Rethinking Th2 antibody responses and allergic sensitization
*Arizona Respiratory Center and Department of Cell Biology and Anatomy, College of Medicine, University of Arizona, Tucson, AZ, USA, {University Children’s Hospital, Munich, Germany, {Molecular Immunoregulation Unit, San Ra¡aele Scienti¢c Institute, Milano, Italy, }Institute of Social and Preventive Medicine, University of Basel, Basel, Switzerland, }Children’s Hospital Salzburg, Paediatric Pulmonology and Allergology, Salzburg, Austria and kInstitute of Occupational and Environmental Medicine, University of Munich, Munich, Germany Abstract. Human Th2 cytokines (interleukins 4 and 13) induce co-expression of IgE and IgG4 through sequential switching. The regulation of IgG4 responses and the role of these responses in the pathogenesis of allergy have not been characterized. We are addressing these issues by comparing and contrasting the expression of allergen-speci¢c IgE and IgG4 in a population of European children thoroughly de¢ned for lifestyle, environmental exposures and allergic phenotypes. The current analysis focused exclusively on children from non-farming families (n ¼ 493) in order to avoid potential e¡ects of exposure to microbial products abundant in farming environments. We found that allergens induce Th2-mediated IgG4 and/or IgE responses in the majority of the population. Approximately two-thirds of the children had allergen-speci¢c IgG4 but not IgE, only a minority had both IgG4 and IgE, only a few were negative for both, and virtually none had only IgE. The prevalence of asthma and hay fever was dramatically higher in children with high IgG4 and IgE compared to children who only mounted IgG4 or low IgG4 and IgE responses. These results appear to recapitulate di¡erent stages of in vivo Th2-dependent sequential switching from IgG4 to IgE. These patterns of Th2-induced antibody responses may warrant a rede¢nition of the notion of allergen sensitization.
Multiple lines of evidence implicate T helper (Th)2 responses in the initiation and/ or ampli¢cation of the pathogenetic processes that result in human allergic disease. Expression of the Th2 cytokines, interleukin (IL)4 and IL13, has been linked to allergic lung in£ammation, rhinitis, and atopic dermatitis, and to dysregulation of immunoglobulin (Ig)E responses (Oettgen & Geha 2001, Vercelli 2001, Vercelli 2002). The role of IgE as a central e¡ector molecule of allergic reactions is undisputed (Gould et al 2003). Understanding the molecular events leading to class switch recombination (CSR) to IgE is therefore critical to develop e¡ective strategies to control and possibly prevent allergic disorders.
In an attempt to de¢ne the mechanisms which regulate CSR to IgE in human B
lymphocytes, a molecular assay to detect recombination between switch (S)m and Se regions was developed which also allowed cloning and sequencing of Sm/Se switch products (Shapira et al 1991, 1992). This approach led to the identi¢cation of genomic DNA fragments in which the Sm and the Se regions were separated by inserts derived from Sg4 (Jabara et al 1993). These results suggested that human CSR can occur sequentially from IgM to IgE via IgG4, a notion that was later independently con¢rmed through a di¡erent approach (Zhang et al 1994). Sequential IgM/IgG4/IgE switching provided a mechanistic explanation for several hitherto unexplained ¢ndings, i.e. the presence of both IgE and IgG4 in IL4-stimulated cultures (Lundgren et al 1989), and the simultaneous production of IgM, IgG4 and IgE in clonal B cell populations stimulated with IL4 and anti-CD40 mAb or activated CD4+ T cell clones (Gascan et al 1991a,b). These data also linked both IgE and IgG4 to Th2 responses regardless of the profound di¡erences that exist between their e¡ector functions. Indeed, IgG4 is functionally monovalent, does not ¢x complement and binds weakly to Fce receptors (Aalberse et al 1983, Schuurman et al 1999, van der Zee et al 1986). Thus, unlike IgE, antigen binding by IgG4 is expected to have no harmful consequences.
IgE and IgG4 levels are known to be co-regulated in vivo in certain diseases, such as chronic parasitic infections. Of note, typical allergic reactions are rare in helminth-infected patients, even though FceRI-bearing cells are sensitized with anti-parasite IgE and are exposed, often continuously, to parasite antigens (Vercelli et al 1998). Inhibition of allergic reactivity has been attributed to ‘blocking antibodies’, predominantly found in the IgG4 subclass (Hussain et al 1992). IgG4 are unusually predominant among anti-¢larial antibodies, representing 50^95% of the total IgG response (Kurniawan et al 1993). Depletion of IgG4 by adsorption on anti-IgG4 a⁄nity columns speci¢cally removed the blocking activity from the sera of micro¢laremic patients (Hussain et al 1992), and IgG4 inhibited the binding of anti-Schistosoma mansoni IgE by over 96% (Rihet et al 1992). Furthermore, a potential role of blocking IgG4 antibodies in allergen immunotherapy was recently suggested (Akdis et al 1998). Indeed, IgG4 with blocking activity is detectable in sera from patients receiving immunotherapy for insect venom and house dust mite hypersensitivity (Akdis & Blaser 1999).
The discovery of sequential IgM/IgG4/IgE switching, and the possibility that IgG4 may act to block IgE-mediated reactions, prompted us to investigate the mechanisms which regulate allergen-speci¢c IgE and IgG4 antibody responses in vitro and in vivo. Our studies on the molecular regulation of e and g4 germline transcription have been discussed elsewhere (Agresti & Vercelli 1999, 2002, Monticelli et al 2002, Thienes et al 1997). Here, we present novel in vitro data in support of preferential IgG4 expression by IL4-stimulated B cells, as well as an in vivo analysis of Th2 antibody responses in a population from rural areas of Germany, Austria and Switzerland thoroughly de¢ned for lifestyle, environmental exposures, and allergic phenotypes (Braun-Fahrlander et al 2002, Riedler et al 2001). Although the ALEX population includes both farmers and non-farmers, our analysis focused exclusively on children from non-farming families, so as to avoid complex e¡ects of exposure to microbial products (Vercelli 2003) on allergen sensitization and type of antibody response. We examined the prevalence of allergen-speci¢c IgE and IgG4 responses, the patterns of IgG4 and IgE co-expression, and the role of IgG4 antibodies in disease pathogenesis. To our surprise, we found that virtually every individual in the population mounted allergen-speci¢c Th2 antibody responses, as indicated by the expression of IgG4, but only a minority of subjects concomitantly expressed IgE. Allergic disease was restricted to the latter group. The Th2 antibody response patterns revealed by our studies may warrant a rede¢nition of the notion of allergen sensitization.
Methods
Peripheral blood mononuclear cells were isolated from normal non-allergic donors by density gradient centrifugation, resuspended at 5^10x106 cells/ml in RPMI1640^10% human AB+ serum, and adhered overnight in plastic Petri dishes. Non-adherent cells (5^10x106 cells/ml) were then incubated on ice for 30 min in the presence of anti-CD3 mAb (OKT-3), washed and incubated for 30 min with magnetic beads coated with goat anti-mouse IgG (Dynal, 8^10:1 bead:cell ratio) at 4 8C with slow rotation. CD3+ cells were then removed using a magnet. This procedure was performed twice. Negative selection with mAb OKM-1 was used to remove CD11b+ cells. The cell populations thus isolated contained 95% CD19+ B cells, as assessed by immuno£uorescence. To isolate sIgD+ B cells, cells were washed, resuspended in labelling bu¡er (PBS, pH 7.2, 2 mM EDTA) and incubated on ice for 30 min with biotin-conjugated goat anti-human IgD (Sigma: 10 mg/sample), washed with labelling bu¡er and then incubated for 20 min on ice with streptavidin-conjugated microbeads (Miltenyi: 10 ml/107 cells). After washing in separation bu¡er (PBS pH 7.2, 0.5% BSA, 2 mM EDTA), sIgD+ cells were collected using a magnetic cell separator. Immuno£uorescence analysis showed that the cell populations thus obtained contained 95% sIgD+ B cells.
In vitro Ig production
sIgD+ B cells were incubated with IL4 (R&D Systems, 10 ng/ml) and/or anti- CD40 mAb 626.1 (5 mg/ml) for 14 days. Culture supernatants were then harvested and assessed for Ig concentrations by enzyme-linked immunosorbent assay (ELISA). For IgE, 96-well plates were coated with anti-IgE mAb 7.12 and 4.15 (ATCC HB-236 and HB-235, 2 mg/ml in 0.1 M carbonate bu¡er, pH 9.0) overnight at room temperature. Wells were then washed twice with PBS^0.05% Tween and blocked with 2% milk^PBS^0.01% azide for 4 h at room temperature. After extensive washing with PBS^0.05% Tween, dilutions of an IgE standard curve (Hybritech) and samples were added to the wells overnight at room temperature. Following extensive washing with PBS^0.05% Tween, a horseradish peroxidase-conjugated rabbit anti-human IgE antiserum (DAKO, 1:1000 in PBS^1% milk^0.05% Tween) was added to the wells for 4 h at room temperature. After washing 10 times with PBS^0.05% Tween, the reaction was developed by incubation with ortho-phenylendiamine (Sigma) for approximately 20 min at room temperature in the dark. After stopping the reaction with 10% sulphuric acid, OD was read at 490 nm. Secretion of IgG subclasses was evaluated using commercially available ELISA kits (The Binding Site). The limits of sensitivity for the Ig assays were: 0.42 (IgG1), 7.35 (IgG2), 0.29 (IgG3), 0.44 (IgG4) and 0.2 (IgE) ng/ml. Control cultures for the evaluation of preformed Ig were set up in the presence of cycloheximide (100 mg/ml). Net Ig synthesis was calculated by subtracting the Ig concentrations detected in cycloheximide-treated cultures from the values found in untreated cultures.
Epidemiological studies
Subjects included in this study were participants in a cross-sectional survey conducted by the Allergy and Endotoxin (ALEX) Study Team in rural areas of Germany, Austria and Switzerland which included both farming and non- farming households (Braun-Fahrlander et al 2002, Riedler et al 2001). Brie£y, parents of 3504 children in school grades 1^6 were invited to answer a questionnaire on respiratory and allergic diseases. 2618 (75%) of the parents elected to participate and were asked to consent to further testing. 1406 (54%) consented. From this group, all children from farming families and a random sample of children from non-farming families from the same rural areas were invited to continue testing. The ¢nal group was restricted to children born in Germany, Austria or Switzerland and who were nationals of those countries (n ¼ 812). Farmers’ children were de¢ned as children whose parents answered ‘yes’ to the question ‘does your child live on a farm?’ The results reported here are limited to children of non-farming families (n ¼ 493).
For serum IgE measurements, each sample was ¢rst tested against a panel of aeroallergens (mixed-grass pollen, birch pollen, mugwort pollen, Dermatophagoides pteronyssinus (Derp), cat dander, dog dander and Cladosporium herbarum) by £uorescence enzyme immunoassay (FEIA, CAP, Pharmacia). In children who had a positive result to this panel, speci¢c IgE responses to timothy grass pollen, cat dander and Derp were measured. Allergen-speci¢c IgE were expressed as kU/L, and the limit of detectability was 0.35 kU/L (Platts-Mills et al 2003). Undetectable samples were assigned a value of 0.30 kU/L.
IgG4 antibodies speci¢c for timothy grass pollen, cat dander and Derp were measured in 487 sera using FEIA (CAP, Pharmacia) and diluting samples 1:10 in diluent provided with the kit. Allergen-speci¢c IgG4 were expressed as mg/L. The limit of sensitivity of the assay was 15 mg/L, and undetectable samples were assigned a value of 10 mg/L.
Disease was de¢ned as ‘ever hay fever’ and ‘ever asthma’.
Statistical analysis
This was performed using the Statistical Package for the Social Sciences (SPSS) for UNIX, version 6.1.3. Values of allergen-speci¢c IgG4 were log- normally distributed and results are reported as geometric means. Due to the large number of undetectable allergen-speci¢c IgE values, detectable IgE values were grouped into half-log intervals and analysed using w-square analysis and contingency tables.
Results
The isotype speci¢city of IL4-dependent CSR in human B cells is controversial. The combination of IL4 and CD40 cross-linking was reported to induce IgE and all IgG subclasses, except IgG2, in tonsillar na|« ve B cells (Fujieda et al 1995). By contrast, molecular analysis identi¢ed IgG4 as the typical IL4-dependent g subclass (Jabara et al 1993), even though other IgG isotypes appear to be occasionally
targeted (Zhang et al 1994). We felt that the source of human B cells used for studies of CSR isotype speci¢city may be a critical variable. Most studies were performed using tonsil B cells. However, this tissue is a site of intense inflammation and cytokine production (Agren et al 1996). It is therefore conceivable that a signi¢cant proportion of tonsil B cells, while still expressing IgM and/or IgD on their membrane and thus na|« ve by commonly accepted criteria, may nonetheless be engaged in cytokine-dependent remodelling of the relevant immunoglobulin loci. In order to analyse CSR in truly resting human B cells, we isolated na|« ve sIgD+ B lymphocytes from peripheral blood, rather than tonsils, and stimulated in vitro with IL4 and/or anti-CD40 mAb 626.1 for 12^14 days. Total IgE and IgG4 secretion in cell culture supernatants was assessed by ELISA. Figure 1 shows the mean SEM of results obtained in six consecutive experiments. Both IgG4 and IgE were strongly (18- and 125-fold, respectively) up-regulated in cultures treated with both IL4 and anti-CD40 mAb. In contrast, only modest enhancement was detected for IgG1 and IgG3, while IgG2 secretion remained una¡ected. As expected, IL4 or CD40 cross-linking alone did not up- regulate Ig secretion. These results demonstrate that, consistent with previous molecular evidence (Jabara et al 1993), the IgG4 subclass is preferentially induced in na|« ve B cells in which both the IL4 and the CD40 receptor have been engaged.
Prevalence of allergen-speci¢c Th2 antibody responses in the ALEX population We then moved on to analyse the in vivo prevalence of antigen-speci¢c IgG4 and IgE responses, the molecular signature of CSR induced by Th2 cytokines. To this end, we measured levels of IgE and IgG4 speci¢c for three common inhalant allergens (timothy grass pollen, Derp and cat dander) in sera from ALEX children. The ALEX population includes both farmers and non-farmers (Braun- Fahrlander et al 2002, Riedler et al 2001). However, we limited our analysis to children from non-farming families (n ¼ 493) so as to avoid potential e¡ects of farm-related environmental exposures. Table 1 (top) shows that 98% of children in the non-farming ALEX population produced IgG4 and/or IgE to one or more of the test allergens (‘Any’ group). Interestingly, the majority of the ALEX children (64.8%) had pure IgG4 responses, whereas only 33.5% of the population expressed both IgE and IgG4 to at least one allergen, and only one child mounted a pure IgE response. These results show allergen-speci¢c, Th2-dependent antibody responses occur in the totality of the non-farming ALEX population. In two- thirds of the children, these responses are uncoupled, i.e. IgG4 antibodies are secreted in the absence of IgE.
We then investigated whether allergen-speci¢c patterns could be detected in IgG4/IgE responses. Table 1 (bottom) demonstrates that only 10.1% of the ALEX children produced IgE as well as IgG4 to cat dander. By contrast, 26.3% of the children had both IgE and IgG4 to timothy grass pollen and 15% had both isotypes to Derp. Of note, timothy grass pollen was the only allergen against which a signi¢cant portion of the population (29.8%) failed to mount a Th2 antibody response. Thus the nature of the allergen, and/or the environmental context in which allergen exposure occurs, appear to have an impact on Th2 response
Th2 antibody responses and disease
In an attempt to characterize the role of Th2 antibody responses in the pathogenesis of asthma and allergy, we then investigated the association between levels of allergen-speci¢c Th2 antibodies and incidence of allergic disease in the non-farming ALEX population. Figure 2 shows that children with high levels of allergen-speci¢c IgE (sum of IgE against the three test allergens) had high incidence of both hay fever and asthma. The IgG4 sum directly correlated with IgE sum (Spearman r ¼ 0.51, P50.001), i.e. the highest IgG4 levels were found in the high IgE groups. However, strong IgG4 responses were also found in the absence of both IgE expression and disease. Overall these results indicate IgG4 responses are not pathogenic, but they are not protective either when coupled with vigorous IgE production.
Discussion
‘Allergen sensitization’ is commonly used as a synonymous term for allergen- speci¢c IgE responses, and these are in turn considered as the outcome of antibody-mediated Th2 immunity. According to this view, Th2 responses would be both relatively infrequent and usually pathogenic. Our current results highlight a di¡erent scenario, in which Th2 antibody responses to allergens (whose signature is the expression of IgG4 as well as, or instead of, IgE) occur frequently and overall invariably in the population. Most importantly, Th2 responses are mostly restricted to the IgG4 isotype, and are non-pathogenic.
The scenario we propose has several implications. The traditional notion that allergens are antigens to which healthy individuals do not develop detectable responses is not supported by our data. Indeed, the majority of the ALEX children mounted IgG4 responses to the three allergens we tested, mostly in the absence of a concomitant IgE response and disease. These children would have been considered allergen non-responders, had we not measured allergen-speci¢c IgG4. Similar conclusions were recently drawn upon examination of antibody expression pro¢les in individuals exposed to domestic animals. High levels of exposure to cat allergen were found to be accompanied by an IgG and IgG4 antibody response without allergic symptoms or risk of asthma (Platts-Mills et al 2001, Platts-Mills et al 2003). Of note, this Th2 response was interpreted as a form of tolerance (Platts-Mills et al 2001). However, in as much as tolerance represents a failure to respond to an antigen, we would argue that expression of allergen- speci¢c IgG4 without IgE would rather represent an intermediate Th2 response, i.e. a situation in which IgM/IgG4/IgE sequential switching induced by Th2 cytokines is arrested at the IgG4 stage.
In this context, allergens may then be better de¢ned as antigens which, because of molecular signatures we have not yet deciphered, evoke Th2 antibody responses (IgG4, with or without IgE). Most allergen-exposed individuals become ‘sensitized’ biologically, even though sensitization may remain clinically silent. That the immune system is quite prompt in mounting Th2 responses to common inhalants suggests a default pre-programming which may have been shaped by evolution, and is supported by recent evidence from animal models (Dabbagh et al 2002, Eisenbarth et al 2002).
The di¡erent clinical outcome of Th2 antibody responses begs the question, what determines whether the response will include both IgG4 and IgE (and will be potentially pathogenic), or only IgG4 (and will be harmless). The nature of the antigen appears to play a role. Furthermore, it is possible that gene^environment interactions, i.e. a complex interplay between genetic makeup and environmental exposure, may tip the balance between different kinds of Th2 responses. The ALEX population, with
FIG. 3. A model of the development of allergen-speci¢c, Th2-mediated antibody responses in humans.
its well characterized pro¢les of environmental exposures, will be ideal to test the multiple facets of these hypotheses.
One puzzling element emerging from our analysis is the lack of protection associated with IgG4 responses. This ¢nding was unexpected, because the in vivo data from patients with chronic parasitic infections suggested a strong blocking e¡ect of IgG4 (Hussain & Ottesen 1986, Hussain et al 1992, Ottesen et al 1985), and similar patterns were observed following bee venom (Akdis et al 1998) and birch pollen (Visco et al 1996) immunotherapy. This discrepancy may result from di¡erences between the immunization routes, target organs, and regulatory networks engaged in these conditions. Furthermore, inhalant and parasite-derived
allergens are likely to di¡er in their biochemical structure and the biological context within which they are presented to the immune system.
We conclude by proposing a model for the generation of allergen-speci¢c antibody responses predicated on our current results (Fig. 3). Allergen-speci¢c, Th2-mediated antibody responses represent the outcome of two fundamental choices made by a developing CD4+ Th cell precursor. The ¢rst choice is whether to di¡erentiate along the Th1 or Th2 pathway. The fact that allergens frequently elicit Th2-mediated, IL4-dependent antibody responses of the IgE and/or IgG4 class is likely to re£ect inherent biochemical properties of these molecules, as well as route and context of exposure. The second choice, perhaps a more complex one, is whether the ultimate outcome of allergen-speci¢c Th2 responses will be expression of IgG4 only, or both IgG4 and IgE. In the ¢rst case, the response will be clinically silent; in the second case, disease may ensue. The molecular mechanisms which activate dfferential isotype switching to IgE and IgG4 in B lymphocytes remain unde¢ned. Our results showing that potentially pathogenic IgE responses are not the inevitable outcome of Th2- mediated immunity should provide the impetus to de¢ne these mechanisms and devise approaches to redirect Th2 e¡ector functions toward intermediate responses and expression of the non-pathogenic IgG4 isotype.
Acknowledgments
This work was supported by National Institutes of Health grant HL67672 (to D.V.), by the Program for Genomic Applications Innate Immunity in Heart, Lung and Blood Disease from the National Heart, Lung and Blood Institute (U01-HL66803, to D.V.), and by the Southwest Environmental Health Sciences Center at the University of Arizona.
References
Aalberse RC, van der Gaag R, van Leeuwen J 1983 Serologic aspects of IgG4 antibodies. I. Prolonged immunization results in an IgG4-restricted response. J Immunol 130:722^726 Agren K, Andersson U, Litton M, Funa K, Nordlander B, Andersson J 1996 The production of immunoregulatory cytokines is localized to the extrafollicular area of human tonsils. Acta Otolaryngol (Stockh) 116:477^485 Agresti A, Vercelli D 1999 Analysis of g4 germline transcription in human B cells. Int Arch Allergy Immunol 118:279^281
Agresti A, Vercelli D 2002 c-Rel is a selective activator of a novel IL-4/CD40 responsive element in the human Ig g4 germline promoter. Mol Immunol 38:849^859 Akdis CA, Blaser K 1999 IL^10-induced anergy in peripheral T cell and reactivation by microenvironmental cytokines: Two key steps in speci¢c immunotherapy. FASEB J 13: 603^609 Akdis CA, Blesken T, Akdis M, Wˇtrich B, Blaser K 1998 Role of interleukin 10 in speci¢c immunotherapy. J Clin Invest 102:98^106
Braun-Fahrlander C, Riedler J, Herz U et al 2002 Environmental exposure to endotoxin and its relation to asthma in school-age children. N Engl J Med 347:869^877
Dabbagh K, Dahl ME, Stepick-Biek P, Lewis DB 2002 Toll-like receptor 4 is required for optimal development of Th2 immune responses: role of dendritic cells. Immunol 168: 4524^4530
Eisenbarth SC, Piggott DA, Huleatt JW, Visintin I, Herrick CA, Bottomly K 2002
Lipopolysaccharide-enhanced, Toll-like receptor 4-dependent T helper cell 2 responses to inhaled antigen. J Exp Med 196:1645^1651
Fujieda S, Zhang K, Saxon A 1995 IL-4 plus CD40 monoclonal antibody induces human B cells g subclass-speci¢c isotype switch: switching to g1, g3, and g4, but not g2. J Immunol 155:2318^ 2328
Gascan H, Gauchat J-F, Aversa G, van Vlasselaer P, de Vries JE 1991a Anti-CD40 monoclonal antibodies or CD4+ T cell clones and IL-4 induce IgG4 and IgE switching in puri¢ed human B cells via di¡erent signaling pathways. J Immunol 147:8^13
Gascan H, Gauchat J-F, Roncarolo MG, Yssel H, Spits H, de Vries JE 1991b Human B cell clones can be induced to proliferate and to switch to IgE and IgG4 synthesis by interleukin 4 and a signal provided by activated CD4+ T cell clones. J Exp Med 173:747^750
Gould HJ, Sutton BJ, Beavil AJ et al 2003 The biology of IgE and the basis of allergic disease.
Annu Rev Immunol 21:579^628
Hussain R, Ottesen EA 1986 IgE responses in human ¢lariasis. IV. Parallel antigen recognition of IgE and IgG4 subclass antibodies. J Immunol 136:1859^1863
Hussain R, Poindexter RW, Ottesen EA 1992 Control of allergic reactivity in human ¢lariasis predominant localization of blocking antibody to the IgG4 subclass. J Immunol 148: 2731^2737
Jabara HH, Loh R, Ramesh N, Vercelli D, Geha RS 1993 Sequential switching from m to e via g4
in human B cells stimulated with IL-4 and hydrocortisone. J Immunol 151:4528^4533 Kurniawan A, Yazdanbakhsh M, van Ree R et al 1993 Di¡erential expression of IgE and IgG4
speci¢c antibody responses in asymptomatic and chronic human ¢lariasis. J Immunol 150: 1941^1950
Lundgren M, Persson U, Larsson P et al 1989 Interleukin 4 induces synthesis of IgE and IgG4 in human B cells. Eur J Immunol 19:1311^1315
Monticelli S, Ghittoni R, Kabesch M, Vercelli D 2002 Myb proteins repress human e germline
transcription by inhibiting STAT6-dependent promoter activation. Mol Immunol 38: 1129^1138
Oettgen HC, Geha RS 2001 IgE regulation and roles in asthma pathogenesis. J Allergy Clin Immunol 107:429^440
Ottesen EA, Skvaril F, Tripathy SR, Poindexter RW, Hussain R 1985 Prominence of IgG4 in the IgG antibody response to human ¢lariasis. J Immunol 134:2707^2712
Platts-Mills T, Vaughan J, Squillace S, Woodfolk J, Sporik R 2001 Sensitisation, asthma, and a modi¢ed Th2 response in children exposed to cat allergen: a population-based cross-sectional study. Lancet 357:752^756
Platts-Mills TA, Erwin EA, Allison AB et al 2003 The relevance of maternal immune responses to inhalant allergens to maternal symptoms, passive transfer to the infant, and development of antibodies in the ¢rst 2 years of life. J Allergy Clin Immunol 111:123^130
Riedler J, Braun-Fahrlander C, Eder W et al 2001 Early life exposure to farming environment is essential for protection against the development of asthma and allergy: A cross-sectional survey. Lancet 358:1129^1133
Rihet P, Demeure CE, Dessein AJ, Bourgois A 1992 Strong serum inhibition of speci¢c IgE correlated to competing IgG4, revealed by a new methodology in subjects from a S. mansoni endemic area. Eur J Immunol 22:2063^2070
Schuurman J, van Ree R, Perdok GJ, van Doorn HR, Tan KY, Aalberse RC 1999 The apparent
monovalency of human IgG4 is due to bispeci¢city. Int Arch Allergy Immunol 118:187^189
Shapira SK, Jabara HH, Thienes CP et al 1991 Deletional switch recombination occurs in IL-4 induced isotype switching to IgE expression by human B cells. Proc Natl Acad Sci USA 88:7528^7532
Shapira SK, Vercelli D, Jabara HH, Fu SM, Geha RS 1992 Molecular analysis of the induction of IgE synthesis in human B cells by IL-4 and engagement of CD40 antigen. J Exp Med 175:289^
Thienes CP, De Monte L, Monticelli S, Busslinger M, Gould HJ, Vercelli D 1997 The transcription factor B cell-speci¢c Activator Protein (BSAP) enhances both IL-4- and CD40-mediated activation of the human e germline promoter. J Immunol 158: 5874^5882
van der Zee JS, van Swieten P, Aalberse RC 1986 Serologic aspects of IgG4 antibodies. II. IgG4 antibodies form small, nonprecipitating immune complexes due to functional monovalency. J Immunol 137:3566^3571
Vercelli D 2001 IgE and its regulators. Curr Opin Allergy Clin Immunol 1:61^65
Vercelli D 2002 Genetics of IL-13 and functional relevance of IL-13 variants. Curr Opin Allergy Clin Immunol 2:389^398 Vercelli D 2003 Learning from discrepancies: CD14 polymorphisms, atopy and the endotoxin switch. Clin Exp Allergy 33:153^155
Vercelli D, De Monte L, Monticelli S, Di Bartolo C, Agresti A 1998 To E or not to E? Can an IL-4-induced B cell choose between IgE and IgG4? Int Arch Allergy Immunol 116:1^4
Visco V, Dolecek C, Denepoux S, Le Mao J, Guret C, Rousset F, Guinnepain MT, Kraft D, Valenta R, Weyer A, Banchereau J, Labecque S 1996 Human IgG monoclonal antibodies that modulate the binding of speci¢c IgE to birch pollen Bet v 1. J Immunol 157:956^962
Zhang K, Mills FC, Saxon A 1994 Switch circles from IL-4-directed e class switching from human B lymphocytes - Evidence for direct, sequential and multiple step sequential switch from m to e Ig heavy chain gene. J Immunol 152:3427^3435
DISCUSSION
Galli: Before I ask Hannah Gould to comment further on the local production of IgE, I’d like us to keep in mind a couple of observations. One is that some patients with anaphylaxis are not atopic. The second is to remind us about a point that Hugh Sampson mentioned earlier, regarding his studies of food allergy. Patients who developed what appeared to be distinct organ-speci¢c patterns of symptoms to two di¡erent foods had detectable levels of circulating IgE against both of the foods: that is, it wasn’t simply that there were not systemic levels of IgE for this food or that food. Now I would like Hannah Gould to comment on the antigen- speci¢city of local as opposed to systemic IgE.
Gould: It is true that some patients with anaphylaxis are not atopic. I said before that locally synthesized IgE contains a signi¢cantly higher ratio of speci¢c/total IgE than serum IgE from the same patient. In fact in one of these patients no IgE antibody at all could be detected in serum whereas it represented a sizeable fraction of the IgE synthesized in biopsies from the nasal mucosa (Smurthwaite et al 2001). In earlier work it was shown that allergen-speci¢c IgE could be often be measured in nasal secretions but not in serum from the same patients (reviewed in Durham 1997). IgE in nasal secretions was taken to represent IgE produced in the tissue.
This seems to have been justi¢ed in view of our results with the tissues themselves.
The recent results in our studies of asthma patients also support the local production of IgE. A sizeable proportion (up to 30%) of asthmatics are non- atopic or intrinsic asthmatics, with no detectable antibodies to any of the common allergens in the conventional skin prick tests or radioallergosorbent test (RAST). However, bronchial biopsies from these patients exhibit IgE e heavy-chain mRNA and FceRI a-chain mRNA (Ying et al 2001, Humbert et al 1999), implying that IgE is synthesized, secreted and bound to mast cells in the tissue.
I would like to mention another observation of possible relevance to the pathogenesis and localization of the allergic response. The studies of IgE VH sequences I mentioned earlier yielded another interesting result, the over- representation of a minor VH family, VH5, in the nasal biopsies from hay fever patients (H.A. Coker, S.R. Durham, H.J. Gould, unpublished results). The total repertoire of VH genes is around 50 and the VH3 family is the largest family, represented in 40% of the immunoglobulins expressed in normal human serum. The VH5 family has only 1 or (in half the population) 2 members and is represented by 3% of the immunoglobulins in serum. In contrast to normal immunoglobulins, VH5 is represented in 8% of serum IgE and in 18% of the IgE in the nasal biopsies of hay fever patients (H.A. Coker, S.R. Durham, H.J. Gould, unpublished results). Similar over-representation of VH5 has been observed in bronchial biopsies from an asthma patient (Snow et al 1997) and the circulation of patients with atopic dermatitis (van der Stoep et al 1993).
The distribution of replacement mutations, those which lead to amino acid substitutions, is also abnormal in the mucosal tissues, being skewed towards the framework regions rather than the complementarity-determining regions, as seen in the antibodies to conventional antigens. Together the over- representation of the expressed IgE VH5 sequences and the pattern of somatic mutations in VH5 suggest the activity of a local B cell superantigen that binds to VH5.
It is possible therefore that B cell superantigens are involved in the pathogenesis and localization of allergic disease. Whether allergens themselves are superantigens or the allergens are able to stimulate an IgE response in the tissues by virtue of local superantigens eliciting hypersensitivity (Genovese et al 2003) is an important question. This might relate to organ-speci¢c responses, as it is conceivable that a complex between a speci¢c allergens and a tissue-speci¢c protein could constitute a superallergen and elicit organ-speci¢c responses to food allergens.
Galli: The question for people doing this kind of investigation is whether there can be a detailed analysis and comparison of the antigen speci¢city and other properties of the IgE antibodies that are present systemically, as opposed to those that are present locally. This is needed to answer the question of whether, given the types of allergens that may be present at a particular site, the locally represented IgE may provide a higher level of reactivity at that site, and thereby explain the local expression of disease. There are other possible explanations of the clinical phenomena, of course. Perhaps at this point such detailed comparative studies of the systemic and the local antibody have not been done.
Gould: The amount of information is perhaps limited, but as I have said before we have shown in our work with hay fever patients that the ratio of speci¢c to total IgE is signi¢cantly greater in the IgE produced in the nasal mucosa than in serum IgE, and in some patients we could only detect speci¢c IgE as a product of local IgE synthesis (Smurthwaite et al 2001). There is also the earlier evidence that in some patients IgE antibodies against allergens are undetectable in serum but detectable in nasal secretions (reviewed in Durham et al 1997). IgE produced in the tissue can di¡use out of the tissue in two possible directions, into the circulation and/or into the secretions. It is likely that the relative rate of di¡usion in these two directions di¡ers from one patient to the next, perhaps determined by the proximity of the particular B cells (or B cell clones) to each of the surfaces.
Further evidence for the production of speci¢c IgE in the target organ comes from measurements in secretions versus serum as a function of time after exposure to allergens. It has been shown that the appearance of speci¢c IgE antibodies in the serum is delayed relative to the appearance in nasal secretions (more representative of the local response than the serum) upon exposure to allergen after a period of allergen avoidance (Sensi et al 1994), pointing to the tissue as the source of speci¢c IgE.
Marone: I would like to comment on the VH5 family, which is apparently overexpressed locally. This is a fascinating observation and also relates to the possibility, mentioned by Hannah Gould, that endogenous, bacterial or viral superantigens can activate IgE on the surface of human basophils and mast cells through the VH family. During the last years we have shown that several of the immunoglobulin superantigens, such as protein A, protein Fv and gp120, basically interact with IgE VH3+ (Genovese et al 2000, 2003, Patella et al 2000). This interaction induces the release of mediators and cytokines from human basophils and mast cells. Interestingly, the VH3 is the most represented VH family in the repertoire of human immunoglobulin. This opens the possibility that the in vivo exposure to certain immunoglobulin superantigens can induce a natural selection of the VH family.
Schwartz: A useful control experiment might be to look at another tissue that has been sensitized, such as the skin. If you do this, do you see the new Ig synthesis there in B cells, or is it really speci¢c for the target organ?
Gould: This is an important question, which we would like to study. We would need to get ethical approval and then do the biopsies, so it would take a little time before we can answer that question.
MacGlashan: You said that the local speci¢c to total IgE ratio was di¡erent to that in the circulation. The local was 30^70%; what did you ¢nd in the circulation?
Gould: We had about 10% in the circulation, and 30^70% in the nose.
Mˇller: It was mentioned just a few minutes ago that not all anaphylaxis is atopic. In fact, there have been quite a few studies on IgG4 in venom anaphylaxis, especially in relation to beekeepers. In highly exposed people like beekeepers, we ¢nd very high levels of IgG4 but little IgE. By passive immunotherapy with beekeeper g globulin, several groups have shown that it is possible to protect even allergic patients. However, we have looked in a number of venom-allergic patients at IgG4-speci¢c antibodies and the relationship between IgE and IgG4 in serum taken directly before a sting challenge. There we could not ¢nd a clear correlation between protection and IgG4.
Galli: Donata Vercelli, I’d like to raise the question of whether there are su⁄cient numbers of subjects in the ALEX study to do a meaningful study of anaphylaxis in that cohort. What are your views?
Vercelli: The ALEX group involves some 800 children, which may not be enough to achieve statistical power, given the low frequency of anaphylaxis. However, the same group is now recruiting another population, which will be named Parsifal, and will involve the same countries plus Sweden. They are intending to recruit a total of 6000 or so children, so it may well be possible to study anaphylaxis in this group.
Simons: Yes. In my presentation I will describe a study in which we tried to address the prevalence of anaphylaxis in a geographically de¢ned population of children. We found that although there was some variation with age, 1.44% of the children had epinephrine dispensed for out-of-hospital use. The highest epinephrine dispensing rate, 5.3%, was found in boys aged 12^17 months (Simons et al 2002). This is really the only data set that exists on the prevalence of anaphylaxis from all triggers in children. As mentioned previously, children are seriously under-represented in retrospective studies of anaphylaxis from all triggers in all ages (Yocum et al 1999, Kemp et al 1995), and two of the three paediatric studies published of anaphylaxis from all triggers are small, involving 55 and 76 children, respectively (Dibs & Baker 1997, Novembre et al 1998). Anaphylaxis does seem to be increasing, and it is in the younger patients that this increase is most signi¢cant.
Lee: This sort of study, with large cohorts, is very powerful. But I would also like to encourage the collection of data on a longitudinal basis. Taking data in one snapshot of time can be misleading. When you have cohorts which you are following up for a long time, it is enormously powerful to have the correlation of immunology with clinical patterns.
Vercelli: I couldn’t agree more. In fact, we are going to do something very similar to the study I just showed for the ALEX group using a Tucson population that has been followed for 25 years. The children were enrolled before they were born, through their parents. The problem is that this is a somewhat smaller population, so there may be an issue of statistical power. But you are absolutely right, longitudinal studies are ideal.
Finkelman: It may not be clear that IgG4 is protective against allergy, but if you look at the reverse side of the coin, protection against worm infections, there are data in humans infected with schistosomiasis that indicate that the IgE:IgG4 ratio tends to correlate better with protection than IgE levels alone. This argues that IgG4 can have some mechanism of down-regulating allergic responses. As you know there is some analogy between IgG4 in humans and IgG1 in mice, in that both can be induced by IL4. With IgG1 there is evidence that in vitro it takes more IL4 to induce a good IgE response than to induce a good IgG1 response. Is this true for IgG4 versus IgE in humans? Related to that, I remember a paper suggesting that gamma interferon was more suppressive of IgE than IgG4 (Akdis et al 1997, Carballido et al 1994). This suggests that you’d see a higher G4:E ratio when both IL4 and IFNg are being produced. Has this been replicated? Is it generally believed?
Vercelli: It’s a complex situation. In vitro at least, when we induce IgE and IgG4 with anti-CD40 antibody and IL4, the response for IgE is much stronger. It goes from virtually zero to a signi¢cant amount, which ends up as about a 200-fold increase. However, serum IgG4 levels are much higher than IgE. Based on the data from the parasite and beekeeper studies, we were expecting some protection, and we were surprised when we did not ¢nd any. However we also saw very di¡erent patterns with di¡erent allergens. I think it is hard to answer this question. In a sense, IgG4 is more complex than IgE: if you take away IL4 in the mouse, IgG1 stays there. There are probably more ways to induce IgG1 than just IL4. IgG4 may have similar characteristics.
Finkelman: As you vary the amount of IL4 that you add to the in vitro culture with the CD40 stimulation, do you vary the IgG4:IgE ratios?
Vercelli: We haven’t done this extensively. What you are asking me is a question about thresholds, and I don’t have an answer yet.
Sampson: I wonder whether we need to go to another level of complexity. We looked at children who were allergic to eggs, and we were trying to do mapping of the IgE binding sites on ovomucoid. Children who had persistent egg allergy - that is, who did not appear to be able to outgrow it as most children do - had IgE binding to speci¢c linear epitopes on the ovomucoid. When we looked at those children for IgG binding to those same epitopes, they bound IgG as well.
However, when we looked for IgG binding to linear epitopes in children who ‘outgrew’ their egg allergy, they had none. If we did an assay, however, in which we measured IgG and IgE to native ovomucoid, we saw these high levels. I almost wonder whether we have to go and look at speci¢c epitopes to look for protection and non-protection.
Vercelli: I agree. This is probably the reason why there are discrepancies. It is possible that we don’t see protection because not enough IgG4 is produced against the epitopes IgE reacts with. What I was intrigued by is that if I believe the data (and I do), the most important conclusion is that the complexity of Th2 responses in vivo is much greater than initially anticipated, and their frequency is much higher. So this idea that there are only a few people who are Th2 responders is incorrect. Then the question becomes why is it that some people go one way or the other? This leads us straight into genetics. My anticipation is that there is a genetic component that shifts people one way or the other.
Schwartz: A critical issue here is how Th2 cells can educate a B cell to be either an IgG4 or an IgE producer. Could you expand more on this? One of the questions that comes to mind is if you take Th2 clones, can you ¢nd one clone that can educate B cells to switch to IgG4 and another one that would switch them to IgE? Or is this not where the regulation is taking place?
Vercelli: It is good that you are asking that: this is exactly what we are doing, trying to dissect this at a clonal level. To rephrase your question, what are the signals that make a response progress, or prevent it from progressing, from an IgG4 to an IgE response? A good candidate in the literature is IL10. This has been described as being a cytokine that is able to speci¢cally suppress IgE expression or secretion and increase IgG4. IL10 has also been an interesting cytokine in patients reacting to parasites. One remarkable observation by Tom Nutman was that peripheral blood mononuclear cells from ¢laria-infected patients, stimulated with antigen in vitro, showed very little proliferation. However, if an anti-IL10 blocking antibody was added, proliferation went through the roof. If he added IL10 in vitro, IgE went down. So IL10 had a blocking e¡ect. The e¡ect that we see with IL10 is very complex: we are now studying this extensively. IL10 has e¡ects at the T and B cell levels, and these e¡ects are opposite. IL10 blocks CD40L induction but has a very strong enhancing e¡ect on expression of IgE in a CD40-based system which bypasses CD40L. We don’t know about IgG4 yet, but I think IL10 may have something to do with this. IL10 would also be a good candidate to mediate the e¡ects of exposure, for obvious reasons: it is coming from APCs and macrophages and so on. It may be an interface between bacterial exposure and whatever mechanism it is that primes the immune response. Of course, IL10 is also produced by T regulatory cells, and these are likely to be critical to modulate allergic responses.
Gould: One thing that hasn’t been mentioned is cell proliferation. IL4 is very good at driving cell proliferation. There is evidence in the literature that more cycles of cell replication are required for the switch to IgG than to IgE (Tangye et al 2002). This could be related to what Fred Finkelman was saying about the concentration of IL4, and whether you get IgG4 or IgE. At relatively low IL4 concentrations there may be fewer cycles of cell proliferation so IgG4 would be favoured. Another reason for the delay in class switching to IgE is likely to be the sequential switching through other isotypes (reviewed in Gould et al 2003).
Ring: We are looking for many in£uences, and you mentioned ‘environmental factors’. What about the dose of allergen as the environmental factor? The more allergen around, the more IL2 or IL10 is formed. Low doses make IgE and high doses give rise to IgG4. This is the essence of immunotherapy.
References
Akdis CA, Blesken T, Akdis M et al 1997 Induction and di¡erential regulation of bee venom phospholipase A2-speci¢c human IgE and IgG4 antibodies in vitro requires allergen-speci¢c and nonspeci¢c activation of T and B cells. J Allergy Clin Immunol 99:345^353 Carballido JM, Carballido-Perrig N, Oberli-Schrammli A, Heusser CH, Blaser K 1994
Regulation of IgE and IgG4 responses by allergen speci¢c T-cell clones to bee venom phospholipase A2 in vitro. J Allergy Clin Immunol 93:758^767
Dibs SD, Baker MD 1997 Anaphylaxis in children: a 5-year experience. Pediatrics 99:E7 Durham SR, Gould HJ, Hamid QA 1997 IgE regulation in tissues. In: Vercelli D (ed) IgE regulation: molecular mechanisms. Wiley, Chichester, p 21^36
Genovese A, Bouvet JP, Florio G, Lamparter-Schummert B, Bjorck L, Marone G 2000 Bacterial immunoglobulin superantigen proteins A and L activate human heart mast cells by interacting with immunoglobulin E. Infect Immun 68:5517^5524
Genovese A, Borgia G, Bjorck L et al 2003 Immunoglobulin superantigen protein L induces IL-4 and IL-13 secretion from human Fc epsilon RI+ cells through interaction with the kappa light chains of IgE. J Immunol 170:1854^1861
Gould HJ, Sutton BJ, Beavil AJ et al 2003 The biology of IgE and the basis of allergic disease.
Annu Rev Immunol 21:579^628
Humbert N, Menz G, Ying S et al 1999 The immunopathology of extrinsic (atopic) and intrinsic (non-atopic) asthma: more similarities than di¡erences. Immunol Today 20:528^533
Kemp SF, Lockey RF, Wolf BL, Lieberman P 1995 Anaphylaxis. A review of 266 cases. Arch Intern Med 155:1749^1754
Novembre E, Cianferoni A, Bernardini R et al 1998 Anaphylaxis in children: clinical and allergologic features. Pediatrics 101:E8
Patella V, Florio G, Petraroli A, Marone G 2000 HIV-1 gp120 induces IL-4 and IL-13 release from human Fc epsilon RI+ cells through interaction with the VH3 region of IgE. J Immunol 164:589^595
Sensi LG, Piacentini GL, Nobile E et al 1994 Changes in nasal speci¢c IgE to mites after periods of allergen exposure-avoidance: a comparison with serum levels. Clin Exp Allergy 24:377^382 Simons FER, Peterson S, Black CD 2002 Epinephrine dispensing patterns for an out-of-hospital population: a novel approach to studying the epidemiology of anaphylaxis. J Allergy Clin Immunol 110:647^651
Smurthwaite L, Walker SN, Wilson DR et al 2001 Persistent IgE synthesis in the nasal mucosa of hay fever patients. Eur J Immunol 31:3422^3431
Snow RE, Chapman CJ, Frew CJ, Holgate ST, Stevenson FK 1997 Pattern of usage and somatic hypermutation in the VH5 gene segments of a patient with asthma: implications for IgE. Eur J Immunol 27:162^170
Tangye SG, Ferguson A, Avery DT, Ma CS, Hodgkin PD 2002 Isotype switching by human B cells is division-associated and regulated by cytokines. J Immunol 169:4298^4306
van der Stoep N, van der Linden J, Logtenberg T 1993 Molecular evolution of the human immunoglobulin E response: high incidence of shared mutations and clonal relatedness among epsilon VH5 transcripts from three unrelated patients with atopic dermatitis. J Exp Med 177:99^107
Ying S, Hunbert M, Meng Q et al 2001 Local expression of e germline gene transcripts (Ie) and RNA for the e heavy chain of IgE (Ce) in the bronchial mucosa in atopic and non-atopic asthma. J Allergy Clin Immunol 107:686^692
Yocum MW, Butter¢eld JH, Klein JS, Volcheck GW, Schroeder DR, Silverstein MD 1999
Epidemiology of anaphylaxis in Olmsted County: a population-based study. J Allergy Clin Immunol 104:452^456
0 comments:
Post a Comment