Fructose biphosphate aldolase: A new cassava allergen

Abstract

Background

Food allergy has considerably increased in recent years and this situation has been aggravated mainly by the consumption of more processed and complex foods, since minor or potentially allergenic foods are not required to be labeled. Manihot esculenta (cassava) is a widely consumed food in South America, Africa, and Asia and can be used in the production of flour and starch, as well as several other products. This root can cause allergic reactions with symptoms ranging from mild to severe.

Methods

Thus, the aim of this study was the characterization of the immunogenic cassava proteins responsible for sensitizing patients allergic to it. Using a 2D-SDS-PAGE based proteomic approach, six proteins were identified, including Fructose Bisphosphate Aldolase (FBA). Recombinant FBA was produced in Expi293 cells and evaluated by immunoblotting with the serum of 10 individual study subjects.

Results

Our results showed six cassava IgE-reactive proteins. From those, recombinant fructose bisphosphate aldolase (FBA) showed a positivity of 80% among tested sera, proving to be a highly sensitizing protein.

Conclusion

The recombinant FBA molecule obtained in this study can be important for in vivo diagnostic assays, by producing more accurate results, and for desensitization protocols, in which the use of the isolated molecule produces more precise results by avoiding secondary sensitization.

Trial registration

All patients signed a consent form approved by the internal ethics committee CAPPesq, Comissão de Ética para Análise de Projetos de Pesquisa do HC FMUSP (CAAE: 10420619.6.0000.0068).

Background

More than 220 million people worldwide suffer from food allergy, and in recent decades this number assumed an increasing prevalence and severity.^1,2 The food allergy situation was worsened mainly by the consumption of more processed and complex foods that may contain minor or potentially allergenic foods, which can be validated by the higher food allergy prevalence in more industrialized and western regions.^2,3

Cassava (Manihot esculenta crantz) is a shrub of the Euphorbiaceae family. Its tuberous roots composed of 85% starch and only 2% protein, and it has a high energy value. Cassava is native to South America, but can also be found in Asia and Africa as it is commonly cultivated in tropical and subtropical regions. It is highly tolerant of drought and harsh weather conditions and very productive in poor soils and marginal lands, being available throughout the year due to the flexibility of planting and harvesting the crop. These nutritional and agronomic characteristics establish cassava as a reliable crop for food security in many countries.^4,5

Cassava is the third largest food source of carbohydrates and vitamins for about 700 million people, especially in developing countries, being an important raw material for starch extraction.5, 6, 7 In Brazil, cassava is known as “mandioca”, but it can be found with others names in other countries such as: “tapioca” in, Thailand and Vietnam, “cassava” in West India, and “yucca” in other South America regions.^8,9 It can be consumed fried or cooked and serves as a basis for the production of several products such as flour, which is part of the daily meals of most Brazilians, animal feed, alcohol, sweets, glue for paper and fabrics, and biodegradable plastics.^5,10,11

So far, 3 allergens have been described for cassava. FBA (Fructose-Bisphosphate Aldolase), with a theoretical molecular mass of 39.9 kDa, GAPDH (Glyceraldehyde-3-Phosphate Dehydrogenase) theoretical molecular mass of 36.8 kDa, and Man e 5 (glutamic acid rich protein) theoretical molecular mass of 30 kDa.⁸ In inhibition assays, Man e 5 partially inhibited IgE-reactivity to Hev b 5 (a described latex allergen), suggesting that the IgE-reactivity to cassava may occur because of prior sensitization by Hev b 5.⁸

Here we characterized immunogenic cassava proteins using sera from cassava-allergic patients followed by the development and IgE-reactivity testing of a recombinant FBA molecule.

Materials and methods

Patient selection

Based on clinical history, 10 patients with cassava allergy confirmed by prick-to-prick test with raw cassava were selected from the Allergy and Immunology Clinic from Ambulatório de Alergia e Imunologia do Hospital das Clínicas da Universidade de São Paulo. Sera from these patients were used in ImmunoCAP 100 tests (Thermo Fisher) for detection of specific IgE for latex and its available allergens: Hev b 1 (rubber elongation factor), Hev b 3 (small rubber particle protein), Hev b 5 (acidic protein), Hev b 6 (hevein precursor), Hev b 8 (profilin) and Hev b 11 (class I chitinase). All patients signed a consent form approved by the internal ethics committee (CAAE: 10420619.6.0000.0068). Besides, 2 individuals, 1 atopic without manic allergy and 1 non-atopic were included as controls for 1D immunoblot.

Cassava extract preparation

Cassava (Manihot esculenta) was purchased from a local supermarket and properly sanitized in running water. The protein extract was produced from maceration of freeze-dried cassava (Super Modulyo – Thermo Savant) using the phenolic extraction and precipitation protocol with 100 mM ammonium acetate/methanol in the presence of protease inhibitors, adapted from Carpentier et al.¹² The protein concentration was measured by Bradford¹³ and the extract was stored at −20 °C until further use.

1D and 2D gel electrophoresis

For 1D gel electrophoresis, 20 μg of cassava protein extract was solubilized in denaturing/reducing sample buffer and applied to each well of a 12% (m/v) SDS-PAGE. For the isoelectric focusing (IEF) of 2D electrophoresis, the proteins were solubilized in a rehydration solution (DeStreak plus 0.5% (v/v) IPG buffer (GE Healthcare Biosciences AB, Uppsala, Sweden). Immobiline Dry Strips (IPG) with a pI range of 4–7 were rehydrated in this protein solution overnight aiming at a good separation of different proteins with different pIs likely to be found in the whole extract. The strips were subjected to IEF in an IPGphor system (GE Healthcare Biosciences AB, Uppsala, Sweden) at 300V, 1000V and 5000V until reaching 5750 Vh. For reduction and alkylation, the two following steps were performed, an incubation with 1% (w/v) DTT for 15 min followed by 4% (w/v) IAA (iodoacetamide) for 15 min, before strips were loaded onto 12% (w/v) polyacrylamide gels for proteins separation by reducing SDS-PAGE. Proteins were stained with Coomassie Blue Colloidal¹⁴ and gel images were captured using ImageQuant LAS 4000 (GE Healthcare Biosciences AB, Uppsala, Sweden).

1D and 2D immunoblot

After separation, proteins from the 1D and 2D gel were electrotransferred onto nitrocellulose membranes and subsequently blocked with TBS added with 0.1% (v/v) Tween 20 and 5% of (w/v) nonfat dry milk (Biorad – Blotting-Grade Blocker) for total extract and at 10% (w/v) for recombinant FBA for 1 h at room temperature. Membranes were incubated overnight at 4 °C with patient sera at 1:5 individually for 1D and (1:10) pooled for 2D. Bound IgE was detected with goat anti-human IgE (ε), HRP conjugate (Invitrogen-California, USA) standardized at 1:5000 as a second antibody. Immunoreactive bands were scanned by ImageQuant LAS 4000 (GE Healthcare Biosciences AB, Uppsala, Sweden) using Enhanced Chemiluminescence (ECL) Kit (GE Healthcare Biosciences, Uppsala, Sweden).

Mass spectrometry and data analysis

For mass spectrometry analyses, IgE-reactive spots from 2D-gels were digested with trypsin (Promega - Madison, WI, EUA) according to the protocol designed by Shevchenko et al,¹⁵ and the mass spectrometry analyses of cassava 2D gel excised spots were performed according to Dias et al.¹⁶ Tryptic digests of the cassava were solubilized in 0.1% (v/v) formic acid solution (solution A) and submitted to Ultimate 3000 HPLC (Dionex, Germering, Germany) coupled to a Q Exactive Orbitrap™ mass spectrometer (Thermo Fisher Scientific, Bremen, Germany). Peptides were loaded on a Trap Column with nano Viper Fitting (P/N 164,649, C18, 5 mm × 30 μm, Thermo Fisher Scientific, Bremen, Germany) and eluted at a flow rate of 300 nL/min using an isocratic gradient of 4% solution B (100% acetonitrile containing 0.1% (v/v) formic acid) for 3 min. Thereafter, peptides were loaded on a C18 PicoChip column (Reprosil-Pur®, C18-AQ, 3 μm, 120 Å, 105 mm, New Objective, Woburn, MA, USA) using a segmented concentration gradient from 4 to 55% B for 30 min, 55–90% B for 1 min, 90% B over 5 min and then returning to 4% B over 20 min at a flow rate of 300 nL/min. Ion polarity was set to positive ionization mode using data-dependent acquisition (DDA) mode. Mass spectra were acquired with a scan range of m/z 200–2000, resolution of 70 000 and injection time of 100 ms. The fragmentation chamber was conditioned with collision energy between 29 and 35% with a resolution of 17 500, 50 ms of injection time, m/z 4.0 of isolation window, and dynamic exclusion of 10 s. The spectrometric data were acquired using the Thermo Xcalibur™ software (version 4.0.27.19, Thermo Fisher Scientific Bremen, Germany).

Raw data were submitted to the PatternLab for Proteomics software version 4.0.0.84¹⁷ and searched against the NCBI database (http://blast.ncbi.nlm.nih.gov) by using the Taxonomy Manihot. The following search parameters were used: semi-tryptic cleavage products (2 tryptic missed cleavage allowed), carbamidomethylation of cysteine as fixed modification, and oxidation of methionine as variable modification. Parent mass tolerance error was set at 40 ppm and fragment mass error at 0.02 ppm. Protein identifications were considered with a minimum of 1 fragment ion per peptide, 5 fragment ions per protein, 2 peptides per protein, and a false discovery rate (FDR) set to lower than 1%, estimated by a simultaneous search against a reversed database.

Construction of FBA expression clones

The gene of fructose bisphosphate aldolase (1164 bp) was optimized with codons preferentially used by mammals, a Kozak sequence included at the 5′ end to favor protein expression, an IgE signal peptide to target proteins for exocytosis, and a histidine tag, for further purification on a nickel column. The designed gene was synthesized (GenOne Biotech©) and cloned into the expression vector pcDNA3.1 (Thermo Fisher), using the restriction enzymes XbaI and ApaI (Invitrogen).

Cell transfection and protein expression

The human cell line Expi293F was cultured in suspension in the recomended Expression medium (Gibco A14351-01) in shake flasks under standard procedures at 125 rpm and 37°C with 8% CO₂.

The pCDNA3.1 plasmid containing the optimized FBA gene was transfected into 300×10⁶ Expi293F cells in 100 ml Expi293 Expression Medium using ExpiFectamine 293 Transfection Kit and Opti-MEM Reduced Serum Medium (GIBCO BRL) for 16–18 h, added with 600 μl Enhancer 1 and 6 ml Enhancer 2. Cell cultures were expanded for 5 days to allow for protein expression. Cell viability was determined by Trypan Blue exclusion assay.

Recombinant protein purification

Due to the presence of histidine tag in the protein, purification of the supernatant was performed by metal affinity chromatography (IMAC). The supernatant containing the proteins was diluted 1:1 in phosphate-buffered saline (PBS, pH 7.4) along with resin (Ni Sepharose 6 Fast Flow - Cytivia) and 5 mM imidazole for every 100 mL of diluted supernatant. This mixture was left under gentle agitation for 1 h and 30 min at 4C. For protein purification, the pre-treated crude supernatant was loaded onto a plastic column (20 mL), and this loaded column was washed with PBS containing 5 mM imidazole. The protein was then eluted in 50 mM, 250 mM and 500 mM imidazole. Finally, the elution was buffer exchanged with PBS by dialysis. Protein concentration was determined in Spectrophotometer (Nanodrop ND1000 - Uniscience), concentration (M) = Absorbance at 280 nm/extinction coefficient.

Simulation of gastric digestion with pepsin

To assess protein stability during gastric digestion, an in vitro simulation of rFBA protein degradation in the presence of the enzyme Pepsin (Sigma - P6887) was performed. The enzyme was diluted in HCL 0.1 M pH 2 to a concentration of 3 μg/μL and stored at −20°C until the time of use as a stock solution. For the experiment, in microtubes, 60 μg of rFBA protein was incubated along with 0.4% pepsin and 114 μL HCL 0.45 M pH 2 solution at 37°C under gentle agitation in ThermoMixer C (Eppendorf). The reaction was then stopped at different timepoints: 10 min, 30 min and 1 h, with 2 μL of NaOH 1 M solution and the microtubes were frozen at −20°C until application on a 12% SDS-PAGE acrylamide gel for visualization, in which 3 μg of protein were loaded per well.

Results

The 10 individuals included in this study had clinical symptoms of cassava allergy which were confirmed by prick-to-prick test with raw cassava. Six were also allergic to latex, in addition to other fruits (Table 1). For this reason, prick to prick test for latex (glove) and ImmunoCAP specific IgE for latex and its individual allergens were performed. All individuals were female with a median age range of 55 years. Demographic data and skin tests and ImmunoCAP results are summarized in Table 1.

Table 1.

Demographic, clinical features and sensitivities of patients

Patient	Age	Health Professional	Symptoms after latex exposure	Symptoms after manioc exposure	Episodes of Anaphylaxis	Prick to Prick Raw Cassava	Prick to Prick Rubber Latex	ImmunoCAP						Other Allergies
								Latex	Hev b 3	Hev b 5	Hev b 6	Hev b 8	Hev b 11
1	72	No	S, NS, EI	D, C, S, CO, A, N	3	+	+	+	-	+	+	-	-	Latex, Papaya, Peach
2	55	Yes	U, A	D, U	2	+	-	+	-	+	+	-	+	–
3	61	Yes	C, S, SI	D, S, CO, A	1	+	-	-	-	-	-	-	-	–
4	55	Yes	SI, D, C	D, S, CO, U, A	1	+	+	+	-	+	-	-	-	Kiwi, Passion Fruit, Pineapple, Avocado, Grape. Eggplant
5	31	Yes	NS	A	0	+	-	-	-	-	-	-	-	–
6	37	Yes	NS	U, A	0	+	+	+	-	+	-	-	-	Latex
7	55	No	U	D, C, S, CO, U, A, N, DI	7	+	+	+	-	+	-	-	-	Latex, Papaya, Peach, Banana
8	52	Yes	D, U, A	D, C, S, CO, U, A	4	+	+	+	+	+	+	-	-	Latex, Papaya
9	70	No	NI	D, C, S, CO, U	1	+	+	-	-	+	+	-	+	Latex, Papaya, Pineapple
10	62	Yes	NI	D, A, N, AP	ND	+	+	+	+	+	-	-	-	Latex, Papaya, Avocado

D: dyspnea; C: cough; S: sneeze; CO: coryza; U: urticaria; A: angioedema; N: nausea; AP: abdominal pain; DI: diarrhea; SI: skin itching; NS: no symptons; NO: nasal obstruction; EI: eye itching; NI: no information; Prick to Prick + positive (≥ 3x3 mm) - negative (≤ 3x3); ND: not done; ImmunoCAP + positive (≥ 0,35 kUA/L); Hev b (Hevea brasiliensis)

We were able to extract 1.5 mg of protein from 2 g of freeze-dried cassava using our precipitation protocol. Cassava proteins were evaluated on 1D SDS-PAGE and showed a wide variety of bands ranging from about 8 to 120 kDa (Fig. 1a). Primary investigation of IgE reactivity by 1D immunoblotting using individual sera from allergic patients revealed different IgE sensitization profiles with various bands being recognized mainly from 25 to 110 kDa. For control, sera from non-allergic individuals were included and no bands were visible. Also, a lane with the secondary antibody alone without any serum was run. In this case it was possible to visualize some very faint bands in the 2nd antibody control lane that should not interfere with the results due to their low density (Fig. 1a). From a 2D immunoblotting approach with pooled sera, seven spots between 39 and 110 kDa and between isoelectric points from 4 to 7 were detected, and therefore excised from the gel for identification (Fig. 1b). Of these, all had positive hits in the database search (Table 2) and those with >2 unique peptides and molecular weight corresponding to that found in the 2D gel were selected. All proteomic data for spots identification are shown in the Supplementary Table 1.

Fig. 1.

SDS-PAGE and Immunoblot analysis of cassava extract. (a) Cassava extract (E) on reducing SDS-PAGE. Immunoblot of individual allergic patients' sera,1, 2, 3, 4, 5, 6, 7, 8, 9, 10 second antibody control (2nd) and patients non-allergic (C1: non-atopic and C2: atopic) (b) 2D-SDS-PAGE of cassava protein extract (I) and Western blotting (II) showing the IgE binding of pooled sera1, 2, 3, 4, 5, 6, 7, 8, 9, 10 to cassava protein. pI, isoelectric point; MW, molecular weight, kDa, kiloDalton

Table 2.

Allergenic proteins identified in cassava extract

Spota	Access Codeb	Identification	Organism	Functional Category	UPc	Experi. Mw/pld	Theo. Mw/ple
1-2	A0A2C9WB58	Alpha-1,4 Glucan Phosphorylase	Manihot esculenta	Carbohydrate metabolism	2-3	100/5-6	107/5.94
3-4	A0A251IQB4	Peptidase_S9	Manihot esculenta	Proteolysis	19-36	72/5-6	74/5.04
5	A0A2C9VPN2	Cytosol_AP	Manihot esculenta	Proteolysis	33	60/6,5	60/6.69
6	A0A2C9VH59	FBA	Manihot esculenta	Glycolysis	9	36/7	38/6.91
6	A0A2C9VQ42	GAPDH	Manihot esculenta	Oxidoreductase	4	36/7	36/7.73
7	A0A2C9UKP7	ATP Synthase	Manihot esculenta	ATP biosynthetic process	10	55/4-5	60/6.26

^aSpot numbers as indicated in Fig. 1b.

^bAccession numbers according to Uniprot database.

^cUnique Peptides.

^dExperimental molecular mass and pI.

^eTheorical mass kDa and pI of identified proteins retrieved from Expasy.org.

Six proteins were identified in the seven excised spots, including apha-1,4 glucan phosphorylase, peptidase_S9, cytosol_AP domain-containing protein, fructose biphosphate aldolase (FBA), glyceraldehyde 3 phosphate dehydrogenase (GAPDH), and ATP synthase. These proteins are related to carbohydrate metabolism, proteolysis, glycolysis, oxidoreductase, and ATP biosynthetic process. According to our research on Allergome, only ATP synthase and FBA have been reported as allergens, and only FBA is included in the World Health Organization (WHO) and International Union of Immunological Societies (IUIS) Allergen Nomenclature Subcommittee database, while the other proteins identified in the 2D WB are not yet reported as allergenic.

Since FBA has already been described as an allergen and as our patients’ sera recognized the protein in the cassava extract, a recombinant FBA protein (rFBA) was produced and expressed in an Expi293F cell system to determine whether this protein could be specifically recognized by IgE derived from patient sera. To investigate this, an immunoblot was performed using the rFBA and individual patient sera. The rFBA protein was specifically recognized by serum IgE from 8 out of 10 patients (Fig. 2). Sera from patients 5 and 9 did not react with FBA.

Fig. 2.

FBA (fructose-bisphosphate aldolase) recombinant protein (R) on reducing SDS-PAGE. Immunoblot of individual allergic patients' sera (n = 10). MW, molecular weight, kDa, kiloDalton

To verify how long rFBA would take to degrade in an environment with a pH similar to that found in the stomach acid. A gastric simulation with the pepsin enzyme was performed. It is possible to observe that after 10 min of incubation, a partial digestion of the rFBA began, at 30 min it was still possible to see some traces of the protein, but after 1 h there was no longer any detection of intact protein (Fig. 3).

Fig. 3.

Gastric simulation with pepsin the protein fructose biphosphate aldolase in SDS gel - PAGE 12 % acrylamide. MM: Molecular marker; C: undigested protein control with pepsin; times: 10 min, 30 min and 60 min - incubation time at 37 °C with pepsin enzyme; kDa: kiloDalton

Discussion

Cassava was elected by the United Nations (UN) as the most relevant food of the 21st century, having been so named it due to its versatility, as it is consumed by people and animals, in addition to being used in the manufacture of various derivatives.¹⁸

In order to investigate potential allergens in cassava, we extracted its proteins and performed immunoblotting with sera from ten patients with confirmed cassava allergy. The blots were incubated with an IgE-specific secondary antibody to ensure that only IgE binding was detected. It is possible to identify different IgE recognition patterns among the individuals and the presence of strongly reactive protein bands (Fig. 1a). Some individuals have several IgE-reactive bands, mainly between 30 and 115 kDa, while others have a cleaner profile, that is, with few IgE bindings (Fig. 1a). Mass spectrometry analysis of the excised 2D-SDS-PAGE spots identified two proteins around 36 and 39 kDa as homologous to GAPDH and FBA respectively, corroborating a previous study¹⁹ when we also showed IgE reactivity to these molecules in another group of patients. As previously reported the protein band around 30 kDa that appears in all samples, except sample 9 is probably Man e 5 that is shown to not be detected on 2D blot (⁸).

The closest species to Manihot esculenta, which also belongs to the Euphorbiaceae family, is Hevea brasiliensis (latex), whose proteome includes FBA and GAPDHs proteins.²⁰ In our cohort, 70% (7/10) were health professionals and of these 42.9% (3/7) from them were are also allergic to latex. It is feasible to infer that sensitization to these allergens form cassava may have occurred due to constant contact with powder from latex glove.

We successfully produced and expressed in an Expi293F cell system an rFBA protein, and this protein can be classified as a major allergen in our cohort, considering the criteria for allergen nomenclature, since 8/10 patients recognized the isolated molecule.²¹ From the 8 individuals who showed IgE reactivity against rFBA protein, 6 are healthcare professionals, showing that FBA can be a cross-reactive allergen. Of note, patient 5 is not allergic to latex and did not react against FBA indicating she is sensitized to another exclusive allergen from manioc, not yet identified, that do not cross react with latex.

FBA is a molecule with native tetrameric conformation, labile and the main routes of exposure described are by inhalation and ingestion.^22,23 This protein has been described by the IUIS as an allergen in several sources, being first described in cockroach²⁴ and also found in other sources such as: shrimp^22,25 being also related to anaphylaxis after physical activity,^26,27 tuna, salmon, and cod.²⁸

Additionally, FBA from wheat flour was identified as IgE-reactive²⁹ in individuals with baker's asthma, one of the most common causes of occupational asthma. However, in our cohort, individuals who presented IgE reactivity to rFBA did not report allergic symptoms to wheat and sensitization to wheat was not tested. Interestingly though, all but one of the rFBA positive individuals have respiratory symptoms. In both cassava and wheat, FBA has a close molecular mass, 36 and 37 kDa respectively,³⁰ with 87% of identity between them considering amino acid sequence (data not shown). As baker's asthma, sensitization occurs in a very specific group of individuals exposed by inhalation, our cohort does not appear to be sensitized by wheat. For cassava, sensitization is more likely have happened by ingestion.

Since FBA proteins from both sources are remarkably similar, differences in allergenicity may be due to cooking process and the presence of different food matrices. It is known that alterations in the molecules such as denaturation, conformational and structural changes due to the loss of secondary and/or tertiary interactions or the formation of new rearrangements of disulfide bonds, can make foods become more or less allergenic.^31,32 Studies have shown that food matrix components can not only alter the structure of allergens, but also change the biological processes of allergens sensitization in the human body, thereby affecting their allergenicity. Processing causes physical-chemical interactions between components released from the food matrix and the allergens. These processes can promote the formation of aggregate between allergens and the matrix, limiting the destruction of the allergen epitope, but can also lead to the generation of new allergens.³³ Further studies are necessary to elucidate the mechanisms that may be involved in this case.

GAPDH that was also herein identified as IgE-reactive has been already recognized by the IUIS as allergen in other sources such as Pan h 13 in Striped catfish, Per a 13 from the American cockroach and Tri a 34 from wheat. Furthermore, although not considered in the official IUIS nomenclature, it has also been reported as an allergen of latex,²⁰ macadamia nut,³⁴ ribwort plantain³⁵ and also Canabis sativa.³⁶ GAPDH is strongly involved in glycolysis/gluconeogenesis and carbon fixation in photosynthetic plants as part of the pentose phosphate reductive cycle for energy metabolism³⁴

Among the proteins identified in 2D immunoblotting, only ATP synthase (protein responsible for ATP synthesis and known to be involved in the regulating plant cell death)³⁷ has already been described as an allergen in pollen³⁷ and in Cannabis sativa.³⁶ Three proteins not previously identified as IgE-reactive were identified for the first time in this study and are potential novel cassava allergens: Alpha-1,4 Glucan Phosphorylase, an enzyme known to play a role in both starch synthesis and degradation and its association with carbohydrate and energy metabolism, can be found in abundance in green fruits, such as mango and banana, where its presence is related to the ripening process of the fruit;^38,39 Peptidase S9, an enzyme involved in proteolysis;⁴⁰ and finally cytosol_AP, which has a molecular function related to metalloaminopeptidase activity.⁴¹

In recent years, the food allergy research focused on the allergenic potential of purified allergens rather than the whole food. The production of recombinant forms has proven to be quite efficient due to their higher purity index when compared to natural extracts, in which protein concentrations are varied.⁴² The use of recombinant allergens in the clinic can increases the specificity of the skin test by allowing only one specific component to sensitize the skin without the risk of exposing the individual to a set of other sensitization components, thus improving the guidance of the desensitization process. Even so, although their standardization may still be a problem, extracts from natural sources are still the only ones available for allergy diagnosis by skin tests, because. Despite the great advances in the characterization of allergens, its recombinants proteins for in vivo use are still difficult to obtain as they demand rigorous and expensive tests are necessary for their manufacture and commercialization.^42,43

Furthermore, our cohort is comprised only of women. It has been described that female sex hormones increase the risk and levels of symptoms in food allergies. In cases of asthma, it has already been proven that estrogen has an aggravating role while testosterone has a protective role.⁴⁴ In addition, the greater permeability of the female skin and mucosa barrier, combined with a lower metabolic capacity, can be decisive for the onset of symptoms, as more disease-causing foods are absorbed and take longer to leave the body.^45,46

Conclusion

Altogether, herein we identified seven IgE-reactive proteins in Manihot esculenta, were described, including two that were previously identified as allergens and included a patient allergic to manioc that do not react to latex indicating the presence of allergens in cassava that are not cross-reacting. We also produced a recombinant allergen that can lead to the development of new diagnostic strategies using the specific molecule, such as more specific in vivo tests and therapies in individuals allergic to Manihot esculenta.

Abbreviations

FBA, Fructose-Biphosphate Aldolase; GAPDH, Glyceraldehyde-3-Phosphate Dehydrogenase; SDS-PAGE, Sodium Dodecyl-Sulfate Polyacrylamide Gel Electrophoresis; IEF, Isoeletric Focusing; pI, Isoeletric Point; 1D, One-Dimensional; 2D, Two-Dimensional; ECL, Enhanced Chemiluminescence; DDA, Data-Dependent Acquisition; FDR, False Discovery Rate; IMAC, Metal Affinity Chromatography; PBS, Phosphate-Buffered Saline; rFBA, Recombinante Protein Fructose-Biphosphate Aldolase; HCL, Hydrochloric Acid; NaOH, Sodium Hydroxide; kDa, Kilodalton; IgE, Immunoglobulin E; WHO, World Health Organization; IUIS, International Union of Immunological Societies.

Funding

This research was funded by Young Research Award of FAPESP (Project 2012/14019-2), Conselho Nacional de Desenvolvimento Científico e Tecnológico (INCT/CNPq-iii grant 465434/2014-2), Ventura, AKRM and Alves, SP were beneficiary of fellowships from CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—Finance Code 001).

Availability of data and materials

The data that support the findings of this study, which are captured in the current article, are available from the corresponding author upon reasonable request.

Author contributions

We hereby state that all authors have contributed substantially for this work to be completed, fulfilling the authorship criteria.

Ethics approval

The study was reviewed and approved by an Hospital das Clínicas da Universidade de São Paulo, approval number 10420619.6.0000.0068.

Authors’ consent for publication

All authors have reviewed this manuscript and consent to its publication.

Declaration of competing interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Appendix A. Supplementary data

The following is the Supplementary data to this article: