In search for predictive factors for atopy in human cord blood
Natalija Novak, MD
Department of Dermatology
Since early prevention is regarded as an important corner stone in the management of atopic diseases, the identification of reliable markers detecting individuals at risk are of major interest. Therefore, many efforts have been made to unravel reliable predictors for atopy which might identify children at risk and allow the initiation of preventive strategies at an early stage. In the past, much scientific energy has been forced in particular on the development of as noninvasive methods as possible to reach this goal. It is obvious that the identification of markers for atopy at the earliest time of life – namely immediately after birth – represents one of the most attractive attempts. In consequence various studies have been initiated to address this issue investigating markers for atopy in cord blood. Most of them have been geared to our current knowledge about cellular and soluble factors which are dysregulated in adolescent atopic individuals. Although the findings of these studies will improve our knowledge about the initial evolution of atopy, several parameters evaluated did not show any association or have led to almost conflicting results. In order to provide an up-date about the current developments in this field, recent research findings on predictive factors for atopy in cord blood are summarized in the following synopsis.
Within the last few decades atopic diseases such as atopic eczema (AE), allergic rhinitis (AR) and allergic asthma bronchiale (AB) showed an increasing prevalence in western societies (1). Atopy describes the personal/familial tendency with a complex genetic background to become sensitized and produce immunoglobulin E (IgE) antibodies in response to common environmental allergens usually occurring in childhood or adolescence (2). Thus the term atopy is strictly linked to IgE mediated sensitization usually documented by positive skin prick testing (SPT) or detection of elevated serum IgE levels by various techniques such as the Radio-Allergo-Sorbent-Test (RAST). The classical atopic career starts during early infancy with a course of AE. During this critical period of life, some atopic children concomitantly develop asthma while their skin lesions improve or completely disappear. Later on symptoms of allergic rhinitis with persisting asthma may emerge (3).
In view of these observations, both – the necessity and impact of preventive measures become clear. Studies confirm the benefit of preventive intervention strategies of allergic rhinitis and asthma in specifically sensitized infants with atopic eczema with oral antihistamines (4, 5). Another promising preventive approach is the early induction of sublingual or subcutaneous immunotherapy to avoid the development of additional sensitizations and slow down the conduction of the atopic career (6, 7). Further on it has been shown in one study that treatment of children at risk until a time period of 6 month after birth with probiotics was effective in prevention of early atopic disease (8).
It is obvious that the identification of early markers of atopy in cord blood (CB) of newborns at delivery would offer the possibility of effective prevention strategies in subjects at risk. Therefore much effort has been made to identify putative predictors of atopy in the human CB. In this manner outstanding pathophysiological factors contributing to atopic disorders have been investigated in CB which will be reviewed here.
Cord blood IGE levels
Atopic disorders are closely linked to elevated serum IgE levels in affected individuals. It is assumed that IgE is not only involved in mediating type I allergic reactions such as in AB and AR but also in type IV allergic reactions which underlie AE (9). Evaluating cord blood IgE (CBIgE), it is important to emphasize that IgE levels in cord bood (CB) profoundly differ from those in adults. Whereas IgE serum levels greater than 100–150 kU/l are considered as elevated in adults, cut-off values in CB are not that clear in the present literature and differ between studies. Further on the follow-up time span for diagnosis of atopy vary in some studies. In one promising report during the early 80th of the last century around 80.0% of newborns with cord blood IgE (IgE) higher than or equal to 0.9 kU/l developed atopic disease before 6 years of age (10). Moreover another study in a large cohort of 1701 newborns reported that 70% of infants with an initially high CBIgE developed atopic diseases within 18 months (11). Another study choosing different CBIgE cut-off values (0.3, 0.5, 0.8, 1.1 kU/l) did not find an increased number of atopy affected infants among those with elevated CBIgE irrespective of the chosen cut-off value. However, in this study the follow-up time span was 18 months which might have been too short and the diagnosis atopy included both IgE and non-IgE mediated disease (12). Nevertheless the same authors could find in the same cohort a correlation between atopy diagnosed by atopic disease combined with elevated total IgE until 18 month of age and elevated CBIgE. In this context, the lowest suggested CBIgE cut-off value (0.3 kU/l) was superior although showing a poor specificity (13). In contrast, a different 18 months follow-up study with high CBIgE cut-off values of 1.2 kU/l raised specificity of CBIgE to 95%, while the sensitivity remained low (14). While the sensitivity might increase by lowering CBIgE cut-off values the specificity decreases which degrades the importance of CBIgE as a predictive factor for atopy. Depending on the cut-off CBIgE level the positive predictive value for atopy ranged between 20 and 95% with altogether very low sensitivity (12–16) so that elevated CBIgE by itself seem to be insufficient as a reliable predictive factor for atopy.
Nonetheless taking positive atopy family history into consideration, the percentage of elevated CBIgE increased with the number of close family members suffering from atopic diseases. Especially a positive history of atopy of the mother was found to be associated with increased CBIgE levels (17). At least, the combination of elevated CBIgE values with a positive mother's history for sensitizations was associated with the development of sensitization to inhalative allergens until the age of 2 years (18, 19). Lately, it has been reported that elevated CBIgE increased the risk especially for aeroallergen sensitization (20) but again CBIgE values showed only a slight sensitivity as a predictive factor for the onset of atopic diseases (21, 22). In contrast to IgE, high levels of cord blood IgG subclass, i.e. IgG4 antibodies to food and inhalant allergens were associated with maternal atopy. In contrast high levels of IgG antibodies to inhalant but not food allergens were associated with a reduced development of atopy in childhood. However, the reasons for the different effects of maternal IgG antibodies to food allergens vs aeroallergens are unknown (23).
Altogether studies investigating CBIgE are difficult to compare mainly because of different CBIgE cut-off values, varying clinical follow-up procedures and definition of atopy (Table 1). Because of their low sensitivity, CBIgE levels appear to be inappropriate as an exclusive reliable predictive marker for atopic diseases.
Table 1. Overview of the cut-off values for cord blood-IgE levels used in the different studies
|Bergmann et al. (17)||0.9||24||1314|
|Croner et al. (11)||0.9||18||1701|
|Edenharter et al. (21)||0.9||60||1314|
|Hansen et al. (13)||0.3||60||251|
|Hansen et al. (12)||0.3||18||688|
|Hide et al. (16)||0.6||12||1111|
|Kjellman et al. (10)||0.9||72||1651|
|Liu et al. (18)||0.54||6||545|
|Tariq et al. (20)||0.5||48||1218|
|Varonier et al. (14)||1.2||18||338|
Soluble mediators of atopy in cord blood
To date several soluble mediators of atopy in the serum or plasma of the patients are known to play a central role in the pathophysiology of atopic diseases. A characteristic feature of atopy is a Th2 immune response and its related cytokines interleukin (IL)-4, IL-5 and IL-13 which are involved in the induction of the IgE synthesis (24, 25). Furthermore, Th2 related soluble cytokine and immunoglobulin receptors like soluble (s-) CD30, sCD23 and sIL-4 receptor (sIL-4R) have been connected to atopic diseases in adults (26). These receptors have also been investigated in CB as putative predictors of atopy, but neither sCD30, sCD23 nor sIL-4R CB levels were sufficient in predicting the onset of atopic diseases during infancy/early childhood (27–29). Some chemokines favouring the outcome or being involved in the maintenance of a Th2 immune response like thymus and activation-regulated chemokine (TARC/CCL17), macrophage-derived chemokine (MDC/CCL22), eotaxin (EOX/CCL11), monocyte chemotactic protein 1 (MCP-1/CCL2), and interferon-gamma-(IFN-γ) inducible protein 10 (IP-10/CXCL10) are known to be elevated in adult atopic individuals especially at the exacerbation state of AD (3). Consequently, a recently published study investigated CB concentrations of these chemokines in newborns. However, serum chemokine concentrations of TARC/CCL17, MDC/CCL22, EOX/CCL11, MCP-1/CCL2 and IP-10/CXCL10 were not associated with enhanced total IgE levels, a positive family history for atopic diseases or the development of atopic disorders. Nevertheless elevated MDC concentrations in CB were associated with wheezing during infancy (30). Serum levels of Th2 related cytokines have also been analysed in CB. In a previous study infants monitored from birth to only 18 month of age, IL-4 failed to serve as a valuable predictor of atopy (28). Nonetheless, a follow-up time span of 18 month appears to be too short to draw a general conclusion on the predictive value of IL-4 for atopy especially since a recent publication reported that detectable IL-4 and IFN-γ in CB of newborns were connected to a lower risk of developing allergic asthma and sensitization to some inhalant allergens after the follow-up period has been expanded to 6 years (31).
Altogether, neither chemokines/cytokines nor soluble cytokine receptors in CB were of significant impact in predicting atopic disorders, so far. However, only few studies actually focused on this topic so that the study power is too low to draw a final conclusion. In view of the complexity of atopic disorders it is tempting to speculate that the combination of different chemokines/cytokines or soluble cytokine receptors and their presence or level in CB may play a central role.
Hygiene hypothesis and cord blood
In view of the hygiene hypothesis which underlines the importance of microbial contact of the immune system in order to induce a Th1 switch, CB and amniotic fluid levels of soluble cytokine receptors connected to bacterial immune defence have also been investigated for a putative protective or antagonizing effect on the development of atopic diseases. CD14 is a multifunctional receptor for endotoxin and exists as soluble (sCD14) as well as membrane-bound (mCD14). MCD14 is expressed on monocytes where it forms a complex with Toll-like receptor (TLR)-4 upon activation (32). TLRs represent a part of the innate immunity and are activated by microbial antigens. Especially TLR-2 polymorphisms have been linked to atopy (33). A recent study reports that a reduced exposure of the gastrointestinal tract via amniotic fluid to sCD14 increases the risk of developing atopic diseases (34). Considering sCD14 levels in CB significant differences have been observed: children with atopic mothers and these mothers themselves had the highest levels of sCD14. Nevertheless, at 2 years of age no significant differences in sCD14 levels were observable anymore between children with atopic mothers and children with nonatopic mothers. Therefore, a final association between sCD14 and atopic disease could not be affirmed (35).
Polyunsaturated fatty acids
There have been many reports connecting an abnormal n-6 and n-3 polyunsaturated fatty acid (PUFA) composition to the occurrence of atopic diseases. Especially the elongated and desaturated products of the essential fatty acids linoleic acid (C18:2n-6) and linolenic acid (18:3n-3) take part in inflammatory mechanisms. For instance, the C18:2n-6 related PUFA arachidonic acid (AA) mediates chemotactic signals and smooth muscle cell contraction, while the 18:3n-3 connected PUFA eicosapentaenoic acid (EPA) and the C18:2n-6 related PUFA dihomo-γ-linolenic acid (DHGLA) are known to be involved in suppressing inflammatory pathways. Recently, an abnormal PUFA composition has been demonstrated in CB serum as well as plasma and red blood cells of infants at high risk for atopic diseases compared with CB of infants at no risk for atopy (36–38). Interestingly, one study reported an impairment of CB serum PUFA composition and the onset of atopic diseases after a follow-up of 1 year in a cohort of 57 newborns at risk (38). Another study reported abnormal proportions of n-3 and n-6 PUFA in the serum of allergic mothers affecting their infants. In this study, a correlation of PUFA from the n-3 series with their products was found in women whose infants did not develop an atopic disease during the first 6 years of life. This correlation was missing in the group of mothers whose children developed atopic disorders (36, 39). A very recently published study in a large cohort of over 1000 newborns also investigated the predictive value of PUFA exposure for atopy (40). By analyzing ratios of different n-6 to n-3 PUFA and their products in red blood cells the authors could detect a significant association of alteration in ratios to wheezing and eczema although these associations were no longer significant after adjustment for multiple comparisons. The authors conclude that exposure to n-6 and n-3 PUFA is not a predictive factor for an early onset of atopic diseases. Since the follow-up time span investigated the infants to a maximum of 42 month the authors concede the case for longer follow-up.
Cytokine production of cord blood mononuclear cells
Th2 type cytokines such as IL-4, IL-5 and IL-13 are primarily involved in the induction and maintenance of the IgE and IgG1 antibody production while Th1 type cytokines such as IFN-γ and IL-12 antagonize Th2 immune responses (41). Many studies have focused on the relevance of Th2 cytokine production in neonates for the development of atopic disorders in early infancy with sometimes conflicting results. In this context, the predictive value of the Th2 cytokine IL-13 as an early marker for atopy has been investigated. It has been shown that the phytohemagglutinin (PHA)-induced production of IL-13 by CB mononuclear cells (CBMC) of term babies with a positive parental history of atopy was significantly lower in neonates who subsequently developed atopic symptoms until the age of 3 years (42). In contrast, three other studies reported a correlation between newborns producing elevated IL-13 levels and a significantly increased risk of developing atopic symptoms (43–45). These conflicting results might arise in main part from the diversity of the study settings and especially the techniques of cytokine analysis. While some studies used enzyme linked immunosorbent assay (ELISA) others investigated messenger RNA levels of cytokines or used intracellular staining methods for fluorescence activated cell sorter (FACS) analyses. In most studies, the cytokine analysis was performed after stimulation. Thereby, different stimulatory reagents were used such as PHA, phorbol myristate acetate (PMA) and ionomycin in addition to allergens, e.g. from house dust mite or staphylococcus enterotoxins which might account to the different heterogeneous results.
An elevated frequency of IL-4 producing CBMC and increased IL-4/IFN-γ ratio in response to PHA as well as lower numbers of IL-12 producing cells after in vitro allergen stimulation has been observed in CB of newborns at risk for atopy (46). In a recently published study low numbers of IL-12 and IFN-γ producing CBMC were connected to IgE sensitization during early childhood (47) and a pronounced production of IL-4 and IL-5 in CBMC and CB lymphocytes was associated with the subsequent development of atopy (48, 49). On the other hand, it has been demonstrated that detection of mRNA of Th2 related cytokines were lower in atopic infants compared to nonatopic infants (50).
The highest consistency in the literature concerning cytokine production was found in the context of decreased Th1 cytokine IFN-γ production in CB. Comparing neonates without risk with neonates at risk for atopy, newborns in the latter group had a lower IFN-γ producing capacity at birth and subsequently developed more often symptoms of atopy (51–56). Further on an increased percentage of CD4+ CD45RO+ memory T cells in CB of newborns who developed atopy has been found and an impaired allergen-induced IFN-γ production was observed. It has also been demonstrated that CD4+ CD45RO T cells from neonates of atopic mothers produce less anti-inflammatory TGF-β in response to cow milk protein (57) in vitro. Another study reported that expression of the gut-homing integrin αEβ7 on milk allergen-stimulated CB T cells preceded the development of early infancy atopic eczema (58).
Taken together, the insufficiency to produce adequate amounts cytokines promoting Th1 immune responses such as IL-12 and IFN-γ and the enhanced expression of integrins enabling T cells to migrate into the tissue could contribute to a higher susceptibility to develop atopic diseases. Further studies will have to elucidate not only the value of low IFN-γ producing CBMC as a predictive marker but also the benefit of IFN-γ provoking agent supplementation during infancy such as microbes.
The impact of antigen presenting cells
Recently, it has been shown that atopic diseases are associated with an abnormal responsiveness of myeloid progenitor cells to hematopoietic growth factors and the genetic risk for atopy seems to be mirrored by a modulated expression of hemopoietic cytokine receptors such as granulocyte-macrophage stimulating factor (GM-CSF) receptor on CB CD34+ stem cells of children with a higher risk to develop atopic diseases (59). GM-CSF not only plays a central role in the differentiation of granulocytes and macrophages but also in the generation of antigen presenting cells (APC) like dendritic cells (DC) (60). Changes in the phenotype of DC in adult atopic individuals such as the expression of the high affinity receptor for IgE (FcɛRI) have been shown to be involved in the pathophysiology of atopic disorders (61) but only little is known about the role of CB APC such as DC in the development of atopy. We could show recently, that FcɛRI expression on CB-derived DC (CBdDC) is highly dependent on supplemented TGF-β concentrations. Low concentrations of TGF-β stabilized FcɛRI on the surface of CBDC whereas high concentration of TGF-β suppressed its expression in our in vitro system (62). This observation is in line with recent publications connecting atopy to a low-producing TGF-β phenotype (63).
Further on, it has been reported very recently that a reduced APC IL-12 production in the perinatal period was associated with reduced T cell activation. Although there was no relationship between atopic risk and the capacity to produce IL-12 in the neonatal period, a (nonsignificant) trend for neonatal IL-12 responses was found to be lower in high-risk children who developed clinical atopy (64). Another study investigated the relationship between the maturation of CB APC at birth to symptoms of atopic disease in a 2 year follow-up. The authors could detect an inverse association between maturation stage of APC reflected by HLA-DR expression on CB monocytes and symptoms of atopic disease (65).
Together these data support the hypothesis that variations in APC function in early stages of life may be associated with the development of atopy later on. Especially the maturation stage of APC at birth might have an impact on the development of atopic diseases as it is well accepted that APC maturation is critical for the quality of immune responses (66).
Genetics and atopy
The genetic background of atopic diseases is well accepted especially since different twin studies have shown a greater prevalence of atopic disorders for monozygotic compared with dizygotic twins (67). Recently, many efforts has been made to identify genetic variability in genes encoding for factors such as cytokines, cytokine receptors and immunoglobulins which play a role in the development of atopic diseases. Single nucleotid polymorphisms (SNP) have been identified in genes encoding for IL-10, TGF-β, IL-4R, IL-13R (68–70).
One recently published study could demonstrate that polymorphisms in the high-affinity IgE receptor beta chain gene (FCER1B) and a silent substitution in the nitric oxide synthase (NOS)2A gene were associated with a reduced IL-13 responses in CB but none of these polymorphisms were associated with an increased risk to develop atopic eczema in the first year of life (71). This is in line with a different observation in which 373 infants at risk for atopy were followed longitudinally from birth to 12 month of age to determine the predictive value of polymorphisms in the TNF, IL4, and FCER1B genes which have been connected to atopy. In this study, only infants that were homozygote for the IL4-589*T allele had an increased risk for the development of rhinitis (72). Further on polymorphisms in the IL18 gene could be related to specific sensitization to common allergens and allergic rhinitis (73). Other studies associated polymorphisms of the IL10 and GMCSF gene with the development of atopic diseases in at-risk children (74, 75). In addition the −28C/G polymorphism in the RANTES gene has been shown to be related to asthma severity, representing a genetic risk factor for life-threatening asthma attacks in chinese children (76).
It is obvious that identification of SNP in CB of newborns at risk for early onset of atopy would offer a powerful tool to develop sufficient prevention strategies.
Since early prevention is a corner stone in the management of atopic diseases the identification of reliable markers detecting individuals at risk especially during early childhood is of major interest. Many efforts have been made in the past to unravel such factors in CB in accordance to what is known or regarded to be typical in the adult course of atopic disorders (Table 2). Although studies investigating CBIgE levels are very heterogenous, most studies confirm that CBIgE has only little value as an exclusive predictive factor for atopy mainly because of its low sensitivity. In consideration of the fact that most atopic children start off their atopic career lacking sensitizations to common aeroallergens and showing regular IgE levels (77) it is not surprising that elevated CBIgE levels at this stage have no predictive value. Particularly a follow-up time span of 18 months for diagnosing atopy might be too short although most studies chose this time frame. It appears that a positive family history of atopy still has the greatest value for atopy prediction (78) and CBIgE might serve as a complementary tool whenever there is an obvious risk for atopy in one parent or both. The abnormal composition of PUFA in atopic individuals is a well accepted phenomenom and the benefit of n-3 PUFA diet supplementation represents a promising approach for atopic individuals to improve their symptoms. It has been shown recently, that n-3 PUFA diet supplementation by atopic mothers reduces the incidence of atopic disorders in their infants indicating the importance of maternal diet. The impact of maternal PUFA metabolism on the infant has been confirmed in one study in a little cohort and partly by large cohort. Nevertheless, since maternal and fetal PUFA metabolism is closely related to each other, CB PUFA appears to be insufficient as a predictive marker for atopy.
Table 2. Summary of the studies showing significant associations of parameters in cord blood with the development of atopic disorders
|CBIgE||Maternal atopy and elevated CBIgE levels||16–18|
|Soluble Mediators||Elevated MDC CB levels||29|
|Microbial contact||Reduced gastrointestinal contact to sCD14||33|
|Th1 Cytokines||Reduced production of IFN-γ and IL-12||45, 46, 50–55|
Predominance of a Th2 immune response in atopic individuals points out the deviated immunological balance and represents an attractive attempt for investigations in CB.
However, most studies conducted to analyze the impact of the Th2 cytokines came to conflicting results. Nevertheless it has been reported recently that season of birth might as well influence cytokine levels in CB and might contribute to this phenomenon (79).
Anyhow, more studies should be performed especially focusing the relation and interaction between different chemokines as well as Th1 and Th2 cytokines so that it is not appropriate to draw a final conclusion. Another important approach is reflected by investigating APC of the CB since they play a major role in the initiation of immune responses. After antigen/allergen-uptake and presentation to T-cells, APC are capable to generate a Th1 or a Th2 response. The insufficient production of IL-12 which promotes immune responses of the Th1 type of CB APC might contribute to the Th2 skewing observed in atopic individuals. In this context the role of plasmacytoid blood DC which are known to induce Th2 immune responses has to be analysed in more detail in the future. Although no differences of circulating DC between newborns at risk and not at risk for atopy have been found, a CD11c- CD123dim+ DC population has been shown to prevail in CB (80). It is tempting to speculate that these cells might be involved in the generation or protection of allergic reaction in newborns as the first line of defence in utero since it is more than likely that these cells get very early in contact with allergens crossing the placenta. Further on the maturation state of different DC populations in CB capturing antigens could play a role in the initiation of allergic reactions since it is known that the DC maturation state influence the outcome of immunity or tolerance (66).
Although in some degree ethically critical, investigation of genetic markers for atopy might represent a useful tool in view of recruitment of cohorts for intervention studies. Moreover, not only the identification of children at risk but also the definition of subgroups of atopic individuals with distinct courses of diseases might result from such studies. This would lead to more precise preventive strategies. For instance evidence arises from the ETAC study that in some but not all atopic infants suffering from atopic eczema antihistamine medication might prevent allergic asthma when conducted early to therapy (5).
In conclusion, it has to be pointed out that the ideal predictive factor for atopy in CB is still to be found. Many promising parameters known to be involved in atopy turned out to be of little value not because of their lack of relevance but because of their probable minor involvement in the initiation of atopy during infancy. Before investigating predictive factors for atopy, more details about the immunological background of atopic diseases during childhood in combination with clear-cut clinical parameters and long-time follow-up studies are necessary to develop more concrete approaches. Furthermore, it would be important to take variants of atopic diseases into consideration to be able to distinguish between subtypes showing sensitization from those lacking sensitizations in order to achieve more homogeneous results.
This work was supported by grants from the Deutsche Forschungsgemeinschaft DFG No/454-2-1, DFG No/454-1-1 and BONFOR.