Age-related differentiations of Th1/Th2 cytokines in newborn infants.

OBJECTIVE: To evaluate age-related differentiation of immune response in newborns by measuring serum concentrations of interleukin-2 (IL-2), interleukin-4 (IL-4) and interferon-gamma (IFN-gamma) during the perinatal period. SUBJECTS AND METHODS: Fifty-seven healthy term neonates, their mothers and 25 healthy adults (controls) age-matched to the mothers were included in the study. Cytokine concentrations were measured in the umbilical cord (UC), and in first-day (1N) and fifth-day (5N) neonatal samples, compared with those in maternal serum (MS) and control serum samples. RESULTS: Serum IL-2 concentrations in the UC were markedly elevated compared with those in MS and controls (p < 0.0001), decreasing significantly thereafter up to 5N (p < 0.001). IL-4 serum concentrations did not differ significantly between the UC, 1N and 5N samples; they were, however, markedly elevated compared with those in MS (p < 0.001, p < 0.0007 and p < 0.0001, respectively) and controls (p < 0.05, p < 0.01 and p < 0.006, respectively). IFN-gamma serum concentrations were significantly lower in the UC compared with those in controls (p < 0.04), increasing significantly up to 5N (p < 0.03). Both IFN-gamma/IL-2 and IFN-gamma/IL-4 ratios increased significantly in 5N, compared with those in the UC (p < 0.001 and p < 0.03). CONCLUSION: Our findings indicate a differential cytokine balance at birth with enhanced expression of IL-2 and IL-4 against IFN-gamma. However, a regularization of immune response seems to proceed quickly during the early neonatal life.


Introduction
Cytokines are the hormonal messengers responsible for most of the biological effects in the immune system, such as cell-mediated immunity and allergic type responses. Based on their preferential functional capacity, cytokines can be divided into two families: type 1 (Th1) cytokines, such as interleukin-2 (IL-2) and interferon-g (IFN-g), which mainly stimulate cellmediated immunity; and type-2 (Th2) cytokines, the main representative of which is interleukin-4 (IL-4), which primarily induce B-cell differentiation and are associated with allergic responses. 1 In the mouse, a CD4 ' cell that produces IL-4 but not IFN-g, or IL-2, is termed Th2, and a cell that produces IFN-g but not IL-4 is termed Th1. 2 In humans, such a distinction between cell types is less clear because, depending on the stimulus, cells can produce both IFN-g and IL-4. Therefore, it is better to use a ratio of IFN-g/IL-4 to define the phenotype of the cell. 3 However, although IFN-g and IL-4 are useful markers for Th1 versus Th2 responses, they do not give definitive descriptions of the responses. On the other hand, in newborns, IFN-g production by cord and peripheral neonatal cells does not correlate with IL-2 production. 1 Thus, the safest way to describe the cytokine response patterns is to determine both the ratios and the absolute amounts of cytokines. 4 For this reason, we focused to study serum concentrations of IL-2, IL-4 and IFN-g, as well as their ratios IFN-g/IL-2 and IFN-g/IL-4, during the perinatal period, in order to evaluate the age-related differentiation of immune response in newborns.

Material and methods
The study was approved by the Ethics Committee of Aretaieion Hospital and informed written consent was acquired from the participating mothers and controls on admission in the Clinic. The study included 57 healthy, appropriate for gestational age term neonates, delivered after a singleton uncomplicated pregnancy from healthy, non-smoking mothers (mean age, 23.89/4.4 years; range, 20 Á/39 years). In all cases, Apgar scores were ]/8 in the first and fifth minutes, and placentas were normal in appearance and weight (mean weight9/SD, 4499/16 g). The demographic data of participating neonates are presented in Table 1. The mothers of the aforementioned neonates also participated in the study, as well as 25 healthy, non-pregnant, non-smoking women acting as controls (mean age9/SD, 23.09/4.8 years; range, 21Á/40 years), with regular menstrual cycles, all in the periovulation period of their cycle.
Blood samples were collected from controls, mothers before delivery (first stage of labor), the doubly clamped umbilical cord (UC) at delivery (mixed arteriovenous blood), and the neonates in the first and fifth day postpartum. Blood was collected in pyrogen-free tubes, was immediately centrifuged after clotting; and the supernatant serum was kept frozen at (/308C until assay. Cytokines were measured using highly sensitive immunoenzyme techniques, with commercially available enzyme-linked immunosorbent assays (EASIA; Medgenix, Fleurus, Belgium for IL-2; Quantikine † hIL-4; HS R&D Systems, MN, USA for IL-4; and Quantikine † IFN-g; R&D Systems for IFN-g). With an extension of the standard curve at very low concentrations and the application of a logistic program for the best fitting, very low sensitivities were achieved. Thus, the new performance characteristics were as follows: sensitivity, intra-assay and inter-assay coefficients of variation were, respectively: 0.1 IU/ml, 5.7% and 7.5% for IL-2; 0.11 pg/ ml, 4.6% and 5.8% for IL-4; and 3 pg/ml, 2.6% and 6.4% for IFN-g.

Statistics
Statistical analysis involved parametric tests for IL-2 (t -test, one-way analysis of variance) and nonparametric tests for IL-4 and IFN-g (Wilcoxon test, and Kruskal Á/Wallis analysis of variance), since data presented normal and abnormal distributions, respectively. Comparisons between IL-2 and IFN-g or IL-4 were performed with non-parametric tests. The level p B/0.05 was considered statistically significant.

Results
Summarized data are presented in Tables 2 and 3:

Discussion
The fetal immune system tends to develop towards a Th2-type immune response, due to the production of the Th2 promoting factors, such as IL-4, IL-10 and prostaglandin E2 from the placenta. 1 Postnatally, newborn infants are selectively impaired regarding development of Th1 memory effector function and are biased to Th2 function in all phases of an immune response. 5 Consequently, umbilical cord white cells would be expected to produce a high level of IL-4, accompanied by a reduced production of IFN-g, resulting in an initial dysregulation of IFN-g and IL-4 in neonates. 6 However, the proliferative response of T lymphocytes on their exposure to antigens is the result of two major events: the generation of the T-cell growth factor IL-2 and the induction of T-cell responsiveness to IL-2, controlled by the expression of an inducible cell surface IL-2 receptor (IL-2R). 7 Cord blood T cells synthesize sufficient amounts of IL-2, express IL-2R and are able to form a functional, intact, high-affinity receptor complex to support T-cell growth and proliferation. However, cord blood T cells lack the capability to induce the production of IFN-g, regardless of IL-2 and IL-2R upregulation of its production. 8 Indeed, IL-2 is a potent IFN-g-inducing cytokine and is produced in nearly equal amounts by both memory and naive T cells from neonatal and adult blood. Therefore, the striking discrepancy between neonatal and adult IFN-g production is not related to deficient IL-2 induction of IFN-g. 9 Consistent with these immunological demands, our findings demonstrate clearly elevated concentrations of IL-2 in umbilical cord and in the first-day samples, as well as significantly increased IL-4 levels in all three neonatal samples, compared with those in maternal and adult serum.
A central biological function of IL-4 is the induction of differentiation of CD4 ' Th precursor T lymphocytes into Th2 cells in both humans and mice. 10 An important finding of this study is that not only memory cells, but also naive human CD4 ' T cells can be an initial source of IL-4 during primary immune responses.
Moreover, an essential biological activity of IL-4 in the development of allergic inflammation is the inhibition of T-cell apoptosis and the ability to drive the differentiation of Th0 into Th2 lymphocytes, which are able to secrete IL-4, IL-5 and IL-13, but not IFN-g. 11 The lack of sufficient IFN-g production after birth may contribute to impaired neonatal antiviral responses and facilitate allergic sensitization. 12 Thus, the IL-4 levels in serum, but not these of IFN-g, were associated with allergic disease in infancy. Elevated concentrations of IL-4 were recorded in atopic neonates, before the onset of clinical symptoms, 13 and the severity of the atopic state correlates with the degree of imbalance in IL-4 and IFN-g production. 14 In agreement with previous studies, 6,15 our findings indicate very low IFN-g concentrations in the UC. IFN-g deficiency in human neonates may represent a developmental phenomenon, possibly associated with cellular responsiveness of neonatal T cells to cytokines, which is crucial to Th1 differentiation as well as in translational and post-translational defects. 15,16 Additional findings of interest in the current study include the dependence of maternal and cord blood serum IFN-g concentrations on the mode of delivery. Cytokine values were higher in mothers giving birth after vaginal delivery than in those delivering by elective cesarean section. Similarly, infants born after vaginal delivery had higher concentrations of IFN-g compared with those born by elective cesarean section. Thus, labor possibly induces the production of such important modulator of immune response, both in mothers and their newborn infants, strengthening the maternal and neonatal defense against perinatal infections. 17 The most important finding of this study was a clear increase of both IFN-g/IL-2 and IFN-g/IL-4 ratios from cord serum up to the fifth day of life. It seems that newborn infant cells can rapidly differentiate into cells with adult-like functional capacities. Thus, the relative immunodeficiency, immaturity, or lack of experience of the neonate may be an adaptive mechanism to optimize survival by balancing the conflicting immunologic requirements of life in utero with those of the external environment. 18 In conclusion, our findings indicate a differential cytokine balance at birth with enhanced expression of IL-2 and IL-4 against IFN-g. However, a regularization of immune response seems to proceed quickly during the early neonatal life.