Th1 and Th2 cytokines secreted by polarized effector T cells play a pivotal role in the development of autoimmune and allergic diseases. However, the genetic basis of cytokine production by T lymphocytes in humans is poorly understood. In this study, we investigated the genetic contribution to cytokine production and regulation of T cell-specific transcription factors in a prospective twin study. We found a substantial genetic contribution to the production of Th1 cytokines such as IFN-γ and TNF-α with heritabilities of 0.85 (95% confidence intervals, 0.74–0.95) and 0.72 (0.50–0.93), respectively, whereas no genetic influence on production of the Th2 signature cytokine IL-4 was observed. Furthermore, the intrapair variability in IFN-γ production by isolated T cells was lower in monozygotic than in dizygotic twins. In contrast to GATA-3, NFAT, and NF-κB, intrapair variability of T-bet, the master transcription factor of Th1 cells, was very low among monozygotic and high among dizygotic twins, indicative of a strong genetic influence on T-bet (heritability 0.93, 95% confidence interval, 0.84–1.0). Our data provide novel insights into the genetic regulation of human Th cell polarization. These data suggest that signature cytokines and cytokine signaling events of Th1 rather than Th2 cells are genetically determined and implicate that Th2-associated diseases in humans might be due to genetic variations in Th1 cytokine regulation via T-bet. This concept is highlighted by the recent finding that inactivation of the T-bet gene in mice results in development of clinical hallmark features of asthma.

In developed countries there has been a steady rise in the incidence of autoimmune and allergic diseases during the last 30 years and a remarkable decrease in the incidence of infectious disorders (1). How infectious agents protect against allergic disorders is unclear, but a number of observations point toward a crucial role of CD4-positive Th cells and T cell-derived cytokines. The development of most autoimmune and allergic diseases depends on T lymphocytes and their cytokines, which play a key role in orchestrating immune responses. Naive Th cells can differentiate into different subsets, each with distinct functions and cytokine profiles (2). Th1 cells are characterized by the secretion of IL-2, IFN-γ, and TNF. Only recently it has become clear that IFN-γ production is under tight control of T-bet, a Th1-specific T box transcription factor, which suppresses early Th2 cytokine production and induces Th1 T cell development via up-regulation of IFN-γ gene transcription and IL-12Rb2 chain expression (3). Th2 cells preferentially secrete IL-4, IL-5, and IL-13 and their differentiation is dependent on induction of the transcription factors STAT6, c-maf, and GATA-3 after stimulation of the IL-4R. In STAT6-deficient T cells, GATA-3 can fully reconstitute Th2 cell development, suggesting that GATA-3 is the key transcription factor for Th2 differentiation (4). However, the reciprocal down-regulation of Th1 cell polarization by Th2 cytokines and of Th2 cell polarization by Th1 cytokines raises the possibility that these cytokines are critical in infection-mediated protection against allergy and autoimmunity.

Maintaining the balance between Th1 and Th2 cytokine production is of critical importance for the immune system, since overproduction of Th1 or Th2 cytokines may result in the development of chronic inflammatory diseases of various organs including the lung, skin, colon, and CNS (2, 5). However, at least in mice, Th cell polarization and subsequent cytokine production are strongly determined by genetic factors. This is best exemplified by the observation that the genetic background of mice strongly modulates Th cell polarization (6). For instance, in murine Leishmania major infections, the effects of the genetic background on Th cell polarization reside within the T cell and not within the APC compartment and control susceptibility of mice to leishmaniasis (6). In contrast to the murine system, the genetic basis of human Th cell polarization is poorly understood, however.

The aim of the present study was to apply a classical twin study approach to define the genetic contribution to Th cell polarization in humans. In this study, we focused on the signature cytokines of polarized effector T cells (TNF-α, IFN-γ, and IL-4) and key transcription factors controlling Th1/Th2 differentiation (T-bet, NF-κB, NFAT, and GATA-3). Such twin studies allow the separation of genetic and environmental components because monozygotic (MZ)3 twins are genetically identical and dizygotic (DZ) twins share on average 50% of their genes. We found low variability of intrapair differences in MZ twins compared with DZ twins for IFN-γ, TNF, and T-bet, indicative of a strong genetic component in the expression of these Th1-associated proteins. There was little or no evidence for a genetic effect on Th2 differentiation, however. These data suggest that the differentiation of Th1 cells is under the control of cytokines and transcription factors with mainly genetically determined activity, whereas Th2 cytokine expression appears to be mainly dependent on environmental factors.

We examined 37 same-sex DZ (26 males, 11 females) and 69 MZ (29 males, 40 females) German twin pairs of Caucasian origin, who were part of a larger twin group that had been recruited for a prospective vaccination study (for further details see Ref.7). The study was approved by the Ethics Committee of the local medical board. After informed consent, twins were personally interviewed and asked for a blood sample prior to vaccination. Twin partners were recruited on the same date and all subsequent analyses were performed in parallel using the same batches of reagents. Interview data included: smoking, alcohol consumption, height, weight, and medical history. None of the twins took any immunosuppressive drugs such as corticosteroids. Zygosity of twin pairs was determined by typing 15 microsatellite loci using the Gene Print Powerplex 16 System kit from Promega. Reactions were analyzed on an ABI 310 sequencer (Applied Biosystems).

Cytokine determinations were done in a whole blood stimulation system as previously described (8). Whole blood samples of 10 ml were drawn using heparinized, endotoxin-free collecting tubes. The blood was diluted 1/1 with RPMI 1640 (no additives). Cells were stimulated with different final concentrations of mitogens (TNF-α: PMA/ionomycin, 25 ng/ml PMA plus 0.5 μg/ml ionomycin; IFN-γ and IL-4, 1 μg/ml PHA; Sigma-Aldrich). Stimulation was performed for 24 h at 37°C in 5% CO2. Plasma supernatants were stored at −70°C before being analyzed.

Cytokine ELISAs were performed according to standard protocols using coating and detection Abs of manufacturers as cited below and enzyme-linked Abs supplied by DakoCytomation. IL-4 ELISAs were performed using BD Pharmingen test systems, TNF-α using R&D Systems Abs, and IFN-γ using Endogen reagents. All samples were analyzed in duplicate, and cytokine concentrations were calculated by reference to a standard curve. A white blood differential cell count was performed the day blood was obtained. Cytokine concentrations were normalized for cell numbers of specific cytokine-producing cells depending on the applied stimulus. PHA-stimulated cytokine production (IL-4 and IFN-γ) was normalized for the number of lymphocytes, PMA/inonomycin stimulation was corrected for the number of lymphocytes plus monocytes (TNF-α).

To determine expression of transcription factors, human PBMC from MZ or DZ twins were isolated using Ficoll-Hypaque gradients. Cell cultures were performed in complete medium consisting of RPMI 1640 (Biochrom) supplemented with 2 mM l-glutamine (BioWhittaker), 100 U/ml each of penicillin and streptomycin (BioWhittaker), and 5% heat-inactivated human serum (PAA Laboratories). T lymphocytes were separated using anti-CD3-coated magnetic beads (Dynabeads; Dynal) and subsequently stimulated in cell culture for 5 h with coated Abs to CD3 (HIT3a, 0.04 μg/ml; BD Pharmingen) and soluble CD28 Abs (1 μg/ml; BD Pharmingen). In addition, rIL-2 (R&D Systems) was used at a final concentration of 40 U/ml in all experiments.

To determine cytokine production by T cells, T lymphocytes were isolated as described above and stimulated in complete RPMI 1640 medium for 3 days with coated Abs to CD3, soluble CD28 Abs (1 mg/ml), and IL-2 (40 U/ml). In some experiments, T cells were cultured in 24-well plates with autologous APCs (isolated using CD14/CD19 immunomagnetic MACS beads) pulsed with 5 μg/ml TT (donated by Dr. W. Böcher, University of Mainz, Mainz, Germany) for 3 days. After 3 days, culture supernatants were removed and assayed for cytokine concentration by ELISA as specified above.

Cells were washed twice in cold PBS and resuspended in 500 μl of buffer A (10 mM HEPES, 1.5 mM MgCl2, 19 mM KCl, 0.5 mM PMSF, and 1 mM DTT) followed by addition of 20 μl of Triton X-100 (Sigma-Aldrich) and incubation on ice for 5 min. Cells were centrifuged for 10 min at 4°C followed by the resuspension of the nuclear pellet in buffer C (20 mM HEPES, 1.5 mM MgCl2, 0.2 mM EDTA, 25% glycerol, 0.5 mM PMSF, 1 mM DTT). Finally, the nuclei were homogenized using ministir bars for 1 h at 4°C followed by centrifugation at 15,000 rpm for 15 min at 4°C. The concentration of nuclear proteins in supernatants was assessed using a Bio-Rad system.

Equal amounts of nuclear extract (15 μg) were added to 8 μl of electrophoresis sample buffer (Roth). After boiling, the extracts were loaded on 10% SDS-PAGE gels and electrophoretically separated. Proteins were transferred to nitrocellulose membranes and detection was performed with the ECL Western blotting analysis system (Amersham Biosciences). Western blots were made using the following Abs: anti-human GATA-3, anti-human NF-κB p65 (1/500 dilution; Santa Cruz Biotechnology), anti-T-bet (1/1000 dilution; donated by Prof. L. Glimcher, Harvard School of Public Health, Boston, MA), and HRP-linked anti-mouse, anti-rabbit, or anti-goat Ig (1/1000 dilution; Amersham Biosciences). Densitometry of Western blots was performed using the ChemiImager 5500 software (Alpha Innotech).

EMSA was performed as previously described (9). Double-stranded oligonucleotides containing consensus binding sites for EMSA were obtained from Santa Cruz Biotechnology. The sequences were as follows: GATA-3, 5′-CAC TTG ATA ACA GAA AGT GAT AAC TCT-3′ and 5′-AGA GAA ATC ACT TTC TGT TAT CAA GTG-3′ and NFAT, 5′-CGC CCA AAG AGG AAA ATT TGT TTC ATA-3′ and 5′-TAT GAA ACA AAT TTT CCT CTT TGG GCG-3′. Oligonucleotides were end-labeled with [γ-32P]ATP (>5000 Ci/mmol; Amersham Biosciences) using T4 polynucleotide kinase (New England Biolabs). Radiolabeled DNA was added to the binding reaction that also contained 1 μg synthetic DNA duplex of poly(dI:dC) (Pharmacia), 15 μg nuclear proteins and binding buffer (25 mM HEPES (pH 7.5), 150 mM KCl, 5 mM DTT, and 10% glycerol). Complex formation was allowed to proceed for 30 min at room temperature. Finally, the complexes were separated from unbound DNA by native PAGE on 6% gels. The gels were exposed to Kodak MS films on intensifying screens at −80°C. Gels were scanned and analyzed by densitometry. The higher densitometric values among twins received an arbitrary value of 100. The value of the corresponding twin was then directly compared with this value of 100 (percent expression level as compared with the other twin). A value of 100 for the other twin therefore means identical expression between twins.

We log transformed the cytokine concentrations for the statistical analysis to obtain approximately normally distributed measures. Linear models were fit to assess the impact of body mass index (BMI), age, and gender. These confounders were first assessed separately (single factor model) and in a second analysis together (full model). To adequately adjust for the dependence of the two measurements in one pair, generalized estimation equations (proc genmod in SAS) (10) were used. Heritability was estimated based on intrapair variance (heritability = phenotypic variance within pairs (VWP) (DZ) − VWP (MZ)/VWP (DZ) (11). Heritability of IFN-γ, IL-4, and TNF-α was also calculated and corrected for the influence of sex, age, and BMI. The confidence limits were calculated using empirical SE estimates. The heritability estimates and the linear models were calculated with SAS 8.01 (SAS Institute).

A total of 69 MZ (29 men, 40 women) and 37 same-sex DZ twin pairs (26 men, 11 women) participated in the study. Mean age was 35.3 years (median, 31 years; range, 18–67 years) for MZ and 42.1 years (median, 39 years; range, 20–65 years) for DZ twin pairs. The mean BMI was 23.9 kg/m2 (median, 23.3; range, 15.7–36.3) for MZ and 25.6 kg/m2 (median, 24.5; range, 18.0–37.5) for DZ twins. Forty-three MZ twins (31%) were active smokers compared with 20 DZ twins (27%).

To determine the genetic impact on TNF-α, IFN-γ, and IL-4 production, we compared cytokine levels after stimulation of whole blood of MZ and same-sex DZ twins. The whole blood cell culture method has proven to be reliable (8, 12) and represents a more physiological situation than examination of isolated PBMC.

We found an influence of gender on IFN-γ and IL-4 production. Male individuals produced more IFN-γ and IL-4 than female twins. In addition, TNF-α production was significantly increased in individuals with a higher BMI (Table I).

Table I.

Influence of gender, age, and BMI on production of IFN-γ, TNF-α, and IL-4

CytokinePredictive VariableEstimated β Coefficientsa95% CIapa
IFN-γ Male 0.311/0.381 0.030–0.591/0.117–0.646 0.030/0.005 
 BMI 0.023/0.033 −0.014–0.059/0.005–0.062 0.225/0.022 
 Age (years) −0.001/0.005 −0.011–0.009/−0.005–0.014 0.867/0.312 
TNF-α Male 0.077/−0.127 −0.443–0.597/−0.586–0.333 0.772/0.590 
 BMI −0.065/−0.060 −0.120–−0.010/−0.106–−0.012 0.020/0.013 
 Age (years) 0.004/−0.004 −0.011–0.020/−0.020–0.012 0.586/0.610 
IL-4 Male 0.376/0.412 −0.022–0.773/0.049–0.776 0.064/0.026 
 BMI −0.007/0.019 −0.056–0.043/−0.022–0.059 0.790/0.374 
 Age (years) −0.011/0.013 −0.003–0.024/0.001–0.025 0.122/0.042 
CytokinePredictive VariableEstimated β Coefficientsa95% CIapa
IFN-γ Male 0.311/0.381 0.030–0.591/0.117–0.646 0.030/0.005 
 BMI 0.023/0.033 −0.014–0.059/0.005–0.062 0.225/0.022 
 Age (years) −0.001/0.005 −0.011–0.009/−0.005–0.014 0.867/0.312 
TNF-α Male 0.077/−0.127 −0.443–0.597/−0.586–0.333 0.772/0.590 
 BMI −0.065/−0.060 −0.120–−0.010/−0.106–−0.012 0.020/0.013 
 Age (years) 0.004/−0.004 −0.011–0.020/−0.020–0.012 0.586/0.610 
IL-4 Male 0.376/0.412 −0.022–0.773/0.049–0.776 0.064/0.026 
 BMI −0.007/0.019 −0.056–0.043/−0.022–0.059 0.790/0.374 
 Age (years) −0.011/0.013 −0.003–0.024/0.001–0.025 0.122/0.042 
a

Coefficients, 95% CI, and p are given for full/single factor model.

Within pair variances for IFN-γ and TNF-α showed striking differences between MZ and DZ twins (Table II). After correction for the influences of gender, age and BMI heritabilities were 0.84 (95% confidence interval (CI), 0.73–0.95) and 0.69 (95% CI, 0.46–0.93) for IFN-γ and TNF-α, respectively.

Table II.

Heritabilities of cytokine productiona

CytokineZygosityNo. of Twin PairsConcentration (pg/ml) Geometric Mean (95% CI)Within-Pair VarianceHeritability (95% CI)
IL-4 MZ 47 90.8 (75.8–108.8) 0.47 0.37 (−0.15–0.89) 
 DZ 26 86.3 (62.6–118.9) 0.75  
TNF-α MZ 54 384.2 (305.1–484.0) 0.28 0.72 (0.50–0.93) 
 DZ 29 397.5 (229.2–540.7) 1.00  
IFN-γ MZ 58 590.5 (518.6–672.5) 0.12 0.85 (0.74–0.95) 
 DZ 35 593.7 (476.3–739.9) 0.80  
CytokineZygosityNo. of Twin PairsConcentration (pg/ml) Geometric Mean (95% CI)Within-Pair VarianceHeritability (95% CI)
IL-4 MZ 47 90.8 (75.8–108.8) 0.47 0.37 (−0.15–0.89) 
 DZ 26 86.3 (62.6–118.9) 0.75  
TNF-α MZ 54 384.2 (305.1–484.0) 0.28 0.72 (0.50–0.93) 
 DZ 29 397.5 (229.2–540.7) 1.00  
IFN-γ MZ 58 590.5 (518.6–672.5) 0.12 0.85 (0.74–0.95) 
 DZ 35 593.7 (476.3–739.9) 0.80  
a

Within-pair variances and heritabilities of IL-4, TNF-α, and IFN-γ production in MZ and DZ twins after stimulation of whole blood cultures (PHA stimulation for IFN-γ and IL-4; PMA/ionomycin for TNF-α; not corrected for age, sex, and BMI). Cytokine determinations were done in a whole blood stimulation system as previously described (8 ).

For IL-4, we obtained more identical correlation coefficients for MZ and DZ twins. Correspondingly, a 95% CI of heritability (−0.15–0.89) including zero points to a very weak or no genetic influence on the observed intrapair variability.

Because the distribution of cell types in the above system using whole blood may influence the results of cytokine assays, we aimed in subsequent studies at characterizing the production of cytokines by purified T lymphocytes. We compared IFN-γ production by anti-CD3 plus anti-CD28-stimulated T cells from 8 MZ and 10 DZ twins. Intrapair similarities in IFN-γ production were significantly (p < 0.05) higher in MZ as compared with DZ twins, suggesting a strong genetic influence. In contrast, no such effect was seen for IL-4 production (Fig. 1). Stimulation of peripheral blood T cells with TT also showed higher intrapair concordance of IFN-γ production in MZ pairs as compared with DZ pairs, whereas very little or no IL-4 production could be found in both MZ and DZ twins upon TT stimulation (<20 pg/ml).

FIGURE 1.

IFN-γ and IL-4 production by purified T lymphocytes. Peripheral blood T lymphocytes were isolated from PBMC of MZ and DZ twins and stimulated with anti-CD3/anti-CD28 Abs (A) or TT (B). Culture supernatants were taken after 3 days and analyzed for cytokine content (IFN-γ, IL-4) by specific ELISA. Values are expressed as percent similarity of intrapair cytokine production, where a value of 100 would mean identical expression within a twin pair. Data represent mean values ± SEM from 6 to 10 patients per group in A and B. There was a significantly (p < 0.05) higher similarity of IFN-γ but not IL-4 production upon anti-CD3/CD28 stimulation in MZ as compared with DZ twins.

FIGURE 1.

IFN-γ and IL-4 production by purified T lymphocytes. Peripheral blood T lymphocytes were isolated from PBMC of MZ and DZ twins and stimulated with anti-CD3/anti-CD28 Abs (A) or TT (B). Culture supernatants were taken after 3 days and analyzed for cytokine content (IFN-γ, IL-4) by specific ELISA. Values are expressed as percent similarity of intrapair cytokine production, where a value of 100 would mean identical expression within a twin pair. Data represent mean values ± SEM from 6 to 10 patients per group in A and B. There was a significantly (p < 0.05) higher similarity of IFN-γ but not IL-4 production upon anti-CD3/CD28 stimulation in MZ as compared with DZ twins.

Close modal

Twenty-one MZ and 16 DZ twin pairs agreed to participate in a second simultaneous blood sampling for the investigation of transcription factor regulation. Results of the EMSAs and Western blot analyses are shown in Fig. 2. The corresponding within-pair variabilities and calculated heritabilities are shown in Table III. Intrapair variabilities in MZ twins for GATA-3, NFAT, and NF-κB expression were higher than or similar to the variabilities observed in DZ twin pairs, thus excluding a significant genetic contribution to their regulation. In contrast, for T-bet, the master transcription factor of Th1 cells, intrapair variability was very low among MZ and high among DZ twins, indicative of a strong genetic influence on T-bet (heritability, 0.93, 95% CI, 0.84–1.0).

FIGURE 2.

EMSA and Western blot analysis of transcription factor expression. PBMC from MZ and DZ twins were isolated from peripheral blood and stimulated in a T cell-specific fashion with anti-CD3 plus anti-CD28 Abs as specified in Materials and Methods. Nuclear proteins were isolated and used for EMSA or Western blot analysis as indicated. For Western blot analysis under denaturing conditions (upper panels) 15 μg nuclear proteins per patient was analyzed using specific Abs for T-bet, GATA-3, and NF-κB p65. Western blots were analyzed by densitometry and the expression of the transcription factors was compared among the twins based on the densitometric data. The higher densitometric values among twins received an arbitrary value of 100. The value of the corresponding twin was then directly compared with this arbitrary value of 100 (percent expression level as compared with the other twin). A value of 100 for the other twin therefore means identical expression between twins. Mean expression values ± SEM is shown (100 = identical expression between twins). The number of patients per group is indicated. For EMSA analysis under nondenaturing conditions (lower panels), 15 μg nuclear proteins per patient were analyzed using specific established dsDNA binding sites for NFAT and GATA-3. EMSA gels were analyzed by densitometry and the expression of the transcription factors was compared among the twins based on the densitometric data. Mean expression values ± SEM is shown (100 = identical expression between twins). Examples of two MZ and two DZ twin pairs are shown. J, Jurkat nuclear extracts; FP, free probe.

FIGURE 2.

EMSA and Western blot analysis of transcription factor expression. PBMC from MZ and DZ twins were isolated from peripheral blood and stimulated in a T cell-specific fashion with anti-CD3 plus anti-CD28 Abs as specified in Materials and Methods. Nuclear proteins were isolated and used for EMSA or Western blot analysis as indicated. For Western blot analysis under denaturing conditions (upper panels) 15 μg nuclear proteins per patient was analyzed using specific Abs for T-bet, GATA-3, and NF-κB p65. Western blots were analyzed by densitometry and the expression of the transcription factors was compared among the twins based on the densitometric data. The higher densitometric values among twins received an arbitrary value of 100. The value of the corresponding twin was then directly compared with this arbitrary value of 100 (percent expression level as compared with the other twin). A value of 100 for the other twin therefore means identical expression between twins. Mean expression values ± SEM is shown (100 = identical expression between twins). The number of patients per group is indicated. For EMSA analysis under nondenaturing conditions (lower panels), 15 μg nuclear proteins per patient were analyzed using specific established dsDNA binding sites for NFAT and GATA-3. EMSA gels were analyzed by densitometry and the expression of the transcription factors was compared among the twins based on the densitometric data. Mean expression values ± SEM is shown (100 = identical expression between twins). Examples of two MZ and two DZ twin pairs are shown. J, Jurkat nuclear extracts; FP, free probe.

Close modal
Table III.

Heritabilities of transcription factor productiona

Transcription FactorZygosityNo. of Twin PairsWithin-Pair VarianceHeritability95% CI
T-bet* MZ 17 0.02 0.93 0.83–1.0 
 DZ 13 0.24   
NFAT MZ 19 0.28 −2.38 −6.45–1.69 
 DZ 16 0.08   
GATA3 MZ 18 0.22 −0.05 −1.40–1.29 
 DZ 16 0.21   
NF-κB p65* MZ 22 0.31 −0.59 −2.70–1.51 
 DZ 15 0.20   
Transcription FactorZygosityNo. of Twin PairsWithin-Pair VarianceHeritability95% CI
T-bet* MZ 17 0.02 0.93 0.83–1.0 
 DZ 13 0.24   
NFAT MZ 19 0.28 −2.38 −6.45–1.69 
 DZ 16 0.08   
GATA3 MZ 18 0.22 −0.05 −1.40–1.29 
 DZ 16 0.21   
NF-κB p65* MZ 22 0.31 −0.59 −2.70–1.51 
 DZ 15 0.20   
a

Statistical analysis of T-bet production by isolated T-cells after stimulation with CD3 and CD28 Abs (comparison of MZ vs DZ twin pairs). Western blots (T-bet and NF-κB p65 (∗)) and EMSAs (NFAT and GATA-3) were analyzed by densitometry and the expression of the transcription factors was compared among the twins based on the densitometric data.

Our results show a strong genetic influence on the production of the cytokines TNF-α and IFN-γ but little or no contribution to IL-4 production in humans. IFN-γ production in CD4 T cells is under transcriptional control of T-bet, which was the only transcription factor that showed evidence for a strong genetic influence on its expression. In contrast, little or no influence of genetic factors was found for transcription factors that are known to be involved in regulation of Th2 cytokines such as NF-κB, NFAT, and GATA-3.

Systematic studies that investigate the genetic background of cytokine production in humans have been hampered by the large interindividual variability in cytokine production. To circumvent this problem, we have used the approach of a classical twin study. The technique of whole blood stimulation applied in our study has been used in previous studies, where it was shown to produce reliable and reproducible results (8, 12). Twin studies allow the estimation of the relative contribution of genetic and environmental components because MZ twins are genetically identical and DZ twins share on average 50% of their genes. Taking the intrapair variance of twins examined at the same time under identical conditions has the advantage of avoiding the pitfalls of large variability in cytokine secretion. Mitogens were chosen as stimulators for peripheral blood cells, since their effect does not rely on previous exposure to certain Ags.

In the present study, we did not find any differences in absolute levels of cytokine production between MZ and DZ twins but differences in similarities of IFN-γ production. In particular, we observed a lower intrapair variability of IFN-γ production in MZ twins as compared with DZ twins, whereas no differences in IL-4 production were found. It should be noted, however, that we assessed cytokine production (rather than T cell polarization) by CD3+ T cells that comprise a mixture of stimulated effector Th1 and Th2 cells (13). Since we found differences in similarity levels of T-bet between MZ and DZ twins and as T-bet is well known to regulate IFN-γ production in mice and humans (14, 15), we believe that the above observation on IFN-γ is due to the effects of T-bet on IFN-γ production of Th1 cells in our system. Whereas T-bet tightly controls IFN-γ expression in humans, only sustained forced overexpression of T-bet in human Th2 cells has any effects on IL-4 production (16). In human Th2 cells, GATA-3 is the key transcription factor that regulates IL-4 production (17). Therefore, different levels of T-bet in Th1 cells among DZ twins may lead to variable outcomes in IFN-γ production, whereas IL-4 production that is mainly driven by GATA-3 rather than T-bet in human Th2 effector cells may be similar among DZ twins.

Heritability of cytokine production has been investigated in a few pedigree studies, including mostly relatives of diseased subjects (12, 18, 19). Our results are in line with published data of TNF-α heritabilities with a high degree of genetic determination between 0.6 and 0.8 for TNF-α (12, 18, 19). For IFN-γ a pedigree study from a community in Uganda with a high prevalence of tuberculosis has reported a heritability of 0.4 (19). Cytokine production in whole blood assays is influenced by cell distribution in the peripheral blood. However, the results obtained with isolated anti-CD3/anti-CD28 or TT-stimulated T cells emphasize the strong genetic impact on IFN-γ production. Our data provide evidence for a mainly genetically determined regulation of two key factors of Th1 cytokine production, IFN-γ and T-bet. These results have implications for our current concepts of the pathogenesis of autoimmune and allergic disorders. Family and twin studies have shown a considerable genetic component in these diseases (20, 21, 22). The development of most autoimmune diseases depends on the cytokines TNF and IFN-γ and the transcription factor T-bet produced by Th1 cells, whereas the development of allergic diseases requires the production of IL-4 and IL-5 under the control of GATA-3 by Th2 cells. In contrast to IL-4 and GATA-3, the production of IFN-γ and T-bet expression is strongly influenced by genetic variability. Our study shows remarkable interindividual differences in the expression of cytokines and transcription factors with up to 50-fold differences in T-bet production between MZ pairs. These differences appear to be largely genetically determined and low and high producer pairs could be identified. This is in accordance with previous observations in mouse models where the genetic background was shown to determine the default pathway of Th cell development (6). For instance, Th2 differentiation and associated pathology can be caused by absent or low expression of IFN-γ and T-bet (23, 24). CD4+ T cells that normally respond to Ags by differentiation into Th1 cells default to the Th2 pathway in the absence of endogenous IFN-γ in IFN-γ knockout mice (23). Furthermore, asthmatic patients with augmented Th2 cytokine production showed reduced numbers of cells staining for T-bet in bronchial biopsies. The functional implication of this observation has been highlighted by the finding that mice carrying only one copy of the T-bet gene develop airway changes similar to those observed in asthma (24). Thus, already subtle changes in T-bet expression can predispose to Th2-type disease manifestations. This is also underlined in studies with human T cells and forced expression of T-bet (25). Thus, subtle changes in T-bet levels may modulate Th1/Th2 polarization in humans. Based on these observations, it will be of interest in future studies to compare the effects of T-bet in human Th cell polarization between MZ and DZ twins under Th1 and Th2 skewing conditions.

Taken together, these observations suggest that genetic control in the direction of human Th cell development resides mainly in the regulation of Th1 cytokines. Individuals with low IFN-γ production and T-bet expression could be prone to the development of Th2-type diseases such as allergies and asthma. It is thus tempting to speculate that the increase of Th2 (e.g., allergic asthma)- mediated atopic and allergic diseases in developed countries could be due to local environmental factors in T-bet low-expressing hosts genetically susceptible to develop such Th2-associated disorders (Fig. 3). This phenomenon might be augmented by the low prevalence of infections that favor Th1 cytokine production. In summary, genetic clues to the pathogenesis of infectious, autoimmune, and allergic diseases in humans have to be sought rather in the regulation of Th1 than Th2 cytokines.

FIGURE 3.

Model of human Th cell development. The direction of human Th cell development resides mainly in the regulation of Th1 cytokines. Individuals with low IFN-γ production and T-bet expression are prone to the development of Th2-type diseases such as allergies and asthma, whereas high T-bet expressors are predisposed to the development of autoimmune diseases. These genetic profiles are modified by infections, endotoxin exposure, and other environmental factors.

FIGURE 3.

Model of human Th cell development. The direction of human Th cell development resides mainly in the regulation of Th1 cytokines. Individuals with low IFN-γ production and T-bet expression are prone to the development of Th2-type diseases such as allergies and asthma, whereas high T-bet expressors are predisposed to the development of autoimmune diseases. These genetic profiles are modified by infections, endotoxin exposure, and other environmental factors.

Close modal

We thank all involved twins for their willingness and patience to participate in this study, which would not have been possible without their support.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by grants from the Deutsche Forschungsgemeinschaft, SFB 490, Project A3. and Project C7 (to T.H. and M.F.N.).

3

Abbreviations used in this paper: MZ, monozygotic; DZ, dizygotic; TT, tetanus toxoid; BMI, body mass index; VWP, variance within pairs; CI, confidence interval.

1
Bach, J. F..
2002
. The effect of infections on susceptibility to autoimmune and allergic diseases.
N. Engl. J. Med.
347
:
911
.-920.
2
Neurath, M. F., S. Finotto, L. H. Glimcher.
2002
. The role of Th1/Th2 polarization in mucosal immunity.
Nat. Med.
8
:
567
.-573.
3
Szabo, S. J., S. T. Kim, G. L. Costa, X. Zhang, C. G. Fathman, L. H. Glimcher.
2000
. A novel transcription factor, T-bet, directs Th1 lineage commitment.
Cell
100
:
655
.-669.
4
Ouyang, W., M. Lohning, Z. Gao, M. Assenmacher, S. Ranganath, A. Radbruch, K. M. Murphy.
2000
. Stat6-independent GATA-3 autoactivation directs IL-4-independent Th2 development and commitment.
Immunity
12
:
27
.-37.
5
Larche, M., D. S. Robinson, A. B. Kay.
2003
. The role of T lymphocytes in the pathogenesis of asthma.
J. Allergy Clin. Immunol.
111
:
450
.-463. qu.
6
Hsieh, C. S., S. E. Macatonia, A. O’Garra, K. M. Murphy.
1995
. T cell genetic background determines default T helper phenotype development in vitro.
J. Exp. Med.
181
:
713
.-721.
7
Hohler, T., E. Reuss, N. Evers, E. Dietrich, C. Rittner, C. M. Freitag, J. Vollmar, P. M. Schneider, R. Fimmers.
2002
. Differential genetic determination of immune responsiveness to hepatitis B surface antigen and to hepatitis A virus: a vaccination study in twins.
Lancet
360
:
991
.-995.
8
van der Linden, M. W., T. W. Huizinga, D. J. Stoeken, A. Sturk, R. G. Westendorp.
1998
. Determination of tumour necrosis factor-α and interleukin-10 production in a whole blood stimulation system: assessment of laboratory error and individual variation.
J. Immunol. Methods
218
:
63
.-71.
9
Atreya, R., J. Mudter, S. Finotto, J. Mullberg, T. Jostock, S. Wirtz, M. Schutz, B. Bartsch, M. Holtmann, C. Becker, et al
2000
. Blockade of interleukin 6 trans signaling suppresses T-cell resistance against apoptosis in chronic intestinal inflammation: evidence in Crohn disease and experimental colitis in vivo.
Nat. Med.
6
:
583
.-588.
10
SAS Institute, Inc.
1997
.
SAS/STAT Software: Changes and Enhancements through Release 6.12
SAS Institute Inc., Cary, NC.
11
Vogel, F., A. Motulsky.
1997
. E. Vogel, ed.
Human Genetics: Problems and Approaches
90
.-95. Springer, Berlin.
12
Westendorp, R. G., J. A. Langermans, T. W. Huizinga, A. H. Elouali, C. L. Verweij, D. I. Boomsma, J. P. Vandenbroucke, J. P. Vandenbrouke.
1997
. Genetic influence on cytokine production and fatal meningococcal disease.
Lancet
349
:
170
.-173.
13
Matsuzaki, S., K. Shinozaki, N. Kobayashi, K. Agematsu.
2005
. Polarization of Th1/Th2 in human CD4 T cells separated by CD62L: analysis by transcription factors.
Allergy
60
:
780
.-787.
14
Mowen, K. A., L. H. Glimcher.
2004
. Signaling pathways in Th2 development.
Immunol. Rev.
202
:
203
.-222.
15
Neurath, M. F., B. Weigmann, S. Finotto, J. Glickman, E. Nieuwenhuis, H. Iijima, A. Mizoguchi, E. Mizoguchi, J. Mudter, P. Galle, et al
2002
. The transcription factor T-bet regulates mucosal T cell activation in experimental colitis and Crohn’s disease.
J. Exp. Med.
195
:
1129
.-1143.
16
Lametschwandtner, G., T. Biedermann, C. Schwarzler, C. Gunther, J. Kund, S. Fassl, S. Hinteregger, N. Carballido-Perrig, S. J. Szabo, L. H. Glimcher, J. M. Carballido.
2004
. Sustained T-bet expression confers polarized human TH2 cells with TH1-like cytokine production and migratory capacities.
J. Allergy Clin. Immunol.
113
:
987
.-994.
17
Crane, I. J., J. V. Forrester.
2005
. Th1 and Th2 lymphocytes in autoimmune disease.
Crit. Rev. Immunol.
25
:
75
.-102.
18
Pantsulaia, I., S. Trofimov, E. Kobyliansky, G. Livshits.
2002
. Genetic and environmental influences on IL-6 and TNF-α plasma levels in apparently healthy general population.
Cytokine
19
:
138
.-146.
19
Stein, C. M., D. Guwatudde, M. Nakakeeto, P. Peters, R. C. Elston, H. K. Tiwari, R. Mugerwa, C. C. Whalen.
2003
. Heritability analysis of cytokines as intermediate phenotypes of tuberculosis.
J. Infect. Dis.
187
:
1679
.-1685.
20
Mumford, C. J., N. W. Wood, H. Kellar-Wood, J. W. Thorpe, D. H. Miller, D. A. Compston.
1994
. The British Isles survey of multiple sclerosis in twins.
Neurology
44
:
11
.-15.
21
Bach, J. F..
1994
. Insulin-dependent diabetes mellitus as an autoimmune disease.
Endocr. Rev.
15
:
516
.-542.
22
Skadhauge, L. R., K. Christensen, K. O. Kyvik, T. Sigsgaard.
1999
. Genetic and environmental influence on asthma: a population-based study of 11,688 Danish twin pairs.
Eur. Respir. J.
13
:
8
.-14.
23
Wang, Z. E., S. L. Reiner, S. Zheng, D. K. Dalton, R. M. Locksley.
1994
. CD4+ effector cells default to the Th2 pathway in interferon γ-deficient mice infected with Leishmania major.
J. Exp. Med.
179
:
1367
.-1371.
24
Finotto, S., M. F. Neurath, J. N. Glickman, S. Qin, H. A. Lehr, F. H. Green, K. Ackerman, K. Haley, P. R. Galle, S. J. Szabo, et al
2002
. Development of spontaneous airway changes consistent with human asthma in mice lacking T-bet.
Science
295
:
336
.-338.
25
Sundrud, M. S., S. M. Grill, D. Ni, K. Nagata, S. S. Alkan, A. Subramaniam, D. Unutmaz.
2003
. Genetic reprogramming of primary human T cells reveals functional plasticity in Th cell differentiation.
J. Immunol.
171
:
3542
.-3549.