Food allergy (FA) is an adverse health effect due to an immune response that occurs reproducibly after exposure to a given food [1]. Allergies have an impact on the health of the patient, cause expenses for being a selective disease, imposing on their patients certain specific eating behaviors and other expenses with diagnosis and treatment [2].

Thus, FA comes with significant social and economic impact, worrying the scientific community and health professionals. Currently, food allergy is present in more than 220 million people in the world [3]. Therefore, approximately 8% of the infant population and 5% of the adult world population present allergy to at least one food [4].

As it is a larger child population, in recent years new perspectives in the diagnosis for food allergy in vivo and in vitro have been elucidated, evaluating not only the immunological mechanism with the presence of IgE antibodies, but also the design of the clinical phenotype of food hypersensitivity reactions, thus ensuring the diagnosis of reliable allergy [3].

However, the methods currently available to diagnose FA are still invasive, causing discomfort to the patient, often present high costs, and are restricted to one food per test. In addition, the test considered gold standard in diagnosis, requires hospital environment and trained staff, and may trigger life-threatening reaction, leading to anaphylactic shock in the patient. Moreover, this variety of methods only evaluates the presence of IgE class specific antigen antibodies [5].

In addition to the existence of diagnostic tests through the detection of IgE, other immunoglobulins are also being studied for this purpose. Mention may be made of food antigen-specific IgG₁ and IgG₄ panels, as these antibodies have been cited by some authors as a possible alternative for the diagnostic of FA, but currently only IgG₁ is being further emphasized by the fact that IgG₄ does not have specific receptors in the cells, which prevents the formation of the clinical reaction of food hypersensitivity [6].

Among the samples investigated for use in the diagnosis of FA [7], saliva has been pointed out by researchers, due to the significant presence of secretory IgA, IgG and IgM immunoglobulins [8] and already have been used for the diagnosis of HIV (human immunodeficiency virus), cancer, tuberculosis and Helicobacter Pylori infection [9].

The racional behind the technique used in this study is that some antibody associated with allergy found in the bloodstream can also be found in the saliva, thus allowing them to be identified through immunologycal testing [10,11]. In view of the above stated, this study aimed to verify the possibility of using saliva to detect IgG₁ for the diagnosis of food allergy.

The quantification of the total proteins of the 14 foods in the study was performed by the Bradford method (1976), presented in Table 1, and the electrophoretic profile for protein characterization were obtained by SDS-PAGE, which revealed the migration of protein fractions with an estimated molecular mass in the range of < 14.4 and < 97.0 kDa, which can be observed in Fig 1.

The ELISA test showed a 100% level in both specificity and sensitivity compared to the RAST test. As shown in Table 2, in all the patients that presented allergy to cow’s milk protein, IgE and IgG₁ was detected in serum and in saliva for the other foods studied, by the ELISA test. On the other hand, no reaction was observed with the control group.

After evaluating the test with cow’s milk protein, it was in the interest of the study to verify if children allergic to cow’s milk protein could present allergy to other foods. The fish, shrimp, egg white, soybean, wheat, peanut, brown, corn, papaya, banana, kiwi, tomato and cocoa proteins were tested in the experimental group, 100% of the sample was also allergic to fish (Table 2). There was a higher prevalence of allergy for ELISA-tested patients in shrimp followed by wheat, egg white, soy, brown, peanut, corn, banana, papaya, tomato, kiwi and cocoa, which presented the lowest percentage of allergic sensitivity, that the animal protein group was the most sensitive when compared to the other groups.

When the means of IgE and IgG₁ concentrations in serum were compared, all foods had no statistical difference (p> 0.05), except for cow’s milk and fish (p < 0.001; p < 0.010), respectively (Table 3).

When comparing the mean of IgE and IgG₁ concentrations in saliva, all foods had no statistical difference (p > 0.05), except for cow’s milk and tomato, which presented a difference (p < 0.001, p < 0.003) (Table 4).

When analyzing the mean of IgE concentration against shrimp, egg, soybeans, wheat, chestnut, peanut, kiwi, banana, tomato and cocoa; no significantly difference between Serum and Saliva was found (p > 0.05). However, the mean of IgE concentration against cow’s milk and papaya in serum was significantly lower than in Saliva (p < 0.05). In addition, the mean of IgE concentration against fish and corn in serum was significantly higher than in Saliva (p < 0.05) (Table 5).

Table 6 shows that the mean concentration of IgG₁ in cow’s milk, fish, shrimp, egg, soy, wheat, brown, peanut, kiwi, papaya, banana, maize and cocoa did not differ significantly between serum and saliva (p > 0,05). But the mean IgG₁ concentration of the tomato in the Serum was significantly lower than in Saliva (p < 0,05).

The diagnosis of food allergy by the detection of specific IgE and IgG₁ immunoglobulins using saliva was reported in this study, making this article the first work that reports the possibility of a noninvasive diagnosis for food allergy. In addition, the interaction of allergenic proteins with serum and saliva can be observed by detecting both IgE and IgG₁.

High levels of salivary IgE and IgG₁ specific for precipitated cow’s milk proteins from 36 allergic children were detected by ELISA using the RAST test. Likewise, 100% of the experimental group also presented high levels of IgE and IgG₁ specific for the fish. Although fish is not usually used in the feeding of children in early childhood, the result suggests that these children have become sensitized by other routes of immunization other than oral, since fish is a food widely used by the population of Ceará. This is according to SICHERER et al. (2011) [15] which showed that most peanut allergy children experience their first allergic reaction to peanuts at the first intake, suggesting that the sensitization that resulted in IgE production must have occurred through exposure through a non-oral route. Two leading theories for the basis of such sensitization are in utero sensitization, or by home exposure through non-oral routes.

When analyzed with maize extract, the serum and salivary levels of specific IgE were observed in 36% and 39% of the samples, repectively. Similarly, IgG₁ were detected in 25% in the allergic group. Corn is a cereal widely used in Brazilian food, however, few studies have addressed specific allergic reactions to this food. Nonetheless, in one of them, performed in a multicenter study in eight different regions of Brazil (n = 306), being tested by means of the prick test, Rizzo et al (1995) [16] found a prevalence of 10.1% positive cutaneous reactivity for corn in the northeast region.

When soybean was used as antigen, the presence of specific serum IgE and IgG₁ were detected in 52% and 55% respectively and salivary IgE and IgG₁ in 55% and 53% of the allergic group. Similarly, for peanuts, serum IgE and IgG₁ reactivity were 36% and 39%, respectively; IgE and IgG₁ salivary 25% and 25% respectively. Still in Rizzo’s study et al (1995) [16] a prevalence of 10.1% of cutaneous reactivity was also found for soybean and 8.4% for egg and peanut in northeastern Brazil.

For the egg white the presence of IgE and IgG₁ in the serum was 39% and 60%, respectively; while in saliva, IgE was observed in 55% and IgG₁ in 44%. When comparing specific IgE and IgG₁ levels in serum and saliva for the foods in this study, there was no significant difference.

These results demonstrated that, when comparing the levels of specific IgE and IgG₁ immunoglobulins in the serum and saliva for the foods tested, no statistical difference between them are found for most foods.

Although IgE is the immunoglobulin identified as the mediator of food allergic reactions, several evidences show that systemic anaphylaxis can also be induced by IgG class immunoglobulins [6]. Mast cells can respond to both IgE and IgG antibodies. IgG antibodies in FA can influence the formation of immunocomplexes contradicting the traditional concept that these antibodies inhibit allergic processes [17]. Therefore, immunoglobulin G serves as a mediating effect rather than inhibiting the hypersensitivity reaction caused by food intolerance. Due to the great diversity of allergenic food products, it is believed that immunological reactions may be very different and their mechanisms may work in a more complicated way, requiring further studies [18].

Corroborating with the theory of IgG participate in the allergic process, Finkelman, Khodoun and Strait (2016) [6], observed in their study with mice that human IgG₁, IgG₂ and IgG₄ can bind to FcγRIII high affinity receptors, activating mast cells, basophils, neutrophils and macrophages, which leads to the production of PAF (platelet activation factor) in anaphylaxis in humans.

Both IgE and IgG antibodies are found in the serum of allergic patients, but the relative concentration of the various classes is poorly understood and rarely investigated. This may have important therapeutic consequences if, in fact, neutrophils can induce IgG-dependent anaphylactic reactions in humans [19]. The allergen-specific IgG₁ detection results shown in this study are relevant, since they show the specific IgE and IgG₁ detection for different foods studied in serum and saliva. In addition, this result shows that saliva can be used to diagnose FA.

Several researchers have been demonstrating the use of saliva for diagnosis of different diseases. As can be observed in the Luzza et al (1997) [20] study who compared salivary IgG levels to diagnose Helicobacter Pilory infections with the gold standard diagnostic test, finding significantly higher salivary IgG levels in people with HP infection when compared to the control group, presenting specificity and sensitivity of the salivary test of 82% and 93%.

According to the study by Cai et al. (2014) [21], a high prevalence of serum IgG antibodies to food allergens were found in allergic patients presenting intestinal symptoms, also reporting that this immunoglobulin is a potent indicator for the state of the allergic disease facilitating orientations during the diet. However, Mekkel et al (2005) [22] considers according to his study that IgG is present, in significant amounts, rarely in immediate type reactions, being considered as the best indicator in this type of reaction to immunoglobulin E. Complementing the previous study, circulating IgG was shown to give a more delayed or even asymptomatic response following exposure to a single food in the Crowe and Perdue study (1992) [23], which report that IgG mediated reactions have a slower response after antigen exposure.

In the study made by Hochwallner et al. (2011) [24], it was investigated whether patients with non-IgE-mediated APLV can be distinguished from non-allergic individuals through serum IgA levels and IgG subclasses, showing higher levels of IgG₁ and IgG₄ for some types of caseins and α-lactoglobulin.

The test to diagnose FA currently considered gold standard is the oral provocation test, but it presents risks, a high cost and there are no practical alternatives, parameters and clear guidelines to discriminate which children should take this test [25]. In contrast, Volppi and Maccari (2009) [25] showed in their study that the ELISA technique is highly sensitive and specific for detecting the presence of serum antibodies, through IgE and IgG isotope tests for 160 different dietary proteins in 6879 individuals, technique is considered low cost and does not cause allergic manifestations in the patient.

According to Hiller et al (2002) [26], the ELISA test allows to verify the profile of reactivity and the detection of allergic components by specific IgE in small amounts of serum but can not identify the allergic disease. But, according to Ramos, Lyra, Oliveira (2013) [27] the ELISA test is a less expensive, simpler and more sensitive feature used by many laboratories to test the presence of antibodies in patients allergic to numerous food antigens, since these have high levels of some IgG subtypes. In addition, some companies also produce kits for the detection of IgE, which, together with clinical examination, will aid in the selection of allergenic foods to be avoided by the patient.

Vojdani A. and Vojodani C. (2015) [28] report that some people are sensitive to lipids, and currently the skin tests are performed only with protein components dissolved in water, and may produce false negative results, making it fundamental to develop a methodology for this type of evaluation.

Therefore, Vojdani A. and Vojodani C. (2015) [28] stated that serological tests measuring both levels of IgG and IgE would make possible a diagnosis for more reliable allergy test, making it possible to diagnose both IgE-mediated, non-mediated and mixed allergies.

It was possible to verify with this study the similarity in the specificity and sensitivity that the ELISA test, using both serum and saliva for the detection of IgE and IgG₁, presented for the RAST test.

In this context, it was found that the use of saliva for the detection of IgE is more sensitive to diagnose CMPA alone because it has a higher amount, however IgG₁ also showed efficiency in the diagnosis of food allergy through saliva.

The use of salivary combined with the method to diagnose food allergy is suggested, with IgE and IgG₁ being measured in order to obtain more reliable results as well as to investigate mediated, non-IgE and mixed mediated allergies.