Asthma is a complex disease whose phenotypic heterogeneity has been shown to depend both on genetic factors, such as those that underlie the appearance of atopy, and on environmental factors 1 2 . Among the environmental factors, one that is sine qua non in the determination of the development, severity and chronicity of allergic asthma is the contact with environmental allergens, especially those derived from indoor mites, cockroaches, pet epithelia and pollens 3 4 . Indeed, atopy is usually defined as an intrinsic condition characterized by the presence of circulating IgE antibodies against those environmental allergens (i.e., antibodies against innocuous antigens) and/or skin reactivity to the same allergens. Although a good association between the presence in the circulation of IgE antibodies and a skin prick test (SPT) positivity for the same aeroallergen has been reported, especially in high-income countries and in affluent sub-populations of developing countries 5 6 , this is not always the case. Thus, a poor association between these two atopy markers in people from rural communities or non-affluent countries has been observed 7 8 . Recently, in partial disagreement with previous studies, the ISAAC research group reported dissociation between these atopy markers in both affluent and non-affluent populations 9 .

Helminth infections usually accompany low economic development in the tropics, and they may play a role in establishing the discrepancy between the presence of positive SPT and allergen-specific IgE antibodies (sIgE) found in low-income communities 10 . Several hypotheses have been proposed to explain why, in some individuals with sIgE, contact of skin mast cells with allergen does not lead to their degranulation. Some of these hypotheses involve: (i) the competition of polyclonal IgE raised by helminths with sIgE for mast-cell Fcε receptors 11 ; (ii) anti-allergen IgG4 antibodies blocking IgE-mediated immunity and allergic processes 12 ; (iii) presence of cross-reactive IgE antibodies that react with glycoprotein carbohydrate moieties [α(1,3)-fucose and ß(1,2)-xylose on N-glycans], which are widely present in plants and invertebrates and produce clinically irrelevant antibodies 13 ; (iv) the induction of regulatory CD4⁺ CD25⁺ Foxp3⁺ T cells that are capable of down-regulating the allergic process 14 .

In the present study, six different parameters were studied in individuals who had circulating IgE antibodies against B. tropicalis extract (BtE) and had either a negative (SPT-negative individuals) or positive (SPT-positive individuals) result in a SPT in which BtE was used as antigen (BtE SPT). The studied parameters were: (i) the levels of circulating anti-B. tropicalis IgE antibodies (αBtE IgE); (ii) the levels of total circulating IgE; (iii) an arbitrary αBtE IgE/total IgE ratio; (iv) the proportion of the αBtE IgE that reacted with carbohydrate antigenic determinants; (v) the production of IL-10 by BtE- and A. lumbricoides extract (AlE)-stimulated blood mononuclear cells; (vi) the proportion of αBtE IgE that cross-reacted with AlE.

IgE antibodies are usually biologically active in remarkably low concentrations 15 . The identification of the underlying factors that in some cases may lead these antibodies to fail to trigger, in the presence of antigen, the most potent of the IgE-associated effector mechanisms, namely the degranulation of mast cells, may be relevant for the development of possible therapies for allergic diseases.

Some aspects of the immune response that may underlie the absence of a positive BtE SPT in the presence of circulating α-BtE IgE were assessed in a group of poor individuals from a Brazilian northeastern large urban center. Both individuals with positive SPT and individuals with negative SPT were selected from the same area, and were probably subjected to the same social and environmental conditions, including exposition to the same pathogens. All studied subjects had serum α-BtE IgE levels clearly above the cut-off of the employed assay.

The total IgE levels did not differ in the two studied groups, although the levels in three out of the 12 SPT-negative individuals were at least 4.7-fold higher than the highest total IgE value in the group of SPT-positive individuals. As there was a great overlap of total IgE levels between the two groups, a possible blocking of IgE receptors in mast cells by allergen-unrelated IgE could not explain the negativity in the SPT in the majority of the individuals. However, since distinct factors may mediate the SPT negativity in different patients, one cannot exclude that the high levels of total IgE in a minority of individuals could be inhibiting, by competition with low levels of specific IgE, the degranulation of mast cells. Indeed, when an arbitrary ratio of specific to total IgE was calculated for each serum, the ratios in two out of 12 sera from the SPT-negative individuals were smaller than the smallest ratio observed in the sera from 20 SPT-positive individuals. In these two individuals, therefore, it is possible that the negativity in the SPT could be due to the blocking of allergen-specific by non-specific IgE. A prediction of this hypothesis is that the sera from these two individuals would fail to sensitize basophils for in vitro allergen-triggered degranulation. A larger sample of SPT-positive and SPT-negative individuals than the one used in the present work, however, should be studied in order to allow one to conclude that a significant proportion of SPT-negative individuals have smaller specific to total IgE ratios than the SPT-positive individuals.

In the studied population, α-BtE IgE levels were significantly higher in those subjects presenting with positive SPT than with negative SPT, indicating that high serum specific IgE antibody levels are associated with SPT positivity. In agreement with this finding, a statistically significant correlation, although weak, between circulating α-BtE IgE levels and the mean wheal size of the SPT reactions was observed. However, as a great overlap of α-BtE IgE levels was found between the SPT-negative and the SPT-positive groups (11 out of 12 sera from the negative SPT group, and 11 out of 20 sera from the positive SPT group, produced ELISA optical densities between the cut-off and 0.3), the simple explanation that low levels of specific IgE led to negative SPT in all the studied individuals does not hold up.

Allergenic cross-reactive carbohydrate determinants (CCDs) are sugar moieties of glycoproteins that induce IgE production and cross-react with a broad range of allergens 13 . CCDs have been implicated in the discrepancy between the presence of specific IgE to pollen and the absence of a positive SPT to that same allergen, as the cross-reactive IgE antibodies seem to be incapable of inducing mast cell degranulation 16 . It is therefore believed that the carbohydrate-reactive IgE antibodies are clinically irrelevant, possibly because they are unable to trigger degranulation by cross-linking the mast-cell receptors or because they have low affinity for the antigen 17 . In the present work, the pre-treatment of BtE with metaperiodate/borohydride caused varying reduction levels (from 10 to 60%) in the optical densities that were obtained in the ELISA performed with sera from both SPT-positive and SPT-negative individuals. However, the difference in percentages of reduction between the positive-SPT and negative-SPT groups was not statistically significant. These carbohydrates-binding IgE antibodies, therefore, do not appear to affect the mast cell-dependent skin response to B. tropicalis antigens, unlike the antibodies against pollen antigens 16 . However, the possibility that the IgE antibodies from the two groups of individuals studied in the present work recognize different carbohydrate moieties in the BtE, and the anti-carbohydrate IgE in the SPT-negative individuals would not cross-link mast-cell receptors in the presence of B. tropicalis antigens, cannot be excluded. It would, therefore, be worthwhile to investigate whether carbohydrate-binding antibodies from the two groups are equally able to sensitize basophils for antigen-triggered degranulation in vitro.

Helminthic infections have the ability to modulate the human immune system and suppress allergic responses, although this does not happen in infections with all helminth species 18 . With regard to asthma and atopy, the role of A. lumbricoides infections is controversial. In studies with Chinese and Costa Rican children, an increased risk of childhood asthma and atopy has been associated with A. lumbricoides infection 19 20 . However, studies with Cuban and Ethiopian children have shown that infection with this worm protects against atopic dermatitis and wheezing 21 22 . In the present work, an A. lumbricoides extract was used in order to study the influence of Ascaris antigens in two aspects of the immune response: IL-10 production and presence of cross-reactive antibodies.

It has been proposed that a balance between different populations of cytokine-producing lymphocytes, such as Th2 and Th1, and the presence of IL-10 - producing Tr1, greatly affect the response to harmless environmental molecules and the development of an allergic inflammatory response 23 . Moreover, the in vitro production of histamine by human mast cells, derived from umbilical cord blood, has been shown to be inhibited by IL-10 24 . In an attempt to assess if there were differences in IL-10 production by peripheral blood cells from individuals with serum α-BtE IgE, with or without positive BtE SPT, their blood were cultivated in the presence of PWM, BCG lysate, BtE, and AlE. The IL-10 production by the blood cells, subjected to any of the four stimuli, of subjects with and without positive BtE SPT, did not differ. This result is consistent with the finding that mesenteric lymph node cells of IL-10^-/- mice that had been chronically infected with the helminth Heligmosomoides polygyrus, when transferred into uninfected allergen-sensitized wild-type recipients, suppressed the allergic inflammation, leading to the conclusion that the cells activated by H. polygyrus infection do not depend on intrinsic IL-10 to suppress the allergic response 14 . This is a controversial issue and probably depends on the species of helminth and/or of the host. For instance, Gabonese children with urinary schistosomiasis have been shown to have a lower prevalence of a positive skin reaction to house-dust mite than those without the disease, a finding that was associated to IL-10 production 25 . However, in our study population, levels of IL-10 do not appear to influence the dissociation between the presence of α-BtE IgE and the absence of BtE SPT. Whether further investigations, studying populations large enough to allow the groups of SPT-positive and SPT-negative individuals to be further stratified in sub-groups of individuals with high and low total IgE levels, would show differences in IL-10 production in response to either helminthic or mite antigens, is open to speculation.

Cross-reactions between helminthic and mite antigens have been reported. A study published in 2001 shows cross-reactivity between the nematode Anisakis simplex and the dust mites Acarus siro, Lepidoglyphus destructor, Tyrophagus putrescentiae and Dermatophagoides pteronyssinus 26 . There is little information, however, about cross-reactivities between B. tropicalis and A. lumbricoides antigens. Only one study describes cross-reactive antigens between these two organisms, namely tropomyosin and glutathione-S-transferase 27 . In accordance with these findings, cross-reactivity between B. tropicalis and A. lumbricoides antigens was demonstrated in the present study. The binding of α-BtE IgE to BtE was reduced in a dose-dependent manner by pre-incubation of the sera with an A. lumbricoides extract, in sera from both positive-SPT and negative-SPT individuals. At the highest concentrations of AlE (30 and 300 μg/mL), the proportion of α-BtE IgE antibodies cross-reacting with AlE reached a plateau (data not shown), with no difference between the two studied groups. However, in the lowest concentration of AlE (0.3 μg/mL), more reduction in α-BtE IgE levels was seen in the SPT- negative group than in the SPT-positive group. This finding cannot be ascribed to differences in pre-absorption levels of α-BtE IgE, because the sera in both groups were paired for this variable. On the other hand, it could be due to a higher avidity for A. lumbricoides antigens of the α-BtE IgE from the SPT-negative group than the α-BtE IgE from the SPT-positive group. This could happen if the cross-reactive antibodies in the SPT-negative individuals were originally produced in response to an A. lumbricoides infection, and therefore could have a lower avidity for B. tropicalis antigens than antibodies elicited by the mite antigens. In any case, it is possible that these cross-reactive antibodies may contribute to the SPT negativity in some individuals of the BtE SPT-negative group. The specificity of the reduction in α-BtE IgE levels by incubation with AlE was proved by assaying the same AlE-treated sera for total IgE content. If IgE bound non-specifically to AlE, similar proportions of reduction in total IgE concentration and in α-BtE IgE activity should be observed in the AlE-treated sera, a fact that did not occur.

Overall, α-BtE IgE levels were found to be directly related to positivity in the BtE SPT and with the average diameter of the BtE SPT wheal. Thus, the more α-BtE IgE an individual has, the more likely he/she will have a positive SPT and react more intensely when coming into contact with that allergen. However, in all or in most of the studied individuals, α-BtE IgE levels, total IgE levels, IL-10 production by peripheral blood cells and carbohydrate-binding IgE did not seem to play an important role in the discrepancy between the presence of specific IgE in the bloodstream and the absence of a positive SPT in the studied individuals. On the other hand, the present results are consistent with the possibilities that a low ratio of specific IgE to total IgE levels (in a minority of individuals), and difference in antigenic specificities of the α-BtE IgE (which would have a high aviditity for A. lumbricoides antigens in BtE SPT-negative individuals), may account for the discrepancy between specific IgE levels and SPT results. The results described in the present paper indicate, therefore, that different factors may be responsible or contribute to the dissociation of positive BtE SPT results and the presence of α-BtE IgE in the circulation. In addition, the fact that a large proportion of the IgE antibodies that reacts with B. tropicalis also reacts with A. lumbricoides antigens in a poor Latin American population may have relevance in the immunodiagnosis of atopy and even in its pathogenesis, a possibility that is only beginning to be investigated. Further studies should be implemented with a larger number of samples to investigate the possibility that the mast-cell degranulation may be inhibited by IgE cross-reacting with mite and helminthes, in order to understand the role played by these antibodies in the pathogenesis of the type-I hypersensitivity to mite antigens.

SPT: Skin prick test; BtE: B. tropicalis extract; α-BtE IgE: Anti-B. tropicalis extract IgE antibodies; AlE: Ascaris lumbricoides extract; IL-10: Interleukin 10

JCMP conducted the experiments under the guidance of NMAN and drafted the manuscript. SBJ helped in the design and in the conduction of the experiments. RVV performed all statistical analysis and reviewed the manuscript. MLB coordinated the field studies. LCPC helped in the design of the experiments and carefully reviewed the manuscript. NMAN conceived and designed the study and coordinated the research team. All authors read and approved the manuscript.