ABSTRACT
Aim: The cluster analysis for the evaluation of the results in Multiplex ISAC test in atopic dermatitis patients. Method: The complete dermatological and allergological examination including the examination of the sensitisation to molecular components with Multiplex ISAC testing was performed. The cluster analysis of molecular components (silhouette value) was processed. Results and conclusion: Altogether 100 atopic dermatitis patients were examined - 48 men, 52 women, the average age 40.9 years. We found 10 clusters with different numbers of molecular components. Fundamental position have the components Phl p 1 (Timothy), Bet v 1 (Birch), Alt a 1 (Alternaria) followed by molecular components of NPC2 family, cystein proteasa, tropomyosin, uteroglobin, lipocalin and PR-10 protein. Our results correspond to the association of molecular components into protein families according to their biochemical structure.
Introduction
Diagnostic work – up for IgE mediated allergic reactions starts with the history, followed by sensitisation tests (skin, IgE and basophil tests) and optional challenge tests (Top – down approach). Molecular allergens for IgE testing provide additional infomation, particularly in polysensitized patients and with allergens of low abundance, low stability or associated risks. IgE reactivity to members of the same allergen family reflect the degree of protein homology and IgE cross – reactivity. If it is high, the relevance needs to be sorted out clinically. In case it is low, selected IgE testing of other family members can provide additional information. Proper interpretation should complete diagnostic testing. Positive sensitisations to allergen extracts or molecules are only clinically relevant in case of corresponding symptoms (Matricardi et al., Citation2016). Testing with molecular components is a part of the development of novel allergy vaccines and identification of non-allergenic isoforms (Matricardi et al., Citation2016). The introduction of allergen molecules has had a major effect on analytic specificity and allergy diagnosis. They are used in both singleplex ImmunoCAP and multiplex ImmunoCAP ISAC assays. The major advantage of ISAC is the comprehensive IgE pattern obtained with a minute amount of serum (van Hage et al., Citation2017). ImmunoCAP ISAC (Thermo Fisher), based on 112 different molecular components (both extracted and recombinant), is the most studied and most frequently used molecular diagnostic tool based on a microarray (Melioli et al., Citation2011). The usefulness of ImmunoCAP ISAC has been validated in a wide spectrum of allergic diseases like asthma, allergic rhinoconjunctivitis, atopic dermatitis, eosinophilic esophagitis, food allergy and anaphylaxis. ISAC provides a broad picture of a patient's sensitization profile from a single test, and provides information on specific and cross-reactive sensitizations that facilitate diagnosis, risk assessment, and disease management. Furthermore, it can reveal unexpected sensitizations which may explain anaphylaxis previously categorized as idiopathic and also display for the moment clinically non-relevant sensitizations. ISAC can facilitate a better selection of relevant allergens for immunotherapy compared with extract testing. Microarray technique can visualize the allergic march and molecular spreading in the preclinical stages of allergic diseases, and may indicate that the likelihood of developing symptomatic allergy is associated with specific profiles of sensitization to allergen components. ISAC is shown to be a useful tool in routine allergy diagnostics due to its ability to improve risk assessment, to better select relevant allergens for immunotherapy as well as detecting unknown sensitization. Multiplex component testing is especially suitable for patients with complex symptomatology (Melioli et al., Citation2011).
Proteins that are described worldwide as allergens can be classified roughly into 30–40 protein families. Based on their molecular, biologic and biochemical properties, allergenic proteins are able to initiate both innate and adaptive immune responses during the sensitisation process that ulimately result in the production of allergen – specific IgE. The most important families and selected allergenic member proteins are shown in .
Table 1. The most important allergen superfamilies and families with their corresponding sources (Matricardi et al., Citation2016).
The list of the molecular components in ImmunoCAP ISAC (Phadia, Thermo Fisher Scientific, Uppsala, Sweden) sorted by protein group is recorded in . We distinguishe these superfamilies: Prolamin (family Cereal prolamins, Bifunctional inhibitors, 2S albumin, Non – specific lipid – transfer proteins), EF hand (family Polcalcins, Parvalbumins), Profilin like (family profilin), Tropomyosin like (family Tropomyosin), Cupin (family Vicilins and Legumins), Bet v 1 – like (family Bet v 1), Calycin (family Lipocalins) and Double – psi beta – barrel (family DPBB and Pollen allergy). The families of highly cross-reactive molecules taken into consideration are as follows: profilins, PR-10-like molecules, nsLTP, serum albumins, tropomyosins, polcalcins, lipocalins, and parvalbumins (Matricardi et al., Citation2016; Nicolaou et al., Citation2010; Vieths et al., Citation2002).
Table 2. The list of the molecular components sorted by protein group. (ImmunoCAP ISAC - Phadia, Thermo Fisher Scientific, Uppsala, Sweden).
The aim of our study is to evaluate the results of examination with Multiplex ImmunoCAP ISAC in atopic dermatitis according to the Cluster analysis. This statistic method is commonly used to group similar samples across a diverse range of applications. Typically, the goal of clustering is to form groups of samples that are more similar to each other than to samples in other groups (Frades & Matthiesen, Citation2010). Silhouette width is a widely used index for assessing the fit of individual objects in the classification, as well as the quality of clusters and the entire classification (Frades & Matthiesen, Citation2010). In our study, we evaluate if these results correspond to the association of molecular components into protein families according to their biochemical structure. Our results can demonstrate the importance of molecular components in atopic dermatitis patients.
Patients and methods
In the period 2018–2019, 100 patients suffering from atopic dermatitis were examined. All these patients were examined in the Department of Dermatology, Faculty Hospital Hradec Králové, Charles University, Czech republic. The diagnosis of atopic dermatitis was made with the Hanifin-Rajka criteria (Hanifin & Rajka, Citation1980). Exclusion criteria were long term therapy with cyclosporin or systemic corticoids, pregnancy, breastfeeding. Patients with atopic dermatitis having other systemic diseases were excluded from the study as well. Complete dermatological and allergological examination was performed in patients included in the study. This study was approved by Ethics commitee of Faculty Hospital Hradec Králové, Charles University of Prague, Czech Republic.
Examination of specific IgE to molecular components
The serum level of the sIgE was measured by the component-resolved diagnosis microarray-based sIgE detection assay ImmunoCAP ISAC (Phadia, Thermo Fisher Scientific, Uppsala, Sweden). ImmunoCAP ISAC is a solid-phase multiple immunoassay which enables to determine 112 different components from 51 allergen sources (Jakob et al., Citation2015). The allergens are applied in triplicates to ensure the test reproducibility. The specific IgE values are presented in arbitrary units called ISAC Standardized Units (measuring range of 0.3 - 100 ISU-E). The level of specific IgE higher than 0.3 ISU-E was assessed as positive. The level of molecular components was evaluated in ISU –E: < 0.3 - negative, 0.3- >0.9 low positivity, 0.9 > 15 moderate positivity, ≥15 ISU –E very high positivity (Choi et al., Citation2014; Jakob et al., Citation2015).
Statistic
We analysed the data to determine 10 clusters with molecular components according to the positivity to molecular components and to their level of specific IgE. The level of molecular components in ISU –E was evaluated: < 0.3 - negative, 0.3-0.9 ISU –E low positivity, 0.9-15 ISU –E moderate positivity, above 15 ISU –E very high positivity. This statistic method for the evaluation of results was used: Clustering by Medoid Partitioning, Method: Kaufman-Rousseeuw, Objective Function: Silhouette, Distance Type: Euclidean, Scale Type: Standard Deviation. Molecular components are ordinal variables (Kaufman & Rousseeuw, Citation1990; Rousseeuw, Citation1987).
Interpreting silhouettes
A silhouette value is constructed for each object. The value can range from minus one to one. It measures how well an object has been classified by comparing its dissimilarity within its cluster to its dissimilarity with its nearest neighbour. When s is close to one, the object is well classified. Its dissimilarity with other objects in its cluster is much less than its dissimilarity with objects in the nearest cluster. When s is near zero, the object was just between clusters A and B. It was arbitrarily assigned to A. When s is close to negative one, the object is poorly classified. Its dissimilarity with other objects in its cluster is much greater than its dissimilarity with objects in the nearest cluster. Why isn't it in the neighbouring cluster? Hence, the silhouette value summarizes how appropriate each object’s cluster is.
Determining the number of clusters
One useful summary statistic is the average value of s across all objects. This summarizes how well the current configuration fits the data. An easy way to select the appropriate number of clusters is to choose that number of clusters which maximizes the average silhouette. We denote the maximum average silhouette across all values of k as SC. Kaufman and Rousseeuw (Kaufman & Rousseeuw, Citation1990) present the following table to aid in the interpretation of SC. We evaluated our data according to this table, it is shown in .
Table 3. Cluster analysis - Proposed Interpretation (Kaufman & Rousseeuw, Citation1990).
Results
Patients
Altogether 112 different components from 51 allergen sources were examined in 100 atopic dermatitis patients included in the study (48 men and 52 women with the average age 40.9 years and with the average SCORAD 39, s.d.13.1 points). The mild form of AD was recorded in 14 patients (14%), moderate form in 58 patients (58%), severe form in 28 patients (28%). In the included patients, 55 patients (55%) suffer from asthma bronchiale and 78 patients (78%) suffer from rhinitis. The positive results in sensitisation to molecular components was recorded in 93 patients (93%). The characteristics of patients is shown in .
Table 4. The characteristic of patients with atopic dermatitis
We found 10 clusters with different numbers of molecular components.
The results of ISAC Muliplex testing evaluated by Clustering by Medoid Partitioning Report are shown in . We show the average of Silhouette Value: Cluster 1: Phlp 1 (Silhouette Value 0. 000), Cluster 2: Phl p 4 (Silhouette Value 0. 000), Cluster 3: Phl p 2, 5, 6, Cyn d 1 (Silhouette Value 0. 1278), Cluster 4: Bet v 1 (Silhouette Value 0. 000), Cluster 5: Alt a 1 (Silhouette Value 0. 000), Cluster 6: Der f 2, Der p 2, Der p 1, Der f 1 (Silhouette Value 0. 4167), Cluster 7: Pen m 2, Mus m 1, Lep d 2, Can f 5, Phl p 11, Pen m 1, Bla g 7, Der p 10, Ole e 1, Blot 5, Ani s 3 (Silhouette Value - 0. 0893). Cluster 8: Can f 1, Fel d 1, Fel d 4, Equ c 1 (Silhouette Value - 0. 0357), Cluster 9: Pru p 1, Cor a 1.010, Mal d 1, Aln g 1, Cor a 1.040 (Silhouette Value 0. 3692). Cluster 10: 82 molecular components in ISAC multiplex testing (Silhouette Value 0. 3743). Groups of 10 Clusters of molecular components with their biochemical name, protein family and allergen source and number of patients with positive results to molecular components are recorded in .
Table 5. Results of ISAC Muliplex testing evaluated by Clustering by Medoid Partitioning Report. There are 10 clusters. Cluster average is shown extra bold.
Table 6. Results of Multiplex ISAC testing - molecular components in 10 clusters with their biochemical name, protein family and allergen source, number of patients with positive results to molecular components. Examination in 100 patients ( = 100%).
Discussion
The sense of our study was to evaluate the results of examination of molecular components with ISAC Multipex testing in atopic dermatitis patients with the use of the Clustering analysis. It needs to be emphasized that in this paper we focus only on sensitization rates and not their clinical relevance without using specific provocation tests.
The clustering problem has been addressed in many contexts and disciplines. Cluster analysis encompasses different methods and algorithms for grouping objects of similar kinds into respective categories. Clustering techniques are widely used in the analysis of large datasets to group together samples with similar properties. There are many algorithms for performing clustering, and the results can vary substantially (Rousseeuw, Citation1987). At our study each cluster is represented by a so-called silhouette, which is based on the comparison of its tightness and separation. This silhouette shows which objects he well within their cluster, and which ones are merely somewhere in between clusters. The entire clustering is displayed by combining the silhouettes into a single plot, allowing an appreciation of the relative quality of the clusters and an overview of the data configuration. The average silhouette width provides an evaluation of clustering validity. Silhouette width is a widely used index for assessing the fit of individual objects in the classification, as well as the quality of clusters and the entire classification. Silhouette combines two clustering criteria, compactness and separation, which imply that spherical cluster shapes are preferred over others—a property that can be seen as a disadvantage in the presence of complex, nonspherical clusters, which is common in real situations (Rousseeuw, Citation1987).
We did not find similar studies dealing with this question in adolescents and adults suffering from atopic dermatitis according to the databates in Medline, Pubmed, and Web of Science.
Our results confirme the fundamental and exceptional position of the components such as Phl p 1, Bet v 1, Alt a 1 – these components were recorded as a principal components without any other components in the separated clusters. The intersting cluster is Cluster No. 4 with the single component of Phl p 4. Important role play also the molecular components of NPC2 family (Niemann-Pick disease type C2), cystein proteasa, tropomyosin, uteroglobin, lipocalin and PR-10 proteins.
The single molecular component Phl p 1 was recorded in Cluster 1. Sensitization to Phl p 1 usually precedes sensitization to other grass pollen allergen and is the most prevalent component sensitization in grass pollen -allergic patients (Hatzler et al., Citation2012). Phl p 1 is a beta-expansin, bound to the cell wall and important for pollen tube penetration; it is a major grass pollen allergen with more than 80% homology to group 1 allergens from the Pooideae subfamily (Focke et al., Citation2001; Hatzler et al., Citation2012). Phl p 1 is in most patients the “initiator” molecule. Moreover, even in the few grass pollen - allergic patients who start their sensitization process with other molecules, IgE against Phl p 1 are produced quite soon. Therefore, IgE to Phl p 1 is an essential marker in grass pollen-allergic patients to establish “true sensitization”. The presence of IgE to Phl p 1 confirms that the patient with a positive skin test or IgE assay to grass pollen extract is truly sensitized to grass pollen. The absence of IgE to Phl p 1 does not exclude “true” sensitization to grass pollen, which might be due (in a few cases) to isolated IgE sensitization to other major allergenic proteins (e.g. Phl p 5) but makes it rather unlikely. The molecular component Phl p 4 is recorded in Cluster 2; it is a tryptase-resistant glycoprotein, berberine bridge enzyme, involved in the synthesis of alkaloids. It can be classified as a major allergen. It shows IgE cross-reactivity with other group 4 grass pollen allergens, including with Cyn 4 to some extent. Moreover, cross-reactivity to the major ragweed allergen Amb a 1 and to Oilseed Rape pollen has been demonstrated. Natural Phl p 4 contains cross-reactive carbohydrate determinants (CCD), which may lead to IgE cross-reactivity with a wide range of plants and plant products. This explains why in several epidemiological studies IgE positivity to Phl p 4 scores over 90% of the grass pollen-allergic patients. However, when the recombinant version of the molecules is used for assays, over 50% of the positivity is not confirmed anymore. Phl p 4 may also serve as a marker for sensitization to Bermuda grass pollen due to its similarity with Cyn d 4, a major allergen of Bermuda grass pollen, but this needs to be investigated in relevant patient populations (Matricardi et al., Citation2016). As Bermuda grass pollen is not present in our region, possible cross- reactivity with β-expansins from other grasses could be the explanation for the results with high sensitisation to Cyn d 1 in in our study.
Molecular components such as Phl p 5, Phl p 6, Cyn d 1 and Phl p 2 were found in Cluster 3. Phl p 5 is another major pollen allergen of temperate grasses with low sensitization prevalence, but often with high IgE levels. Phl p 5 is a cytoplasmatic ribonuclease, important in the enzymatic degradation of RNA. It shows broad IgE cross-reactivity with other group 5 allergens from the Pooideae subfamily of temperate grasses. Although IgE to Phl p 5 usually appear later than those to Phl p 1 in the sensitization process, their concentration grows in many patients rapidly and higher and their contribution to patients' symptoms has been demonstrated. Testing IgE to Phl p 5 can be useful as a second-line test and has been shown to be useful for distinguishing between allergy to grass and olive pollen in southern Europe. Phl p 5-specific IgE may have some prognostic value for indicating disease severity or likely progression from allergic rhinitis to asthma, but this needs to be confirmed with well-designed studies. Phl p 6 is another major grass pollen allergen, specific for the Pooideae subfamily. Its function has not yet been described (Matricardi et al., Citation2016). Phl p 2 is rarely the only molecule inducing grass pollen sensitization and the presence of IgE antibodies to this Phl p 2 – observed in around 60–80% of the European grass pollen-allergic patients – just confirms that a positive SPT reaction is the expression of true sensitization to grass. Phl p 6 is highly cross-reacting with Phl p 5 and does not add more diagnostic information, once IgE to Phl p 5 has been tested (Matricardi et al., Citation2016). The only component found in Cluster 4 is Bet v 1. Birch, followed by alder and hazel, represents the most potent cause of tree pollen allergy. In Europe, the prevalence of positive skin prick test to birch pollen allergens ranges from 5% in The Netherlands to 54% in Switzerland, while Scandinavian countries have the highest number of patients with exclusive sensitization to Bet v 1 (Asam et al., Citation2015; Breiteneder et al., Citation1988). Bet v 1 in birch pollen is responsible for most of the IgE binding to the allergen source. For tree pollen allergic patients in Northwestern and Central Europe, Bet v 1 is of decisive clinical importance because there is no “competing” major allergen (Matricardi et al., Citation2016). Pathogenesis-related protein group 10 PR-10 molecules (i.e. Bet v 1 and homologous allergens) are the major allergens in Fagales pollen and are recognized by virtually all allergic patients, thus representing the major cause of clinical allergy (D'Amato et al., Citation1998). PR-10 proteins defend plants against fungi and other microorganisms (Asam et al., Citation2015; Breiteneder et al., Citation1988). Their homologs are also present in a large number of plant-derived foods, and thus frequently cause cross-sensitization and consequently plant-food allergy (oral allergy syndrome, in most cases). In our study, molecular components from PR – 10 protein family (Pru p 1, Cor a 1.010, Mal d 1, Aln g 1, Cor a 1.040, Ara h 8 and Gly m 4) are recorded in Cluster 9.
Exceptional position was recorded for molecular component Alt a 1, which was found as a single component in Cluster 5. Sensitisation to Alternaria alternata spores are considered a well-known biological contaminant and a very common potent aeroallergen source that is found in environmental samples. The major allergen, Alt a 1, has been reported as the main elicitor of airborne allergies in patients affected by a mold allergy and considered a marker of primary sensitization to Alternaria alternata. Moreover, Alternaria alternata sensitization seems to be a triggering factor in the development of poly-sensitization, most likely because of the capability of Alternaria alternata to produce, in addition to Alt a 1, a broad and complex array of cross-reactive allergens that present homologs in several other allergenic sources (Chruszcz et al., Citation2012). According to our previous results, the sensitisation to molecular fungal allergen was recorded altogether in 58% atopic dermatitis patients (Čelakovská et al., Citation2019). The study and understanding of Alternaria alternata allergen information may be the key to explaining why sensitization to Alternaria alternata is a risk factor for asthma and also why the severity of asthma is associated to this mold. Recent research on the identification and characterization of Alternaria alternata allergens has allowed for the consideration of new perspectives in the categorization of allergenic molds, assessment of exposure and diagnosis of fungi-induced allergies (Chruszcz et al., Citation2012; Gabriel et al., Citation2016).
The molecular components such as Der f 2, Der p 2, Der p 1 and Der f 1 were recorded in Cluster 6. Der p 1, Der f 1 are cystein proteasa, Der p 2, Der f 2 are from NPC2 protein family (epidermal secretory proteins). For D. pteronyssinus and D. farinae, the fecal particles are the cardinal form in which the allergens became airborne (de Blay et al., Citation1997; Tovey, Chapman, and Platts-Mills, Citation1981; Tovey, Chapman, Wells, et al., Citation1981). In addition, for Der p 1, there is specific evidence about the effects of reducing exposure from 13 to 0.2 μg/g on symptoms among mite-allergic individuals (Banerjee et al., Citation2015; de Blay et al., Citation1997; Tovey, Chapman, & Platts-Mills, Citation1981; Tovey, Chapman, Wells, et al., Citation1981). The nature of these particles is relevant to both the induction of the IgE response and the subsequent contribution to inflammation of the nose and lungs. For dust mite particles, it is clear that they not only carry a high concentration of several allergens but also are an important source of other substances both nucleic acids and proteins that act as toll receptor ligands or PAMPs. While this has been well recognized for mites, it may be just as important for cockroach allergens (Banerjee et al., Citation2015; de Blay et al., Citation1997; Matricardi et al., Citation2016; Spieksma & Voorhorst, Citation1969; Tovey, Chapman, & Platts-Mills, Citation1981; Tovey, Chapman, Wells, et al., Citation1981).
The components such as Pen m 2, Mus m 1, Lep d 2, Can f 5, Phl p 11, Pen m 1, Bla g 7, Der p 10, Ole e 1, Blo t 5 and Ani s 3 are found together in the Cluster 7. Phl p 11 belongs to the Ole e 1 related proteins and hence exhibits a broad range of cross-reactivities to pollen from different plants as olive, ash, privet, saffron crocus, thistle, plantain, and corn. It is an acidic polypeptide with homology to the tryptase inhibitor of soybean. The shrimp (Penaeus aztecus) major allergen, Pen a 1, is one of the most clinically relevant allergenic tropomyosins (Ahumada et al., Citation2015; Daul et al., Citation1994; Reese et al., Citation2006). Tropomyosins from invertebrates are allergenic for genetically susceptible individuals, and due to their extensive cross-reactivity, are considered panallergens. Most allergenic tropomyosins are major shellfish allergens. Symptoms may be induced by very low amounts of the offending food and sometimes by inhalation. In Europe, sensitization to mite tropomyosin Der p 10 is low and has been considered an effect of cross-reactivity but also a marker for broad sensitization among house dust mite allergic patients. Component-resolved diagnosis of D. pteronyssinus allergens Der p 1, Der p 2, and Der p 10 has been suggested for selecting patients for house dust mite immunotherapy. The prevalence of sensitization to Der f 10 was found around 80% in Japan. In addition, sensitization to Der p 10 was found 55% in Zimbabwe and 34% in Colombia, probably because of perennial exposure to shellfish and helminth infections. Therefore, the clinical impact of non-food allergenic tropomyosins may be greater than previously thought. In fact, it has been suggested that sensitization to tropomyosin from mite cockroach, Ascaris (Ahumada et al., Citation2015), and mosquito could influence the prevalence and severity of asthma in places where coexposition to several sources of tropomyosin occurs. The frequency of IgE sensitization to tropomyosins in shellfish-allergic patients ranges from 50% to 100%. In addition, Pen a 1 binds up to 75% of all shrimp-specific IgE antibodies, which is supported by histamine release experiments (Daul et al., Citation1994; Reese et al., Citation1997; Reese et al., Citation2006). Molecular components such as Can f 1 (Lipocalin), Fel d 1 (Uteroglobin), Fel d 4 (Lipocalin) and Equ c 1 (Lipocalin) were found in Cluster 8. The majority of the mammalian allergens are lipocalins (Hilger, Swiontek, et al., Citation2012). Lipocalins are proteins that are ubiquitous; they are present also in arthropods, plants and bacteria, and have very diverse functions. They are characterized by a common tertiary structure composed of a central β-barrel formed of eight antiparallel β-strands. The internal binding pocket carries small hydrophobic molecules such as retinol, steroids, pheromones, and odorants. Mammalian allergens isolated so far are mostly odorant and pheromone binding lipocalins. Only few natural ligands have been characterized. Lipocalin allergens are present in dander, saliva, and urine. They stick to particles and become easily airborne and transported to public places such as schools or day care centers (Hilger, Kuehn, et al., Citation2012; Hilger, Swiontek, et al., Citation2012; Konradsen et al., Citation2014; Nilsson et al., Citation2012; Saarelainen et al., Citation2008; Zahradnik & Raulf, Citation2014). Lipocalins are small, secreted molecules of 150–250 amino acids. Representatives of this cross-reactive group are Equ c 1, Fel d 4 and Can f 6. However, even between molecules of low general sequence identity such as Fel d 4 and Can f 2 (25% identity), single epitopes may have short stretches of sequence identity and lead to patient-dependent IgE cross-reactivity. (Hilger, Kuehn, et al., Citation2012; Nilsson et al., Citation2012; Saarelainen et al., Citation2008). In Cluster 9, these components were found: Pru p 1 (Peach), Cor a 1.010 (Hazelnut), Mal d 1 (Apple), Aln g 1 (Alder), Cor a 1.040 (), Ara h 8 (Peanut), Gly m 4 (Soy) – all these components belong to the family PR −10 protein. In 1980, pathogenesis-related proteins (PR proteins) were defined as “proteins encoded by the host plant but induced only in pathological or related situations” and subsequently grouped into families. Today, the list of PR proteins comprises 14 families. When the sequence of Bet v 1 was discovered in 1989, the PR-10 family had not been defined yet but it was noted that Bet v 1 was homologous to a PR protein from pea Bet v 1 is constitutively expressed in pollen at rather high concentrations. Hence, the term PR-10 for the Bet v 1 homologous allergens is not entirely correct. These constitutively expressed proteins are referred to as PR-10 like proteins.. Positive specific IgE to Bet v 1 homologs plant-food allergens (i.e. Pru p 1 from peach) demonstrates allergic sensitization, being clinically relevant in case of corresponding symptoms (Antoniw et al., Citation1980; Gepp et al., Citation2014; Geroldinger-Simic et al., Citation2011; Kitzmüller et al., Citation2015; van Loon & van Strien, Citation1999).
In Cluster 10, 82 molecular components from different pollen and food allergen sources were recorded. Phl p 7 and Phl p 12 are minor allergens, representing panallergens from the plant world. Phl p 7, polcalcin, is a calcium-binding protein present in many different pollens, hence representing a broad cross-reacting allergen: birch, alder, juniper, ragweed, mugwort, olive, goosefoot, etc. Phl p 7 sensitization can be used as a marker for a more general pollen sensitization. Phl p 12 is a member of the profilin family, an actin-binding protein that is present throughout the whole plant world. As profilins are ubiquitous in plant cells, profilin sensitization gives rise to a long range of cross-reacting plants and plant products as birch, soybean, corn, latex, and plant foods. Phl p 12 is the highly cross-reacting profilin of Phleum pratense. As a heat-labile, relatively weak allergenic molecule, IgE sensitization to profilin comes later in the molecular spreading process, reaches only moderate levels of IgE antibodies and only in a minority of patients. Hence, IgE to Phl p 12 mark in general patients with a higher atopic background and/or longer disease duration. Patients with a positive skin test/IgE to grass pollen extract but no IgE to Phl p 1 and Phl p 5 must be tested for IgE to Phl p 12 as these antibodies – that can be induced by other pollens containing profilin – are the first cause of “false” positivity to assays based on grass pollen extract. In the presence of IgE to Phl p 12, patients should be asked about oral allergy syndrome triggered by the ingestion of fruit and vegetables containing profilin, with a focus on apple, celery, and birch. Phl p 7, is the highly cross-reacting polcalcin of Phleum pratense. This is a heat-stabile, relatively potent allergen that can induce quite high IgE antibody levels. An IgE response to Phl p 7 is observed only infrequently among grass pollen-allergic patients and usually many years after the disease onset. IgE to Phl p 7 marks a relatively distinct category of grass pollen-allergic patients, with more severe symptoms, a higher prevalence of asthma, and a higher frequency of additional allergic comorbidities. Moreover, many other pollens and allergenic sources contain polcalcin so that the origin sensitization to polcalcin in a grass pollen-allergic patient must be carefully searched. These allergenic sources could be indeed responsible of a more severe disease (Bousquet et al., Citation2007; Davies, Citation2014; Devanaboyina et al., Citation2014; Popescu, Citation2014).
In our previous studies we evaluated the occurrence of sensitisation to food and inhallant allergens in patients suffering from AD. The challenge test was performed according to the results of examinations sIgE and APT atopy patch tests with suspected foods. The diagnostic work-up should comprise not only the laboratory methods, but also the diagnostic hypoallergenic diet and the challenge test in patients with suspected food allergy (Čelakovská, Ettlerová, Ettler, and Bukač, Citation2015; Čelakovská, Ettlerová, Ettler, Vaněčková, et al., Citation2015). Our previous study demonstrates that there is a significant association between the severity of AD and the incidence of perennial rhinitis, asthma bronchiale, and the worsening of atopic dermatitis in relation to food (Čelakovská & Bukač, Citation2011).
According to the literature, quantitative competition immunoassays with appropriate combinations of antibodies give consistent dose-response patterns which may be used to identify and estimate amounts of cross-reacting compounds. Previously reported methods of analyzing cross-reaction patterns include multiple regression, principal components analysis and minimum estimates of variance (MEV). Four other techniques which are preferable in theory have been surveyed: discriminant analysis (DA), maximum likelihood estimates (MLE), classification and regression trees (CART), and computational neural networks (NN). MLE and simple back-propagation neural networks can estimate the concentration, as well as the identity, of individual compounds. These four methods worked well with unfitted, unscaled data from monoclonal assays of triazines, phenylureas and avermectins. Immunoassays must be properly designed to provide adequate data for pattern recognition. Cross-reactivity pattern analysis will make multi-analyte, multi-antibody immunoassays feasible for many applications in toxicology and hazard assessment (Karu et al., Citation1994).
According to the literature, little is known about pollen-food allergy syndrome (PFS) in China. Ma Shikun et al. investigated the clinical characteristics, as well as sensitization patterns, of PFS in China. Clinical parameters and serum Immunoglobulin E (IgE) responses to prevalent pollens, plant foods and corresponding allergen components were evaluated. The top three most common pollen-associated allergenic foods were peach, apple and pear. Peach was the most common allergenic food in PFS patients. Patients with PFS in China showed an LTP-dominant sensitization profile and usually presented systemic reactions upon consumption of the allergenic foods (Ma et al., Citation2018).
Food allergies, including kiwi fruit allergy, have been the subject of extensive research in the last few years. The aim of Gavrovic-Jankulovic ´s study was to examine a possible relationship between the developmental stage of kiwi fruit and its allergenic potency. The protein and allergen patterns of kiwi fruit extracts in September, October, November and December fruit in the period from 2000–2002 were analysed. Two major allergens of kiwi fruit, Act c 1 and Act c 2, mainly accounted for the highest allergenic potential of November kiwi extract in vivo and in vitro. Not only the content of major allergens, but also the ratio of different proteins and even isoforms of the same allergen (Act c 2) change with fruit ripening. These findings should be taken into account during preparation of extracts for allergy diagnosis (Gavrovic-Jankulovic et al., Citation2005).
The prevalence of fish allergies has become a serious health problem and has increased alarmingly over the past few years. Liu, Chu-Yi ´s study attempted to identify and purify the major allergen implicated in the allergic response to largemouth bass (Micropterus salmoides), a freshwater fish widely consumed in China. According to their results, nucleoside diphosphate kinase B was identified as a novel fish allergen in largemouth bass. This finding is important for allergy diagnoses and the treatment of freshwater fish–allergic disorders (Liu et al., Citation2014).
The quality of an applied protein extract is important in both serological and in vivo diagnosis of allergy, and for allergen detection methods. In Dooper ´s study the effects of the extraction procedure and hazelnut source on antibody binding to hazelnut (Corylus avellana) proteins were investigated. This study indicates that for the production of hazelnut protein extracts, the use of fresh hazelnuts is important. A quick, vigorous extraction in a tris buffer might contribute positively, at least for extraction of Cor a 9 (Dooper et al., Citation2008).
Conclusion
The results of ISAC multiplex examination in 100 atopic dermatitis patients were processed with cluster analysis, we found 10 clusters with different numbers of molecular components. Our results correspond to the association of molecular components into protein families according to their biochemical structure. Fundamental position have the components Phl p1, Bet v 1, Alt a 1 followed by molecular components of NPC2 family, cystein proteasa, tropomyosin, uteroglobin, lipocalin and PR-10 protein. These molecular components play the most important role in atopic dermatitis patients.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Additional information
Funding
References
- Ahumada, V., García, E., Dennis, R., Rojas, M. X., Rondón, M. A., Pérez, A., Peñaranda, A., Barragán, A. M., Jimenez, S., Kennedy, M. W., & Caraballo, L. (2015). Ige responses to Ascaris and mite tropomyosins are risk factors for asthma. Clinical & Experimental Allergy, 45(7), 1189–1200. https://doi.org/10.1111/cea.12513
- Antoniw, J. F., Ritter, C. E., Pierpoint, W. S., & van Loon, L. C. (1980). Comparison of three pathogenesis-related proteins from plants of two cultivars of tobacco infected with TMV. Journal of General Virology, 47(1), 79–87. https://doi.org/10.1099/0022-1317-47-1-79
- Asam, C., Hofer, H., Wolf, M., Aglas, L., & Wallner, M. (2015). Tree pollen allergens-an update from a molecular perspective. Allergy, 70(10), 1201–1211. https://doi.org/10.1111/all.12696
- Banerjee, S., Resch, Y., Chen, K.-W., Swoboda, I., Focke-Tejkl, M., Blatt, K., Novak, N., Wickman, M., van Hage, M., Ferrara, R., Mari, A., Purohit, A., Pauli, G., Sibanda, E. N., Ndlovu, P., Thomas, W. R., Krzyzanek, V., Tacke, S., Malkus, U., … Vrtala, S. (2015). Der p 11 is a major allergen for house dust mite-allergic patients suffering from atopic dermatitis. Journal of Investigative Dermatology, 135(1), 102–109. https://doi.org/10.1038/jid.2014.271
- Bousquet, P.-J., Chinn, S., Janson, C., Kogevinas, M., Burney, P., & Jarvis, D. (2007). Geographical variation in the prevalence of positive skin tests to environmental aeroallergens in the European Community Respiratory health Survey I. Allergy, 62(3), 301–309. https://doi.org/10.1111/j.1398-9995.2006.01293.x
- Breiteneder, H., Hassfeld, W., Pettenburger, K., Jarolim, E., Breitenbach, M., Rumpold, H., Kraft, D., & Scheiner, O. (1988). Isolation and characterization of messenger RNA from male inflorescences and pollen of the white birch (Betula verrucosa). International Archives of Allergy and Immunology, 87(1), 19–24. https://doi.org/10.1159/000234643
- Čelakovská, J., & Bukač, J. (2011). Food adverse reactions in patients over 14 years of age suffering from atopic eczema. Food and Agricultural Immunology, 22(2), 105–114. https://doi.org/10.1080/09540105.2010.533751
- Čelakovská, J., Bukač, J., Vaňková, R., Krcmova, I., Krejsek, J., & Andrýs, C. (2019). Sensitisation to molecular allergens of Alternaria alternata, Cladosporium herbarum, Aspergillus fumigatus in atopic dermatitis patients. Food and Agricultural Immunology, 30(1), 1097–1111. https://doi.org/10.1080/09540105.2019.1660624
- Čelakovská, J., Ettlerová, K., Ettler, K., & Bukač, J. (2015). Food allergy, asthma bronchiale, and rhinitis in atopic dermatitis patients with total immunoglobulin E under and above 200 IU/ml. Food and Agricultural Immunology, 26(5), 671–681. https://doi.org/10.1080/09540105.2015.1007447
- Čelakovská, J., Ettlerová, K., Ettler, K., Vaněčková, J., & Bukač, J. (2015). Evaluation of food allergy to wheat, cow milk, egg, soy and peanuts in patients suffering from atopic dermatitis. Food and Agricultural Immunology, 26(1), 26–37. https://doi.org/10.1080/09540105.2013.864603
- Choi, J. S., Roh, J. Y., & Lee, J. R. (2014). Clinical availability of component-resolved diagnosis using microarray technology in atopic dermatitis. Annals of Dermatology, 26(4), 437–446. https://doi.org/10.5021/ad.2014.26.4.437
- Chruszcz, M., Chapman, M. D., Osinski, T., Solberg, R., Demas, M., Porebski, P. J., Majorek, K. A., Pomés, A., & Minor, W. (2012). Alternaria alternata allergen Alt a 1: A unique β barrel protein dimer found exclusively in fungi. Journal of Allergy and Clinical Immunology, 130(1), 241–247. https://doi.org/10.1016/j.jaci.2012.03.047
- D'Amato, G., Spieksma, F. T. M., Liccardi, G., Jäger, S., Russo, M., Kontou-Fili, K., Nikkels, H., Wüthrich, B., & Bonini, S. (1998). Pollen-related allergy in Europe. Allergy, 53(6), 567–578. https://doi.org/10.1111/j.1398-9995.1998.tb03932.x
- Daul, C. B., Slattery, M., Reese, G., & Lehrer, S. B. (1994). Identification of the major brown shrimp (Penaeus aztecus) allergen as the muscle protein tropomyosin. International Archives of Allergy and Immunology, 105(1), 49–55. https://doi.org/10.1159/000236802
- Davies, J. M. (2014). Grass pollen allergens globally: The contribution of subtropical grasses to burden of allergic respiratory diseases. Clinical & Experimental Allergy, 44(6), 790–801. https://doi.org/10.1111/cea.12317
- de Blay, F., Sanchez, J., Hedelin, G., Perez-Infante, A., Vérot, A., Chapman, M., & Pauli, G. (1997). Dust and airborne exposure to allergens derived from cockroach (Blattella germanica) in low-cost public housing in Strasbourg (France). Journal of Allergy and Clinical Immunology, 99(1), 107–112. https://doi.org/10.1016/S0091-6749(97)70307-5
- Devanaboyina, S. C., Cornelius, C., Lupinek, C., Fauland, K., Dall'Antonia, F., Nandy, A., Hagen, S., Flicker, S., Valenta, R., & Keller, W. (2014). High-resolution crystal structure and IgE recognition of the major grass pollen allergen Phl p 3. Allergy, 69(12), 1617–1628. https://doi.org/10.1111/all.12511
- Dooper, M. M. B. W., Plassen, C., Holden, L., Moen, L. H., Namork, E., & Egaas, E. (2008). Antibody binding to hazelnut (Corylus avellana) proteins: The effects of extraction procedure and hazelnut source. Food and Agricultural Immunology, 19(3), 229–240. https://doi.org/10.1080/09540100802243325
- Focke, M., Mahler, V., Ball, T., Sperr, W. R., Majlesi, Y., Valent, P., Kraft, D., & Valenta, R. (2001). Nonanaphylactic synthetic peptides derived from B cell epitopes of the major grass pollen allergen, Phl p 1, for allergy vaccination. The FASEB Journal, 15(11), 2042–2044. https://doi.org/10.1096/fj.01-0016fje
- Frades, I., & Matthiesen, R. (2010). Overview on techniques in cluster analysis. In Bioinformatics methods in clinical research (pp. 81–107). Humana Press.
- Gabriel, M., Postigo, I., Tomaz, C., & Martínez, J. (2016). Alternaria alternata allergens: Markers of exposure, phylogeny and risk of fungi-induced respiratory allergy. Environment International, 89-90, 71–80. https://doi.org/10.1016/j.envint.2016.01.003
- Gavrovic-Jankulovic, M., Polovic, N., Prisic, S., Jankov, R. M., Atanaskovic-Markovic, M., Vuckovic, O., & Velickovic, T. C. (2005). Allergenic potency of kiwi fruit during fruit development. Food and Agricultural Immunology, 16(2), 117–128. https://doi.org/10.1080/09540100500090804
- Gepp, B., Lengger, N., Bublin, M., Hemmer, W., Breiteneder, H., & Radauer, C. (2014). Chimeras of Bet v 1 and Api g 1 reveal heterogeneous IgE responses in patients with birch pollen allergy. Journal of Allergy and Clinical Immunology, 134(1), 188–194. https://doi.org/10.1016/j.jaci.2013.12.1073
- Geroldinger-Simic, M., Zelniker, T., Aberer, W., Ebner, C., Egger, C., Greiderer, A., Prem, N., Lidholm, J., Ballmer-Weber, B. K., Vieths, S., & Bohle, B. (2011). Birch pollen-related food allergy: Clinical aspects and the role of allergen-specific IgE and IgG4 antibodies. Journal of Allergy and Clinical Immunology, 127(3), 616–622. https://doi.org/10.1016/j.jaci.2010.10.027
- Hanifin, J., & Rajka, G. (1980). Diagnostic features of atopic dermatitis. Acta Dermato-Venereologica, 92, 44–47.
- Hatzler, L., Panetta, V., Lau, S., Wagner, P., Bergmann, R. L., Illi, S., Bergmann, K. E., Keil, T., Hofmaier, S., Rohrbach, A., Bauer, C. P., Hoffman, U., Forster, J., Zepp, F., Schuster, A., Wahn, U., & Matricardi, P. M. (2012). Molecular spreading and predictive value of preclinical IgE response to Phleum pratense in children with hay fever. Journal of Allergy and Clinical Immunology, 130(4), 894–901.e5. https://doi.org/10.1016/j.jaci.2012.05.053
- Hilger, C., Kuehn, A., & Hentges, F. (2012). Animal lipocalin allergens. Current Allergy and Asthma Reports, 12(5), 438–447. https://doi.org/10.1007/s11882-012-0283-2
- Hilger, C., Swiontek, K., Arumugam, K., Lehners, C., & Hentges, F. (2012). Identification of a new major dog allergen highly cross-reactive with Fel d 4 in a population of cat- and dog-sensitized patients. Journal of Allergy and Clinical Immunology, 129(4), 1149–1151. https://doi.org/10.1016/j.jaci.2011.10.017
- Jakob, T., Forstenlechner, P., & Matricardi, P. (2015). Kleine – Tebbe J. Molecular allergy diagnostics using multiplex assays: Methodological and practical consideration for use in research and clinical routine. Allergo Journal International, 24(8), 320–332. https://doi.org/10.1007/s40629-015-0087-8
- Karu, A. E., Lin, T. H., Breiman, L., & Mark, T. (1994). Muldoon & Jean Hsu Use of multivariate sttaistical methods to identify immunochemical cross-reactants. Food and Agricultural Immunology, 6(4), 371–384. https://doi.org/10.1080/09540109409354849
- Kaufman, L., & Rousseeuw, P. J. (1990). Finding groups in data. John Wiley.
- Kitzmüller, C., Zulehner, N., Roulias, A., Briza, P., Ferreira, F., Faé, I., Fischer, G. F., & Bohle, B. (2015). Correlation of sensitizing capacity and T-cell recognition within the Bet v 1 family. Journal of Allergy and Clinical Immunology, 136(1), 151–158. https://doi.org/10.1016/j.jaci.2014.12.1928
- Konradsen, J. R., Nordlund, B., Onell, A., Borres, M. P., Grönlund, H., & Hedlin, G. (2014). Severe childhood asthma and allergy to furry animals: Refined assessment using molecular-based allergy diagnostics. Pediatric Allergy and Immunology, 25(2), 187–192. https://doi.org/10.1111/pai.12198
- Liu, C.-Y., Tao, S., Xue, J.-Y., Zhang, H., Xue, W.-T., & Chen, F.-S. (2014). Identification and purification of a novel fish allergen from largemouth bass (Micropterus salmoides). Food and Agricultural Immunology, 25(1), 70–81. https://doi.org/10.1080/09540105.2012.745122
- Ma, S., Wang, R., Nie, L., & Yin, J. (2018). Pollen – food allergy syndrome in China. Food and Agricultural Immunology, 29(1), 281–293. https://doi.org/10.1080/09540105.2017.1372372
- Matricardi, P. M., Kleine-Tebbe, J., Hoffmann, H. J., Valenta, R., Hilger, C., Hofmaier, S., Aalberse, R. C., Agache, I., Asero, R., Ballmer-Weber, B., Barber, D., Beyer, K., Biedermann, T., Bilò, M. B., Blank, S., Bohle, B., Bosshard, P. P., Breiteneder, H., Brough, H. A., … Ollert, M. (2016). EAACI molecular Allergology User's guide. Pediatric Allergy and Immunology, 27(Suppl 23), 1–250. https://doi.org/10.1111/pai.12563
- Melioli, G., Bonifazi, F., Bonini, S., Maggi, E., Mussap, M., Passalacqua, G., Rossi, E. R., Vacca, A., & Canonica, G. W. (2011). The ImmunoCAP ISAC molecular allergology approach in adult multi-sensitized Italian patients with respiratory symptoms. Clinical Biochemistry, 44(12), 1005–1011. https://doi.org/10.1016/j.clinbiochem.2011.05.007
- Nicolaou, N., Poorafshar, M., Murray, C., Simpson, A., Winell, H., Kerry, G., Härlin, A., Woodcock, A., Ahlstedt, S., & Custovic, A. (2010). Allergy or tolerance in children sensitized to peanut: Prevalence and differentiation using component-resolved diagnostics. Journal of Allergy and Clinical Immunology, 125(1), 191–197. https://doi.org/10.1016/j.jaci.2009.10.008
- Nilsson, O. B., Binnmyr, J., Zoltowska, A., Saarne, T., van Hage, M., & Grönlund, H. (2012). Characterization of the dog lipocalin allergen Can f 6: The role in cross-reactivity with cat and horse. Allergy, 67(6), 751–757. https://doi.org/10.1111/j.1398-9995.2012.02826.x
- Popescu, F. D. (2014). Molecular biomarkers for grass pollen immunotherapy. World Journal of Methodology, 4(1), 26–45. https://doi.org/10.5662/wjm.v4.i1.26
- Reese, G., Jeoung, B. J., Daoul, C. B., & Lehrer, S. B. (1997). Characterization of recombinant shrimp allergen Pen a 1 (tropomyosin). International Archives of Allergy and Immunology, 113(1-3), 240–242. https://doi.org/10.1159/000237558
- Reese, G., Schicktanz, S., Lauer, I., Randow, S., Luttkopf, D., Vogel, L., Lehrer, S. B., & Vieths, S. (2006). Structural, immunological and functional properties of natural recombinant Pen a 1, the major allergen of Brown Shrimp, Penaeus aztecus. Clinical and Experimental Allergy, 36(4), 517–524. https://doi.org/10.1111/j.1365-2222.2006.02454.x
- Rousseeuw, P. J. (1987). Silhouettes: A graphical aid to the interpretation and validation of cluster analysis. Journal of Computational and Applied Mathematics, 20, 53–65. https://doi.org/10.1016/0377-0427(87)90125-7
- Saarelainen, S., Rytkönen-Nissinen, M., Rouvinen, J., Taivainen, A., Auriola, S., Kauppinen, A., Kinnunen, T., & Virtanen, T. (2008). Animal-derived lipocalin allergens exhibit immunoglobulin E cross-reactivity. Clinical & Experimental Allergy, 38(2), 374–381. https://doi.org/10.1111/j.1365-2222.2007.02895.x
- Spieksma, F., & Voorhorst, R. (1969). Comparison of skin reactions to extracts of house dust, mites, and human skin scales. Allergy, 24(2), 124–146. https://doi.org/10.1111/j.1398-9995.1969.tb03762.x
- Tovey, E. R., Chapman, M. D., & Platts-Mills, T. A. (1981). Mite faeces are a major source of house dust allergens. Nature, 289(5798), 592–593. https://doi.org/10.1038/289592a0
- Tovey, E. R., Chapman, M. D., Wells, C. W., & Platts-Mills, T. A. (1981). The distribution of dust mite allergen in the houses of patients with asthma. American Review of Respiratory Disease, 124, 630–635.
- van Hage, M., Hamsten, C., & Valenta, R. (2017). ImmunoCAP assays: Pros and cons in allergology. Journal of Allergy and Clinical Immunology, 140(4), 974–977. https://doi.org/10.1016/j.jaci.2017.05.008
- van Loon, L. C., & van Strien, E. A. (1999). The families of pathogenesis-related proteins, their activities, and comparative analysis of PR-1 type proteins. Physiological and Molecular Plant Pathology, 55(2), 85–97. https://doi.org/10.1006/pmpp.1999.0213
- Vieths, S., Scheurer, S., & Ballmer-Weber, B. (2002). Current understanding of cross-reactivity of food allergens and pollen. Annals of the New York Academy of Sciences, 964(1), 47–68. https://doi.org/10.1111/j.1749-6632.2002.tb04132.x
- Zahradnik, E., & Raulf, M. (2014). Animal allergens and their presence in the environment. Frontiers in Immunology, 5, 76. https://doi.org/10.3389/fimmu.2014.00076