IgE, IgG1, and IgG4 Reactivity to Dermatophagoides pteronyssinus Glycosylated Extract in Allergic Patients

Background House dust mites are important allergen sources and some of these allergenic proteins may contain carbohydrate moieties, which are able to be isolated using lectins, as Concanavalin A (ConA). This study aimed to investigate allergenicity (IgE) and antigenicity (IgG1 and IgG4) of ConA-unbound and ConA-bound Dermatophagoides pteronyssinus (Dpt) crude extracts using sera of mite-allergic patients as well as inhibition capacity of antibody binding. Material and Methods We obtained mannose-enriched and mannose-depleted fractions from Dpt by ConA affinity chromatography. Both ConA-bound and ConA-unbound fractions were evaluated by ELISA and Western Blotting for specific IgE, IgG1, and IgG4 reactivity with sera obtained from 95 mite-allergic patients (DP+) and 92 nonallergic (NA) subjects. Inhibition ELISA was used to assess cross-reactivity between Dpt extract and its fractions. Results Among the DP+ patients, no difference was found between ConA-unbound and ConA-bound fractions regarding the levels of specific IgE, IgG1, and IgG4. Nonallergic subjects had the same levels of specific IgG1 to both ConA-unbound and ConA-bound fractions, although for specific IgG4, values were higher for ConA-bound. A positive correlation was found among specific IgE, IgG1, and IgG4 levels when Dpt was compared to ConA-unbound and ConA-bound fractions. Recognition of crude Dpt by IgE, IgG1, and IgG4 was highly inhibited by ConA-unbound and ConA-bound fractions. Western Blotting revealed a broad spectrum of bands ranging from 14 to 116 kDa recognized by specific IgE and IgG4. However, IgG1 reached higher frequency values on high molecular weight polypeptides. Conclusion ConA-unbound and ConA-bound fractions derived from D. pteronyssinus crude extract revealed important components involved in the IgE recognition in allergic patients as well as IgG1 and/or IgG4 in allergic and healthy subjects.


Introduction
House dust mites (HDM) have been described as important allergen sources causing respiratory allergies, such as asthma and rhinitis, and this association has led to the realization of several studies aiming to elucidate the role of the exposure to these allergens on the synthesis of components of the immune system, including cytokines, such as IL-4, IL-5, and IL-13 as well as different immunoglobulin classes [1]. In prevalence studies, mites from Pyroglyphidae (particularly Dermatophagoides pteronyssinus and Dermatophagoides farinae) and Echimyopodidae (Blomia tropicalis) families have been found in several countries, including areas in Latin America [2][3][4][5]. Allergens derived from these mites are characterized according to their biochemical and structural properties in association with the mechanism of IgE induction in the allergic response [6,7]. More than 30 allergens from HDM have been described so far and Der p 1 and Der p 2, from D. pteronyssinus, are responsible for more than 50% of IgE reactivity in patients sensitized to HDM [8], although these major allergens present distinct biochemical features. Der p 1 is a glycosylated protein with cysteine-protease function that exhibits high identity to Der f 1, from D. farinae, and it is involved in the barrier disruption and Th2 cell differentiation [9] while Der p 2 is likely associated with Tolllike receptors, particularly TLR-4, by triggering components in the innate immunity [10]. Consequently, the association of these findings and the use of these proteins on diagnosis and treatment of the allergic diseases have determined most of the approaches in clinical immunology.
The contribution of IgG antibodies on the allergic response has also been investigated, although the mechanism of activation as well as interaction with allergens is controversial. Correlations between specific IgG and IgE antibodies propose that IgG is similarly important in the immune response triggered by airborne allergens, suggesting that IgE should not be considered an exclusive marker for the allergy status [11]. In addition to that, IgG4 is an important marker for immunotherapy efficacy and may act as a blocking element in the allergen presentation while IgG1 should be involved in the upregulation of the IgE/B cell binding complex [12]. Despite this, reports using IgG and IgG1/IgG4 subclasses as an alternative or complementary tool for allergy diagnosis either determination of predictive values have not been consistent [13].
Some allergenic extracts may contain carbohydrate molecules associated with proteins with a wide-ranging immune response profile, including IgE recognition [14][15][16]. The isolation and purification of these glycoconjugates have been frequently conducted using plant lectins as a ligand, which includes Concanavalin A (ConA) [17][18][19], and IgE reactivity to these components has been investigated in some allergen sources [20,21]. Nevertheless, only few studies have been conducted aiming to investigate IgE reactivity to glycosylated components derived from HDM. The first D. farinae glycoconjugates were isolated by ConA-Sepharose affinity chromatography [22], and later, this technique was also used to isolate a high molecular weight antigen from D. pteronyssinus [23]. However, none of these studies demonstrated an IgE/IgG-binding profile against HDM crude extract as well as ConA-related fractions neither inhibition assays. Thus, we sought to investigate allergenicity (IgE) and antigenicity (IgG1 and IgG4) of ConA-bound and ConA-unbound D. pteronyssinus crude extracts using sera of mite-allergic patients and healthy subjects as well as inhibition capacity of antibody binding.

Concanavalin A-Sepharose Affinity Chromatography.
ConA-Sepharose 4B matrix (5 ml; GE Healthcare, Uppsala, Sweden) was packed and equilibrated according to manufacturer instructions. To prevent ConA subunit coelution, matrix was treated with 25% glutaraldehyde as described elsewhere [27]. Dpt extract (5 mg protein) was loaded onto the column and washed with binding buffer containing 1 mM MgCl 2 , 1 mM MnCl 2 , and 1 mM CaCl 2 in PBS while absorbance was monitored by UV at 280 nm. Each 2 mlfraction was collected at 4 ∘ C and pooled, labeling ConAunbound fraction. Once the fractions have reached background levels, 0.1 M methyl -D-mannopyranoside (Ferro Pfanstiehl Laboratories, Inc., Waukegan, USA) was added and ConA-bound fractions were eluted and pooled. Protein and carbohydrate contents were also determined in the pooled fractions.

Subjects and Skin Prick Test (SPT).
A total of 187 subjects were recruited at the Clinics Hospital, Federal University of Uberlândia, Brazil, and submitted to SPT with Dpt, Df, and Bt extracts at 2 mg/ml in PBS added with 0.4% phenol and 50% glycerol (FDA Allergenic, Rio de Janeiro, Brazil). Histamine (10 mg/ml) and buffered glycerine-saline solution were used as positive and negative controls, respectively (FDA Allergenic). Wheal sizes 3 mm larger than the negative control were considered positive. Blood samples were obtained from all subjects and sera were stored at -20 ∘ C until serologic evaluations.
Patients with respiratory allergy (n=95; 35 men and 60 women; age: 25 ± 16 years) were grouped based on their history of rhinitis and asthma, physical examination, and SPT positivity to D. pteronyssinus (DP+ group). Healthy subjects (n=92; 22 men and 70 women; age: 28 ± 22 years) were included in the nonallergic (NA) group, based on the negative SPT results to airborne allergens and absence of clinical To evaluate cross-reactivity among Dpt extract and ConA-unbound and ConA-bound Dpt fractions as well as crude Df and Bt extracts, inhibition ELISA was conducted as previously described [24]. Dpt extract was loaded on the solid phase (1 g/well) and six pooled sera (diluted 1:5) selected from patients of the group DP+ were preadsorbed 18 h at 4 ∘ C with 10-fold-diluted Dpt extract and ConA-unbound and ConA-bound fractions as well as crude Df and Bt extracts as inhibitors. Tetanus toxoid (TT) was used as unrelated antigen. Residual uninhibited antibody reactivity to Dpt extract was measured by ELISA as described above and results were defined as percentage of reactivity to the respective extracts in the absence of inhibitors (inhibition %).

SDS-PAGE and Western
Blotting. Polypeptides in Dpt extract and ConA-unbound and ConA-bound fractions were separated on sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) under denaturing conditions and electrophoretic profile was visualized by silver staining. To identify IgE-, IgG1-, and IgG4-reactive polypeptide fractions in Dpt extract and ConA-unbound and ConAbound fractions, 10 sera from the group DP+ and 10 sera from the group NA were selected for Western Blotting and performed as described elsewhere [24]. Briefly, Dpt extract and its fractions previously separated on 12% SDS-PAGE were electrotransferred onto a nitrocellulose membrane (Bio-Rad Laboratories, Inc., Hercules, USA) followed by blocking with 5% skimmed milk (SM) in 0.05% Tween 20 (PBS-T   Table 1. ConA-bound fraction presented the lowest amount of total protein and carbohydrate (500 g/ml and 200 g/ml, respectively) and the highest protein/carbohydrate ratio (2.5), indicating that carbohydrate levels in relation to protein levels are lower in this fraction, comparing to ConA-unbound either Dpt extract. Although the difference between relative amounts of Der p 2 (6.1%) and Der p 1 (5.9%) in the Dpt extract was subtle, both ConA-unbound and ConA-bound fractions show higher levels of Der p 1 (10.3 % and 5.0 %, respectively) than Der p 2 (6.7% and 0.8%, respectively). As expected, only negligible levels of Der f 1 and Blo t 5 were found.

IgE, IgG1
, and IgG4 Antibodies to Dpt Extract and ConA-Unbound and ConA-Bound Fractions. Specific IgE, IgG1, and IgG4 levels to Dpt, ConA-unbound, and ConA-bound fractions were determined in sera from both allergic patients (DP+ group) and nonallergic subjects (NA group) by ELISA ( Figure 2). Specific IgE levels to Dpt extract and its fractions were significantly higher in sera from the group DP+ when compared to the group NA (p < 0.001). Allergic patients were included in the group DP+ based on their IgE positivity to Dpt extract; thus, all of 95 sera within this group reached specific IgE levels to the Dpt extract above the positivity threshold value (EI = 1.2) while only 61 and 40 serum samples had positive IgE to ConA-unbound and ConAbound fractions, respectively (p < 0.0001). Noticeably, no serum sample had positive IgE within the group NA. In the group DP+, mean IgE levels to Dpt extract were higher than those to ConA-unbound or ConA-bound fractions (p < 0.001), although no significant difference was found between both fractions (Figure 2(a)).
Patients from the group DP+ had higher IgG1 levels to Dpt extract, when compared to ConA-unbound and ConAbound fractions, with no significant difference between the fractions. A similar profile of these differences was observed in the group NA, although with a distinct statistical pattern (Figure 2(b)). Likewise, number of sera with positive IgG1 levels was higher in the Dpt comparing to both ConA-bound (p < 0.05) and ConA-unbound (p < 0.001) fractions in the group DP+ as well as in the group NA. When comparing both fractions, no difference was found between them in the group DP+. However, in the group NA, number of sera positive to ConA-bound fraction was higher than to ConA-unbound fraction (63 vs. 45, respectively; p < 0.05).
In the group DP+, IgG4 mean levels to Dpt extract were higher than to ConA-unbound fraction (p < 0.01), but not to ConA-bound fraction. IgG4 positivity to ConA-unbound fraction (67 positive sera) was also lower than to ConAbound fraction (79 positive sera) (p < 0.05). In the NA group, IgG4 levels to Dpt extract and ConA-bound fraction were higher than to ConA-unbound fraction (p < 0.001) and the same was observed for positivity outcomes. When both groups were compared, IgG4 mean levels to Dpt extract and ConA-unbound fraction were higher in the group DP+ than in the group NA (p< 0.001), but no significant difference was found in the IgG4 levels to ConA-bound fraction between DP+ and NA groups (Figure 2(c)).
IgE/IgG1, IgE/IgG4, and IgG1/IgG4 ratios to Dpt extract and ConA-unbound and ConA-bound fractions were calculated in both DP+ and NA groups. In general, ConAunbound and ConA-bound ratios were similar within these groups and only differences among Dpt extract and its fractions were seen.
In the group DP+, there was no difference between IgE/IgG1 ratios to ConA-unbound/ConA-bound fractions and the Dpt extract. However, IgE/IgG1 ratio to ConA-bound fraction was slightly higher than Dpt extract in the NA group (p < 0.01; Figure 4(a)). Group DP+ presented the higher IgE/IgG1 ratios compared to the group NA for all extracts (Figure 4(a)).
IgE/IgG4 ratio to Dpt extract was higher than ConAbound fraction in the group DP+ (Figure 4(b)). When the distinct groups were compared, IgE/IgG4 ratio to both Dpt extract and ConA-unbound fraction were higher in the group DP+ than NA (Figure 4(b)).
A higher IgG1/IgG4 ratio to both Dpt and ConAunbound fraction was found in the group NA when compared to DP+ group (Figure 4(c)). Moreover, ratios were also significantly lower to ConA-bound fraction than Dpt extract in the NA group (1.0 and 1.6, respectively; Figure 4(c)).

IgE-, IgG1-, and IgG4-Binding to Polypeptides of Dpt Extract and ConA-Unbound and ConA-Bound Fractions.
Polypeptide profiles of Dpt extract and ConA-unbound and ConA-bound fractions were visualized in 12% SDS-PAGE ( Figure 6(a), lanes 1, 2, and 3, respectively). A broad range of bands were detected in the Dpt extract, although some of them were found thicker (14 kDa; 25 to 29 kDa) and others were found sharp, such as high molecular weight bands (66 kDa, 72 kDa, and 99 kDa). Some polypeptides from the Dpt extract were also found in the ConA-unbound fraction, particularly those above 66 kDa and below 20 kDa. Most of those high molecular weight bands are also found in the ConA-bound fraction, although a polypeptide profile complementary among both fractions has been noticed as prominent, particularly 32 kDa, 35 kDa, 45 kDa, 72 kDa, 80 kDa, 95 kDa, and 105 kDa. IgE-, IgG1-, and IgG4-binding activity of Dpt, ConAunbound, and ConA-bound fractions was determined by Western Blotting in 10 serum samples from each DP+ and NA groups. Representative serum samples in each group and each extract for IgE (Figure 6(b)), IgG1 (Figure 6(c)), and IgG4 (Figure 6(d)) are demonstrated. Frequency of band recognition is also represented for IgE- (Figure 7(a)), IgG1- (Figure 7(b)), and IgG4- (Figure 7(c)) binding activity. Immunodominant bands (≥ 50%) were recognized by IgE, including the 14 kDa and 22-25 kDa bands in both Dpt and ConA-unbound fraction, 80 kDa in Dpt and ConAbound fraction, 99 kDa in the ConA-unbound fraction, and 116 kDa in all extracts (Figure 7(a)). As expected, no polypeptide bands were recognized by IgE in the group NA. Interestingly, IgG1 mostly recognized bands above 42 kDa, with immunodominance for 57 kDa and 116 kDa bands in the Dpt, 80 kDa in the Dpt and ConA-unbound fraction, and 78 kDa, and 95-99 kDa for all extracts. Polypeptides with 35 and 50 kDa were weakly recognized by IgG1 (Figure 7(b)).
Band labeling by IgG4 was mostly faint in all extracts ( Figure 6(d)), although frequently reaction was seen in sera from the group NA to the 22-25 kDa and 88 kDa bands in the Dpt extract (Figure 7(c)). Along with the 50 kDa band, the 88 kDa was predominantly recognized in all extracts. Only the 50 kDa and 88 kDa bands were immunodominant in the ConA-unbound fraction and same was seen in the ConAbound fraction for 35 kDa, 50 kDa, 88 kDa, and 95-99 kDa

Discussion
Glycoproteins often constitute immunogenic (innate immunity cells and T-cells) epitopes or IgE-binding sites in allergens [28,29]. Instrumentally, ConA (from Canavalia ensiformis) lectin has been a valuable molecule for glycoconjugate purification as it commonly binds to D-mannose and D-glucose or its stereochemically related residues present in these molecules [30]. In our study, ConA-unbound and ConA-bound fractions were obtained from Dpt crude extract by ConA affinity chromatography and carbohydrate content found in both fractions revealed the presence of glycoconjugates even in the ConA-unbound fraction, with no affinity to ConA.
Up to date, at least 20 groups of D. pteronyssinus allergens have been characterized with different structural and biochemical properties, but IgE-binding capacity is the most prominent feature to distinguish among major, mid-term or minor allergens. Group 1 (Der p 1; Der f 1) and group 2 (Der p 2; Der f 2) major allergens are the most important sensitizing agents of Dermatophagoides species since these allergens are strongly recognized by IgE antibodies in allergic patients [8]. Although Der p 1 is a glycosylated molecule, it was mostly found in the ConA-unbound fraction (10.3% of the total protein content), suggesting that carbohydrate moieties present in the native form of this allergen might have poor affinity to ConA. Controversially, mannose was reported as a mediator on Der p 1 uptake by dendritic cells [31], but Al-Ghouleh and colleagues [32] found a poor reactivity to native Der p 1 using anti-mannose antibody whereas a stronger signal was seen using anti-1,3-Fucose. We did not perform qualitative assays in order to characterize this allergen in our samples, but it is possible that ConA is not appropriate for Der p 1 isolation, based on our quantitative findings. Despite that, group 3 and nonglycosylated allergens are described with same molecular weight as Der p 1 [33,34]; thus, it is likely that the stack of protein observed within this range on SDS-PAGE and Western Blotting for both Dpt extract and ConAunbound fraction has been composed by, at least, those three allergens. Furthermore, as expected, only background levels of Der p 2 were found in the ConA-bound fraction (0.8% of the total protein content), suggesting that this allergen might not be associated with ConA due to absence of glycosylation sites. In addition to that, another glycoprotein has been identified and characterized as a major allergen [35]. It is termed Der p 23, and it was referred as a 14 kDa-protein that is difficult to extract from the fecal pellets and, consequently, not easy to use in its native form for laboratory approaches.
Additional studies, such as direct binding with specific antibodies and mass spectrometry are worthy of further considerations to provide a better characterization of these proteins.
The presence of carbohydrate moieties in HDM allergens has been strongly reported but the role of these molecules on IgE recognition has remained controversial. In our previous studies, fractionation of components from Blomia tropicalis crude extract using ConA affinity chromatography revealed that IgE response to the ConA-bound Bt fraction is similar to the crude Bt extract, indicating that this fraction is important for allergenicity [36]. On the other hand, a given structural approach has revealed a potential N-glycosylation site for the mature group 1 allergen, but this glycan has not affected Der p 1 and Der f 1 allergenicities on recombinant form since the position in which this sugar was found has not corresponded to a IgE-binding site [37].
Not only IgE but IgG antibodies are also important for allergy studies. Boost of IgG levels, particularly IgG4, during or after immunotherapy suggests a protective status against clinical symptoms, including a shift of the Th2 profile towards Th1, upregulation of immunosuppressive cytokines, and downregulation of IgE [12]. Moreover, we previously [38] revealed that D. pteronyssinus-specific IgG antibodies in allergic patients are able to inhibit reaction mediated by specific IgE. Based on these findings, we have evaluated IgE, IgG1, and IgG4 responses to Dpt extract and ConA-related fractions aiming to investigate if glycosylation could be important on IgE/IgG1 and IgE/IgG4 ratios, which are considered valuable 14   parameters for immunotherapy. We revealed that allergic patients can exhibit similar IgG4 levels to both Dpt extract and glycosylated fraction, indicating a potential use of the ConA-bound fraction for immunosuppressive therapy. We also demonstrated that IgG1 antibodies are not relevant for the clinical status of allergy to D. pteronyssinus since no difference in the IgG1 levels has been found between allergic patients and healthy subjects. Furthermore, positive correlations found between IgE levels to the Dpt extract and ConA-related fractions among allergic patients reveal that both fractions can generate IgE response.
Due to differences in the molecular structure of its allergens, cross-reactivity between Dermatophagoides and B. tropicalis is considerably low [39]. However, the majority of allergic patients are cosensitized to D. pteronyssinus and D. farinae, mostly due to high cross-reactivity, 80% and 88%, observed in group 1 and group 2 allergens, respectively [40,41]. Although we have not clearly identified those allergens in our samples, we have found a high IgEbinding inhibition between these Dermatophagoides species, by ELISA. In addition, there was a total cosensitization to D. pteronyssinus and D. farinae, determined by Skin Prick Test, as we previously noticed in our prevalence studies conducted in Brazil [42,43], showing that Pyroglyphidae family could be considered an important risk factor for sensitization of our patients.
In the present study, IgE antibodies have recognized a reasonably high amount of polypeptide bands in Dpt extract and ConA-related fractions, indicating the presence of allergenic molecules in all of them, although ConA-bound fraction contains a reasonably small number of allergenic bands. Interestingly, we have shown that IgE antibodies recognize a broader range of polypeptide bands when compared to the other antibody classes, but IgG1 recognition has been noticed preferably by high molecular weight bands while only few polypeptide bands have had more than 50% of recognition frequency by IgG4. This might be a useful data for further studies involving IgE/IgG binary in the allergic response. Despite this, further identification of allergens, such as Der p 1, Der p 2 and Der p 23 in association to IgE and IgG levels, mainly in ConA-related fractions, should be important to reveal components that cause sensitization in these patients.
Taking together, our data indicate that fractionation of D. pteronyssinus crude extract into mannose/glucose-enriched and -depleted fractions implicates in distinct IgE and IgG recognition patterns, mainly revealing a predominance of IgG antibodies to mannose-enriched fraction, which indicates a protective potential of this fraction. Furthermore, as IgE is described to have poor affinity to carbohydrate moieties [44], it was not expected that ConA-bound fraction elicits higher IgE levels than ConA-unbound one. However, ability of ConA-bound fraction to elicit IgG levels is an important finding as it may constitute a source for immunotherapy by acting on upregulation of immunomodulatory components to reduce the allergic symptoms.

Conclusions
The isolation of native components from D. pteronyssinus using ConA affinity chromatography should be a potential tool for immunochemical analysis of allergenic and antigenic molecules. These studies may be useful to determine new perspectives for immunotherapy using natural allergens from HDM.

Data Availability
All data that support our results are available upon request.

Conflicts of Interest
The authors declare that there are no conflicts of interest regarding the publication of this paper.