genetic control of both serum IgE levels and expression of FcεR1, as well as a linkage between variations in FCER1A, the gene encoding the antibody-binding α subunit of FcεR1 and serum IgE levels (31). The present study confirmed that both serum IgE levels and basophil FcεR1 expression are genetically regulated; however, limited change in the variance composition was observed for expression on basophils when adjusting for IgE levels, suggesting that these may be independently regulated. In the same analysis, large environmental influence was observed for expression of FcεR1 on cDC2 and pDCs, particularly when IgE was included in the model as a covariate.
To assess how children with atopy responded to allergen exposure, HDM was used as a model allergen, as this was the most common allergen of sensitization (88% of the atopic children showing HDM reactivity). Following in vitro cultures, HDM-atopic children displayed elevated levels of TH2 cytokines, including IL-5 and IL-13 as well as TNF and IL-10. Evaluating the genetic and environmental influences on these responses suggested limited genetic influence on atopic effector responses that appeared mainly influenced by environmental or individual factors. These responses may thus be susceptible to modulation through environmental exposures.
The analyses outlined here have the major strengths inherent in a twin study design, namely, the capacity to largely control for genetic and shared environmental factors. In addition, by using twins in childhood, we were able to minimize the impact of nonshared environmental exposures that accumulate over time to our outcomes of interest. Nevertheless, our study also has limitations including a limited variation in age (7.2 to 13.0 years), which may affect the length of time the children are exposed to a shared environment. Older children may also experience hormonal changes, although the impact of sex discordance appeared limited. To minimize the influence on age, sex, and technical variation, these were included as covariates into the statistical models. In addition, the cohort was recruited on the basis of atopic sensitization, and this may skew some of the findings. However, including atopy as a covariate in the models had, in general, limited impact. Last, it is well established that MZ twins have somatic variation at the T cell and B cell receptor repertoires and the heritable influences on immune traits we have observed may be determined by both genetic and/or epigenetic mechanisms.
In conclusion, using a unique cohort of young twin pairs, this study has provided important insight on the genetic influence on immune responses associated with atopy. Notably, our data show a strong genetic influence on the abundance of basophils and their expression of FcεR1, whereas FcεR1 expression on other cell types instead appears sensitive to environmental exposures. This suggests that modulation of some aspects of atopic immune profiles may be possible to reduce the risk of atopy for a given genetic predisposition in the future.
MATERIALS AND METHODS
Study cohorts
A subset of school-age individuals (age range, 7 to 13 years) from the WATCH was established as a population-based twin register to examine the genetics of asthma and attention-deficit/hyperactivity disorder (32). This was extended to include a follow-up subsample of participant families enriched for (but not exclusively) physician-diagnosed asthma and atopic sensitization (33), the focus of the present study. A total of 93 twin pairs (186 individuals) were included in the study, with patient demographics and clinical parameters summarized in Table 1. Although not specifically asked at the time of the recruitment, the study population is likely to reflect the population of Western Australia, where most individuals are of Caucasian ethnicity. The study was approved by the UWA HREC #RA/4/1/6527. All work on human samples was carried out after informed consent was given in accordance with the declaration of Helsinki.
Table 1. Patient demographics in a subset of the WATCH cohort used in this study.
| Total cohort | |
|---|---|
| Number of individuals, n | 186 |
| Number of twin pairs, n | 93 |
| Sex, n males (% males) | 103 (55.3%) |
| Age, median years (range) | 10.6 (7.2–13.0) |
| Zygosity, number of MZ pairs (% MZ pairs) | 45 (48.4%) |
| Sex concordance, number of DZ pairs with same sex (% of DZ pairs) | 27 (56.3%) |
| Total IgE, median IU/ml (range) | 61 (1–747) |
| SPT+, n (%) | 138/184 (75.0%)* |
| Atopy, n (%) | 139 (74.7%) |
| Total PBMC cell count, ×106/ml median (range) | 1.1 (0.6–65.3) |
*Data missing from two individuals.
Clinical parameters
Atopy was defined as either (i) a positive SPT (wheal size ≥ 3 mm) for any of the most common aeroallergens including HDM, cat dander, dog dander, fungi (Alternaria alternata, Aspergillus spp.), cockroach, perennial rye grass, mixed grass, and mixed mould or (ii) a total IgE level above 300 IU/ml as measured by clinical pathology laboratory PathWest in Perth, Australia.
PBMC collection, storage, and thawing
Peripheral blood was collected in sodium heparin tubes during a period without clinical manifestations of atopy or other inflammatory events and processed within 18 hours of collection. PBMCs were isolated using a density gradient, Lymphoprep (Axis-Shield), as per the manufacturer’s instructions. After separation, PBMCs were thoroughly washed with RPMI 1640 supplemented with 2% heat-inactivated fetal bovine serum (Hi-FBS) to deplete platelets and subsequently resuspended in freezing medium (Hi-FBS supplemented with 42.5% RPMI 1640 and 7.5% dimethyl sulfoxide) before freezing in an isopropanol filled container. The frozen samples were subsequently transferred to liquid nitrogen for long-term storage. Frozen PBMCs were thawed in a 37°C water bath, suspended in RPMI 1640 at 4°C, centrifuged and resuspended in RPMI 1640 supplemented with 10% Hi-FBS for further analysis. Cell viability postthaw was assessed using trypan blue exclusion and generally exceeded 80%.
Flow cytometry analysis
Approximately 1 × 106 thawed PBMCs were stained as previously described (34). Briefly, thawed PBMCs were washed using phosphate-buffered saline supplemented with 0.5% bovine serum albumin and 0.1% NaN3 (flow buffer), incubated with the extracellular antibody mixture (table S4) at 4°C for 30 min, washed and incubated in fixation/permeabilized solution for 30 min at 4°C (Thermo Fisher Scientific) followed by intracellular FoxP3-phycoerythrin staining at 4°C for 30 min, and washed and resuspended in flow buffer for flow cytometric data generation. Flow data were acquired using the LSRFortessa II with FACSDiva software (BD Biosciences). Quality control was performed before each cytometry run using CS&T beads (BD Biosciences) to ensure standardized setup and technical performance, as well as rainbow calibration beads (BD Biosciences) to determine alignment and track performance across detection channels and lasers. Processing of samples was block randomized, ensuring that all twin pairs were processed in the same batch to reduce technical variation in the paired analysis. Following acquisition, data were analyzed using FlowJo 10.3 (BD) and CATALYST (35) in R (version 4.2.2). Immune cells were clustered using FlowSOM (36) and displayed using UMAP. All analysis was performed using conventional gating strategies.
Cell culture and cytokine quantification
Thawed PBMCs were resuspended in RPMI 1640 with 10% FBS and cultured at 0.25 × 106 cells per well, in 250 μl, in round-bottom 96-well culture plates. Cells were stimulated with PHA (1 μg/ml; Remel-Thermo), LPS (25 pg/ml; Alexis-Enzo), or HDM (0.5 mg/ml; CSL) and incubated for 24 hours at 37°C and 5% CO2. Following culture, supernatants were collected for cytokine assays and stored at −80°C. From the HDM-stimulated cultures, cells were also washed and prepared for flow cytometry analysis, as above. For evaluation of cytokine content in cell supernatant, a 13-plexed cytokine panel (in-house; table S5) was used and analyzed on a Luminex 100 System. Briefly, postculture supernatants and in-house quality controls (one high and one low made from recombinant human standards) were diluted 1:2 in culture medium. Pooled standards were also diluted in culture medium, and cytokines were quantified as concentration of picograms per milliliter with reference to a standard curve. Antibodies were coupled to MagPlex beads (Luminex Corp.), and the Luminex assay was performed using the manufacturer’s instructions. Standard curves and quantitation were determined using BioPlex Pro Manager software. In-house quality controls (high and low) were included across each plate and accepted if their values fell within 2 SDs of the mean.
Statistical analysis
Student’s t test or the Mann-Whitney U test was used to test for differences between groups, depending on the data distribution; for comparison of three groups, the Kruskal-Wallis test followed by Dunn’s post hoc test was used; for comparison of paired samples, Wilcoxon signed-rank test was used; and for comparison of categorical data between groups, Pearson’s χ2 test was used. Where relevant, P values were adjusted for multiple comparisons using the Benjamin-Hochberg procedure. Linear regression was also used to evaluate correlation between atopic and nonatopic twin pairs enabling adjustment for serum IgE levels. Allergen response clusters were generated using the variable clustering algorithm (JMP17, JMP Statistical Discovery LLC). Within-pair variation was determined as the absolute difference in cell abundance or marker expression; these were subsequently normalized to the mean abundance/expression for each cell subset. Unrelated pairs were generated to evaluate the effect of a shared environment. For this, all individuals were randomly assigned to a random, unrelated, pair. This was done for two age blocks to minimize age-related effects yet ensuring random distribution of sex, atopic status, and original zygosity. Evaluation of genetic, environmental, or individual factors contributing to a given trait was evaluated using an ACE model. The ACE model was run through the UMX package (37) in R, using the umxACE() function with age, sex, and technical batches included as covariates. With exception for the ACE modeling, all statistical analysis was performed using JMP 17 (JMP Statistical Discovery LLC), and data were visualized using Prism 9.5.0 (GraphPad).
Acknowledgments
We would like to acknowledge all study participants and families involved in the study, their treating physicians and technicians involved in sample processing, N. de Klerk (The Kids Research Institute Australia) for contributing to setting up the WATCH study, and M. Cooper (The Kids Research Institute Australia) for statistical advice.
Funding: The study was funded by the Australian National Health and Medical Research Council (APP1063394), The University of Western Australia, and the Kids Research Institute Australia. J.L. was supported by fellowships from the WA Future Health Research and Innovation fund as well as the Stan Perron Charitable Foundation. A.B. was supported by a Fellowship from the Simon Lee Foundation, and D.H.S. was supported by a fellowship from the Kids Research Institute Australia.
Author contributions: Conceptualization: A.B., P.G.H., R.S., and D.H.S. Funding acquisition: D.H.S. and R.S. Methodology: D.M., M.S., A.B., R.S., and D.H.S. Resources: D.M., P.G.H., R.S., and D.H.S. Validation: D.M., M.S., and D.H.S. Investigation: J.L., J.F.R., D.M., M.S., P.G.H., and R.S. Data curation: J.L., J.F.R., R.S., and D.H.S. Software: J.F.R. Formal analysis: J.L., J.F.R., and R.S. Visualization: J.L., J.F.R., and R.S. Supervision: A.B., P.G.H., R.S., and D.H.S. Project administration: D.H.S. Writing—original draft: J.L., J.F.R., and R.S. Writing—review and editing: J.L., J.F.R., D.M., P.G.H., R.S., and D.H.S.
Competing interests: A.B. and J.F.R. are cofounders, equity holders, and directors of the startup company Respiradigm Pty Ltd. (and subsidiary First Breath Health Pty Ltd) that is unrelated to this work. A.B. is the founder of the startup company INSiGENe Pty Ltd. that is unrelated to this work. The other authors declare that they have no competing interests.
Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Datasets supporting the conclusions of this article are available at Datadryad.org (https://doi.org/10.5061/dryad.0gb5mkmch).
Supplementary Materials
This PDF file includes:
Figs. S1 to S6
Tables S1 to S5
REFERENCES AND NOTES
- 1.Leffler J., Read J. F., Jones A. C., Mok D., Hollams E. M., Laing I. A., Le Souef P. N., Sly P. D., Kusel M. M. H., de Klerk N. H., Bosco A., Holt P. G., Strickland D. H., Progressive increase of FcεRI expression across several PBMC subsets is associated with atopy and atopic asthma within school-aged children. Pediatr. Allergy Immunol. 30, 646–653 (2019). [DOI] [PubMed] [Google Scholar]
- 2.Shin J.-S., Greer A. M., The role of FcεRI expressed in dendritic cells and monocytes. Cell. Mol. Life Sci. 72, 2349–2360 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Greer A. M., Wu N., Putnam A. L., Woodruff P. G., Wolters P., Kinet J.-P., Shin J.-S., Serum IgE clearance is facilitated by human FcεRI internalization. J. Clin. Invest. 124, 1187–1198 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gill M. A., Bajwa G., George T. A., Dong C. C., Dougherty I. I., Jiang N., Gan V. N., Gruchalla R. S., Counterregulation between the FcεRI pathway and antiviral responses in human plasmacytoid dendritic cells. J. Immunol. 184, 5999–6006 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Coenen I., de Jong E., Jones A. C., Khoo S.-K., Foo S., Howland S. W., Ginhoux F., Le Souëf P. N., Holt P. G., Strickland D. H., Laing I. A., Leffler J., Impaired interferon response in plasmacytoid dendritic cells from children with persistent wheeze. J. Allergy Clin. Immunol. 153, 1083–1094 (2024). [DOI] [PubMed] [Google Scholar]
- 6.Holt P. G., Sly P. D., Viral infections and atopy in asthma pathogenesis: New rationales for asthma prevention and treatment. Nat. Med. 18, 726–735 (2012). [DOI] [PubMed] [Google Scholar]
- 7.Vercelli D., Discovering susceptibility genes for asthma and allergy. Nat. Rev. Immunol. 8, 169–182 (2008). [DOI] [PubMed] [Google Scholar]
- 8.Hinds D. A., McMahon G., Kiefer A. K., Do C. B., Eriksson N., Evans D. M., St Pourcain B., Ring S. M., Mountain J. L., Francke U., Davey-Smith G., Timpson N. J., Tung J. Y., A genome-wide association meta-analysis of self-reported allergy identifies shared and allergy-specific susceptibility loci. Nat. Genet. 45, 907–911 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Tang H. H. F., Teo S. M., Sly P. D., Holt P. G., Inouye M., The intersect of genetics, environment, and microbiota in asthma-perspectives and challenges. J. Allergy Clin. Immunol. 147, 781–793 (2021). [DOI] [PubMed] [Google Scholar]
- 10.Chatkin J., Correa L., Santos U., External environmental pollution as a risk factor for asthma. Clin. Rev. Allergy Immunol. 62, 72–89 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Renz H., Holt P. G., Inouye M., Logan A. C., Prescott S. L., Sly P. D., An exposome perspective: Early-life events and immune development in a changing world. J. Allergy Clin. Immunol. 140, 24–40 (2017). [DOI] [PubMed] [Google Scholar]
- 12.Khan S. J., Dharmage S. C., Matheson M. C., Gurrin L. C., Is the atopic march related to confounding by genetics and early-life environment? A systematic review of sibship and twin data. Allergy 73, 17–28 (2018). [DOI] [PubMed] [Google Scholar]
- 13.Ingelfinger F., Gerdes L. A., Kavaka V., Krishnarajah S., Friebel E., Galli E., Zwicky P., Furrer R., Peukert C., Dutertre C.-A., Eglseer K. M., Ginhoux F., Flierl-Hecht A., Kümpfel T., De Feo D., Schreiner B., Mundt S., Kerschensteiner M., Hohlfeld R., Beltrán E., Becher B., Twin study reveals non-heritable immune perturbations in multiple sclerosis. Nature 603, 152–158 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Brodin P., Jojic V., Gao T., Bhattacharya S., Angel C. J. L., Furman D., Shen-Orr S., Dekker C. L., Swan G. E., Butte A. J., Maecker H. T., Davis M. M., Variation in the human immune system is largely driven by non-heritable influences. Cell 160, 37–47 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Eaves L. J., Last K. A., Young P. A., Martin N. G., Model-fitting approaches to the analysis of human behaviour. Heredity (Edinb) 41, 249–320 (1978). [DOI] [PubMed] [Google Scholar]
- 16.Bapat S. P., Whitty C., Mowery C. T., Liang Y., Yoo A., Jiang Z., Peters M. C., Zhang L. J., Vogel I., Zhou C., Nguyen V. Q., Li Z., Chang C., Zhu W. S., Hastie A. T., He H., Ren X., Qiu W., Gayer S. G., Liu C., Choi E. J., Fassett M., Cohen J. N., Sturgill J. L., Crotty Alexander L. E., Suh J. M., Liddle C., Atkins A. R., Yu R. T., Downes M., Liu S., Nikolajczyk B. S., Lee I. K., Guttman-Yassky E., Ansel K. M., Woodruff P. G., Fahy J. V., Sheppard D., Gallo R. L., Ye C. J., Evans R. M., Zheng Y., Marson A., Obesity alters pathology and treatment response in inflammatory disease. Nature 604, 337–342 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hansen J., de Klerk N., Croft M., Alessandri P., Burton P., The Western Australian Twin Child Health (WATCH) study: Work in progress. Australas. Epidemiol. 7, 16–20 (2000). [Google Scholar]
- 18.Silva S. L., Fonseca M., Pereira M. L. M., Silva S. P., Barbosa R. R., Serra-Caetano A., Blanco E., Rosmaninho P., Pérez-Andrés M., Sousa A. B., Raposo A. A. S. F., Gama-Carvalho M., Victorino R. M. M., Hammarstrom L., Sousa A. E., Monozygotic twins concordant for common variable immunodeficiency: Strikingly similar clinical and immune profile associated with a polygenic burden. Front. Immunol. 10, 2503 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yan Z., Maecker H. T., Brodin P., Nygaard U. C., Lyu S. C., Davis M. M., Nadeau K. C., Andorf S., Aging and CMV discordance are associated with increased immune diversity between monozygotic twins. Immun. Ageing 18, 5 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Evans D. M., Frazer I. H., Martin N. G., Genetic and environmental causes of variation in basal levels of blood cells. Twin Res. 2, 250–257 (1999). [DOI] [PubMed] [Google Scholar]
- 21.MacGillivray D. M., Kollmann T. R., The role of environmental factors in modulating immune responses in early life. Front. Immunol. 5, 434 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Ter Horst R., Jaeger M., Smeekens S. P., Oosting M., Swertz M. A., Li Y., Kumar V., Diavatopoulos D. A., Jansen A. F. M., Lemmers H., Toenhake-Dijkstra H., van Herwaarden A. E., Janssen M., van der Molen R. G., Joosten I., Sweep F. C. G. J., Smit J. W., Netea-Maier R. T., Koenders M. J. F., Xavier R. J., van der Meer J. W. M., Dinarello C. A., Pavelka N., Wijmenga C., Notebaart R. A., Joosten L. A. B., Netea M. G., Host and environmental factors influencing individual human cytokine responses. Cell 167, 1111–1124.e13 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Li Y., Oosting M., Smeekens S. P., Jaeger M., Aguirre-Gamboa R., Le K. T. T., Deelen P., Ricaño-Ponce I., Schoffelen T., Jansen A. F. M., Swertz M. A., Withoff S., van de Vosse E., van Deuren M., van de Veerdonk F., Zhernakova A., van der Meer J. W. M., Xavier R. J., Franke L., Joosten L. A. B., Wijmenga C., Kumar V., Netea M. G., A Functional genomics approach to understand variation in cytokine production in humans. Cell 167, 1099–1110.e14 (2016). [DOI] [PubMed] [Google Scholar]
- 24.Hopp R. J., Bewtra A. K., Watt G. D., Nair N. M., Townley R. G., Genetic analysis of allergic disease in twins. J. Allergy Clin. Immunol. 73, 265–270 (1984). [DOI] [PubMed] [Google Scholar]
- 25.Strachan D. P., Wong H. J., Spector T. D., Concordance and interrelationship of atopic diseases and markers of allergic sensitization among adult female twins. J. Allergy Clin. Immunol. 108, 901–907 (2001). [DOI] [PubMed] [Google Scholar]
- 26.Gamez C., Leffler J., Clark S., Corscadden K., Prescott S. L., Palmer D. J., Strickland D., Egg-sensitised infants have elevated CD4+ effector memory T regulatory cells from birth. Clin. Exp. Allergy 54, 34–45 (2024). [DOI] [PubMed] [Google Scholar]
- 27.Reilly K. M., Yap P. L., Dawes J., Barnetson R. S., MacKenzie F., Allan T. L., Circulating basophil counts in atopic individuals. Int. Arch. Allergy Appl. Immunol. 84, 424–426 (1987). [DOI] [PubMed] [Google Scholar]
- 28.Dehlink E., Baker A. H., Yen E., Nurko S., Fiebiger E., Relationships between levels of serum IgE, cell-bound IgE, and IgE-receptors on peripheral blood cells in a pediatric population. PLOS ONE 5, e12204 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Berings M., Gevaert P., De Ruyck N., Derycke L., Holtappels G., Pilette C., Bachert C., Lambrecht B. N., Dullaers M., FcεRI expression and IgE binding by dendritic cells and basophils in allergic rhinitis and upon allergen immunotherapy. Clin. Exp. Allergy 48, 970–980 (2018). [DOI] [PubMed] [Google Scholar]
- 30.Jacobsen H. P., Herskind A. M., Nielsen B. W., Husby S., IgE in unselected like-sexed monozygotic and dizygotic twins at birth and at 6 to 9 years of age: High but dissimilar genetic influence on IgE levels. J. Allergy Clin. Immunol. 107, 659–663 (2001). [DOI] [PubMed] [Google Scholar]
- 31.Weidinger S., Gieger C., Rodriguez E., Baurecht H., Mempel M., Klopp N., Gohlke H., Wagenpfeil S., Ollert M., Ring J., Behrendt H., Heinrich J., Novak N., Bieber T., Krämer U., Berdel D., von Berg A., Bauer C. P., Herbarth O., Koletzko S., Prokisch H., Mehta D., Meitinger T., Depner M., von Mutius E., Liang L., Moffatt M., Cookson W., Kabesch M., Wichmann H.-E., Illig T., Genome-wide scan on total serum IgE levels identifies FCER1A as novel susceptibility locus. PLOS Genet. 4, e1000166 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hansen J., Alessandri P. T., Croft M. L., Burton P. R., de Klerk N. H., The Western Australian register of childhood multiples: Effects of questionnaire design and follow-up protocol on response rates and representativeness. Twin Res. 7, 149–161 (2004). [DOI] [PubMed] [Google Scholar]
- 33.McClenaghan J., Warrington N. M., Jamrozik E. F., Hui J., Beilby J. P., Hansen J., de Klerk N. H., James A. L., Musk A. W., Palmer L. J., The PHF11 gene is not associated with asthma or asthma phenotypes in two independent populations. Thorax 64, 620–625 (2009). [DOI] [PubMed] [Google Scholar]
- 34.Leffler J., Jones A. C., Hollams E. M., Prastanti F., Le Souëf P. N., Holt P. G., Bosco A., Laing I. A., Strickland D. H., Basophil counts in PBMC populations during childhood acute wheeze/asthma are associated with future exacerbations. J. Allergy Clin. Immunol. 142, 1639–1641.e5 (2018). [DOI] [PubMed] [Google Scholar]
- 35.Nowicka M., Krieg C., Crowell H. L., Weber L. M., Hartmann F. J., Guglietta S., Becher B., Levesque M. P., Robinson M. D., CyTOF workflow: Differential discovery in high-throughput high-dimensional cytometry datasets. F1000Res 6, 748 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Van Gassen S., Callebaut B., Van Helden M. J., Lambrecht B. N., Demeester P., Dhaene T., Saeys Y., FlowSOM: Using self-organizing maps for visualization and interpretation of cytometry data. Cytometry A 87, 636–645 (2015). [DOI] [PubMed] [Google Scholar]
- 37.Bates T. C., Maes H., Neale M. C., umx: Twin and path-based structural equation modeling in R. Twin Res. Hum. Genet. 22, 27–41 (2019). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figs. S1 to S6
Tables S1 to S5