Allergenicity of Deamidated and/or Peptide-Bond-Hydrolyzed Wheat Gliadin by Transdermal Administration

Hydrochloric acid (HCl)-treated wheat protein (HWP) is widely used in various products, including foods, cosmetics and shampoos. Recently, immediate hypersensitivity towards facial soap containing HWP has been reported. HCl treatment of protein causes hydrolysis not only of main-chain amide bonds (peptide-bond hydrolysis) but also of side-chain ones (deamidation). We have already reported that gliadin, the main allergen in wheat, reduces allergenicity and increases digestibility by deamidation, indicating that deamidation and peptide-bond hydrolysis are effective to reduce the allergenicity of wheat protein. However, transdermally administered HWP is assumed to induce sensitization to orally administered wheat protein even in those who have been taking wheat products daily before sensitization. The present study was conducted to examine which structural change is responsible for the induction of cutaneous sensitization by comparing the allergenicity of deamidated and/or peptide-bond-hydrolyzed wheat gliadin. Because we have developed a deamidation method without causing peptide-bond hydrolysis, only deamidated wheat gliadin is available. Therefore, after deamidated-only, hydrolyzed-only, and deamidated and hydrolyzed gliadins were transdermally administered to mice for several weeks, the corresponding gliadin was intraperitoneally administered and allergenicity was evaluated. Transdermal administration of deamidated and hydrolyzed gliadin induced severe allergic reaction, while that of deamidated-only and hydrolyzed-only gliadin showed almost no allergic response. This result indicates that both deamidation and peptide-bond hydrolysis are necessary to increase the allergenic potency of transdermally administered wheat gliadin.

The degree of deamidation was determined as the ratio of the amount of removed acid amide of DG, HG and HDG to the total acid amide of UG, as described previously [17]. Each gliadin was completely deamidated by heating in a 4 M HCl solution, and the amount of ammonia produced was measured by Conway's micro-diffusion method [34,35]. The amount of removed acid amide was calculated by subtracting the amount of ammonia produced from DG, HG and HDG from that produced from UG. The degree of hydrolysis was measured by the method described by Kumagai et al. [13]. The degree of hydrolysis was expressed as the proportion of nitrogen contained in the 0.6 M trichloroacetic acid-soluble fraction to the total nitrogen content.

Animals
Seven-week-old female BALB/c mice were purchased from Japan SLC, Inc., (Shizuoka, Japan). All animal experiments were performed following the Guidelines for Animal Experiments of the College of Bioresource Sciences of Nihon University (approval number: AP14B003, AP15B017, and AP15B028), which meet the ethical guidelines for experimental animals in Japan. Water was made available from bottles, and a commercial mouse diet (CLEA Rodent Diet CE-2, CLEA Japan, Inc., Tokyo, Japan) was provided ad libitum.

Transdermal Sensitization
Transdermal sensitization was performed according to the method proposed by Adachi et al. [22] As illustrated in Figure 1, the mice were divided into five groups (n = 8): the distilled water (DW) group was patched with 80 µL of DW containing 0.5% (w/v) SDS to be unsensitized; the UG, DG, HG, and HDG groups were sensitized with 800 µg of the corresponding gliadin in 80 µL of DW containing 0.5% (w/v) SDS. The stratum corneum of the dorsal shaved skin of mice was removed by tape-stripping, repeated 20 times before sensitization. The gliadin suspension was applied to the shaved area of each mouse with a patch (CARELEAVESTM, Nichiban Co., Ltd., Tokyo, Japan), and the patched area was Foods 2020, 9,635 4 of 14 covered with surgical tape. The mice were exposed to each antigen for 3 days and were then rested for the next 4 days by removing the patch. Transdermal exposure to antigen was repeated for four cycles.
Foods 2020, 9, x FOR PEER REVIEW 4 of 14 0.5% (w/v) SDS. The stratum corneum of the dorsal shaved skin of mice was removed by tapestripping, repeated 20 times before sensitization. The gliadin suspension was applied to the shaved area of each mouse with a patch (CARELEAVESTM, Nichiban Co., Ltd., Tokyo, Japan), and the patched area was covered with surgical tape. The mice were exposed to each antigen for 3 days and were then rested for the next 4 days by removing the patch. Transdermal exposure to antigen was repeated for four cycles.

Systemic Anaphylaxis Induction
The transdermally sensitized mice were intraperitoneally challenged with 1 mg of antigen in 100 μL of DW 30 min on the 28th day after the final sensitization cycle.
The rectal temperature of mice was measured with a rectal thermometer (SK-1260, Sato Keiryoki Mfg Co., Ltd., Tokyo, Japan) 30 min after intraperitoneal challenge.
The anaphylactic response to the antigen-challenged mice was evaluated by scoring symptoms at the same time. The anaphylactic scores were categorized into the following six levels: Level 0, no symptoms; level 1, scratching and rubbing around the snout and head; level 2, showing puffiness around the eyes and snout, pilar erection, reduced activity or standing still with an increased respiratory rate; level 3, motionless for more than 1 min, transient convulsion, wheezing, labored respiration, or lying prone on stomach; level 4, no response to whisker stimuli, slight or no activity after prodding, tremor, convulsion, or loss of consciousness; level 5, death.

Measurement of Serum Histamine and Gliadin Specific IgE and IgG1 Levels
The blood was collected from the postcaval vein under anesthesia after the measurement of rectal temperature and anaphylactic scores. The blood samples for the measurement of gliadin specific IgE and IgG1 levels were centrifuged at 1300×g and 20 °C for 15 min to give the sera, which were stored at −80 °C until assay. The levels of gliadin-specific IgE and IgG1 in the sera were determined as described previously [17]. For the measurement of histamine levels, blood from individual mouse was collected into 0.5-mL heparinized tubes. The amount of histamine in the plasma was measured by high-performance liquid chromatography according to the on-column conversion method by using o-phthalaldehyde according to the manufacturer's instruction.

Systemic Anaphylaxis Induction
The transdermally sensitized mice were intraperitoneally challenged with 1 mg of antigen in 100 µL of DW 30 min on the 28th day after the final sensitization cycle.
The rectal temperature of mice was measured with a rectal thermometer (SK-1260, Sato Keiryoki Mfg Co., Ltd., Tokyo, Japan) 30 min after intraperitoneal challenge.
The anaphylactic response to the antigen-challenged mice was evaluated by scoring symptoms at the same time. The anaphylactic scores were categorized into the following six levels: Level 0, no symptoms; level 1, scratching and rubbing around the snout and head; level 2, showing puffiness around the eyes and snout, pilar erection, reduced activity or standing still with an increased respiratory rate; level 3, motionless for more than 1 min, transient convulsion, wheezing, labored respiration, or lying prone on stomach; level 4, no response to whisker stimuli, slight or no activity after prodding, tremor, convulsion, or loss of consciousness; level 5, death.

Measurement of Serum Histamine and Gliadin Specific IgE and IgG1 Levels
The blood was collected from the postcaval vein under anesthesia after the measurement of rectal temperature and anaphylactic scores. The blood samples for the measurement of gliadin specific IgE and IgG1 levels were centrifuged at 1300× g and 20 • C for 15 min to give the sera, which were stored at −80 • C until assay. The levels of gliadin-specific IgE and IgG1 in the sera were determined as described previously [17]. For the measurement of histamine levels, blood from individual mouse was collected into 0.5-mL heparinized tubes. The amount of histamine in the plasma was measured by high-performance liquid chromatography according to the on-column conversion method by using o-phthalaldehyde according to the manufacturer's instruction.

Measurement of Mcpt8 Expression Level
Skin biopsy specimens of the sensitized site were obtained after blood collection. To extract RNA and measure Mcpt8 expression level, the specimens were homogenized in Trizol (GIBCO BRL, Grand Island, NY, USA) using a Polytron (NS-52K, Microtec Co., Ltd., Chiba, Japan) and RNA was reverse transcribed into cDNA using a thermal cycler (MJ Research, Watertown, MA, USA).

In Vitro Skin Sensitization Test
To evaluate the immunological response of skin macrophage to HG and HDG, the human cell-line activation test (h-CLAT) was conducted according to the method described by Ashikage et al. [36]. Briefly, THP-1 cells from ATCC (Manassas, VA, USA) were cultured in RPMI-1640 (Sigma-Aldrich, Dorset, UK) supplemented with 10% fetal bovine serum, 100 units/mL penicillin, and 100 µg/mL streptomycin (Wako Pure Chemical) at 37 • C under 5% CO 2 in 12-well plates (1.0 × 10 6 cells/mL/well) with 1.0 mg/mL HG or HDG for 18 h. In the control group, HG and HDG were not added. Thereafter, the cells were washed twice with phosphate-buffered saline containing 0.1% bovine serum albumin. Next, total RNA was extracted from the cells.

Statistical Analysis
Statistical analyses were performed using the Mac Toukeikaiseki Version 3.0 software package (Esumi, Co., Ltd., Tokyo, Japan). All data were expressed as mean ± SE, and the significance of the differences (p values) among groups was evaluated by one-way ANOVA followed by a Tukey-Kramer test.

Degrees of Deamidation and/or Peptide-Bond Hydrolysis of Wheat Gliadins
The degrees of deamidation of UG, DG, HG, and HDG were 0%, 26%, 0%, and 39%, respectively, while those of peptide-bond hydrolysis of UG, DG, HG, and HDG were 0%, <1%, 26%, and 13%, respectively ( Figure 2). Figure 3 shows the apparent molecular weight distributions for each gliadin, measured by the ultrafiltration method. The abundance ratios of peptides with molecular weights lower than 10 kDa in UG, DG, HG, and HDG were 0%, <1%, 42%, and 9%, respectively. This distribution was consistent with the order of degree of peptide-bond hydrolysis of gliadins. The abundance ratios of peptides with molecular weight higher than 100 kDa in UG, DG, HG, and HDG were 44%, 45%, 0%, and 31%, respectively, which was consistent with the inverse order of degree of peptide-bond hydrolysis of gliadins. HDG possessed both high-(>100 kDa) and low-(<10 kDa) molecular-weight peptides and the range of molecular weight was widely distributed.   Figure 4 shows the change in rectal temperature of mice 30 min after intraperitoneal challenge. The average rectal temperature of HDG-sensitized mice decreased by approximately 4.2 °C for 30 min, while the average rectal temperature of UG-, DG-and HG-sensitized mice was at the same level as that of unsensitized mice (DW). The rectal temperature of HDG-sensitized mice at 30 min was significantly lower than that of the other groups at the same time point (p < 0.01). After four transdermal sensitizations of each gliadin, immediate hypersensitivity reactions were examined by intraperitoneal challenge of the same antigens used for the sensitization. Figure 5 shows the anaphylactic scores of these mice. The DW group showed no symptoms, with an average score of 0, while DG-and HG-sensitized mice presented moderate symptoms, with an average score of 1. Severe symptoms were observed in HDG-sensitized mice, with an average score of 4. The average score of the HDG group was significantly higher than that of the other groups (p < 0.05). Figure 6 shows the plasma histamine levels of mice after intraperitoneal challenge with each gliadin. The mean value of    Figure 4 shows the change in rectal temperature of mice 30 min after intraperitoneal challenge. The average rectal temperature of HDG-sensitized mice decreased by approximately 4.2 °C for 30 min, while the average rectal temperature of UG-, DG-and HG-sensitized mice was at the same level as that of unsensitized mice (DW). The rectal temperature of HDG-sensitized mice at 30 min was significantly lower than that of the other groups at the same time point (p < 0.01). After four transdermal sensitizations of each gliadin, immediate hypersensitivity reactions were examined by intraperitoneal challenge of the same antigens used for the sensitization. Figure 5 shows the anaphylactic scores of these mice. The DW group showed no symptoms, with an average score of 0, while DG-and HG-sensitized mice presented moderate symptoms, with an average score of 1. Severe symptoms were observed in HDG-sensitized mice, with an average score of 4. The average score of the HDG group was significantly higher than that of the other groups (p < 0.05). Figure 6 shows the plasma histamine levels of mice after intraperitoneal challenge with each gliadin. The mean value of  Figure 4 shows the change in rectal temperature of mice 30 min after intraperitoneal challenge. The average rectal temperature of HDG-sensitized mice decreased by approximately 4.2 • C for 30 min, while the average rectal temperature of UG-, DG-and HG-sensitized mice was at the same level as that of unsensitized mice (DW). The rectal temperature of HDG-sensitized mice at 30 min was significantly lower than that of the other groups at the same time point (p < 0.01). After four transdermal sensitizations of each gliadin, immediate hypersensitivity reactions were examined by intraperitoneal challenge of the same antigens used for the sensitization. Figure 5 shows the anaphylactic scores of these mice. The DW group showed no symptoms, with an average score of 0, while DG-and HG-sensitized mice presented moderate symptoms, with an average score of 1. Severe symptoms were observed in HDG-sensitized mice, with an average score of 4. The average score of the HDG group was significantly higher than that of the other groups (p < 0.05). Figure 6 shows the plasma histamine levels of mice after intraperitoneal challenge with each gliadin. The mean value of plasma histamine level of HDG-sensitized mice was significantly higher than that of UG-, DG-and HG-sensitized mice and unsensitized mice (p < 0.05).

Cutaneous Sensitivity of Deamidated and/or Peptide-Bond-Hydrolyzed Gliadin
Foods 2020, 9, x FOR PEER REVIEW 7 of 14 plasma histamine level of HDG-sensitized mice was significantly higher than that of UG-, DG-and HG-sensitized mice and unsensitized mice (p < 0.05).   Foods 2020, 9, x FOR PEER REVIEW 7 of 14 plasma histamine level of HDG-sensitized mice was significantly higher than that of UG-, DG-and HG-sensitized mice and unsensitized mice (p < 0.05).

mRNA Level of Mcpt8 at the Skin Site
To investigate the migration of basophils at the skin site after antigen exposure, we examined the expression level of Mcpt8, a basophil maker. Figure 9 shows mRNA expression level of Mcpt8 at the skin site after final transdermal sensitization with UG, DG, HG and HDG. The Mcpt8 mRNA expression level increased in the HG and HDG groups but not in the other groups.  Figure 10 shows the CD54 and CD86 mRNA expression levels in the THP-1 cells after HDG or HG exposure in the h-CLAT method. The mRNA expression levels of CD54 in the HG and HDG group were 5.6-and 15.6-fold higher than that in the DW group, respectively. The mRNA expression level of CD86 in the HDG group was 1.8-fold higher than that in the DW group, while the mRNA expression level of CD86 in the HG group was not increased.

mRNA Level of Mcpt8 at the Skin Site
To investigate the migration of basophils at the skin site after antigen exposure, we examined the expression level of Mcpt8, a basophil maker. Figure 9 shows mRNA expression level of Mcpt8 at the skin site after final transdermal sensitization with UG, DG, HG and HDG. The Mcpt8 mRNA expression level increased in the HG and HDG groups but not in the other groups.

mRNA Level of Mcpt8 at the Skin Site
To investigate the migration of basophils at the skin site after antigen exposure, we examined the expression level of Mcpt8, a basophil maker. Figure 9 shows mRNA expression level of Mcpt8 at the skin site after final transdermal sensitization with UG, DG, HG and HDG. The Mcpt8 mRNA expression level increased in the HG and HDG groups but not in the other groups.  Figure 10 shows the CD54 and CD86 mRNA expression levels in the THP-1 cells after HDG or HG exposure in the h-CLAT method. The mRNA expression levels of CD54 in the HG and HDG group were 5.6-and 15.6-fold higher than that in the DW group, respectively. The mRNA expression level of CD86 in the HDG group was 1.8-fold higher than that in the DW group, while the mRNA expression level of CD86 in the HG group was not increased. 3.5. mRNA Levels of CD54 and CD86 in THP-1 Cells Figure 10 shows the CD54 and CD86 mRNA expression levels in the THP-1 cells after HDG or HG exposure in the h-CLAT method. The mRNA expression levels of CD54 in the HG and HDG group were 5.6-and 15.6-fold higher than that in the DW group, respectively. The mRNA expression level of CD86 in the HDG group was 1.8-fold higher than that in the DW group, while the mRNA expression level of CD86 in the HG group was not increased. Foods 2020, 9, x FOR PEER REVIEW 10 of 14 Figure 10. CD54 and CD86 mRNA expression levels in THP-1 cells after HDG or HG exposure in the h-CLAT method. The value with two asterisks is significantly different from that of the DW group at p < 0.01 by the Tukey test. Each value is the mean of four experiments, with SE shown as a vertical bar.

Discussion
HCl-treated wheat protein had increased cutaneous permeability and allergic potency. This treatment of proteins causes both peptide-bond hydrolysis and side-chain deamidation. Therefore, we focused on peptide-bond hydrolysis and side-chain deamidation of wheat gluten, which might be the key structural change to induce cutaneous sensitization. We prepared UG, DG, HG, and HDG from gluten to clarify the structural feature responsible for the cutaneous sensitization process. HCl treatment of gliadin afforded HDG with a 13% degree of peptide-bond hydrolysis and a 39% degree of side-chain deamidation. Enzyme and cation-exchange resins selectively hydrolyzed peptide bonds and side-chain amide bonds, respectively. Enzymatic treatment of gliadin produced HG with a 26% degree of peptide-bond hydrolysis and a 0% degree of side-chain deamidation, while cationexchange-resin treatment produced DG with a <1% degree of peptide-bond hydrolysis and a 26% degree of side-chain deamidation. These results indicate that deamidated-only gliadin, hydrolyzedonly gliadin, and hydrolyzed and deamidated gliadin could be prepared.
Using each gliadin, we evaluated the allergenicity after cutaneous immunization followed by intraperitoneal injection of gliadin. The anaphylactic score in the HDG group was about 3, which was significantly higher than that in the other groups (p < 0.05). Although the UG, DG, and HG groups showed slight anaphylactic symptoms after intraperitoneal injection of each gliadin, scoring about 1, no significant difference was found to the DW group. In addition, the rectal temperature of the DW, UG, DG, and HG groups was retained at almost the same level for 30 min, while that of the HDG group dropped by about 4.2 °C after 30 min, showing significant difference from the other groups (p < 0.01). The severe allergic responses observed in the HDG group were in good accordance with those reported in the literatures using HCl-treated wheat protein [37]. Therefore, the sharp contrast between the HDG and other groups regarding allergic reactions demonstrated that both peptidebond hydrolysis and side-chain deamidation are crucial for cutaneous sensitization. In other words, our results indicate that DG and HG have a lower risk of inducing allergic reactions after cutaneous immunization compared with HDG.
The severe allergic reactions observed only in the HDG group could be attributed to the high sensitizing potency of HDG. The plasma histamine, gliadin-specific IgE, and IgG1 levels in the HDG group were significantly higher than those in the other groups (p < 0.05), indicating that mice in the HDG group were sensitized with HDG and induced a Th2-immune response. However, the IgE and IgG1 levels specific to native gliadin in the UG, DG and HG groups were significantly lower than those in the HDG group and were at similar levels to those in the DW group. These results suggest that UG, DG and HG have low sensitizing abilities.

Discussion
HCl-treated wheat protein had increased cutaneous permeability and allergic potency. This treatment of proteins causes both peptide-bond hydrolysis and side-chain deamidation. Therefore, we focused on peptide-bond hydrolysis and side-chain deamidation of wheat gluten, which might be the key structural change to induce cutaneous sensitization. We prepared UG, DG, HG, and HDG from gluten to clarify the structural feature responsible for the cutaneous sensitization process. HCl treatment of gliadin afforded HDG with a 13% degree of peptide-bond hydrolysis and a 39% degree of side-chain deamidation. Enzyme and cation-exchange resins selectively hydrolyzed peptide bonds and side-chain amide bonds, respectively. Enzymatic treatment of gliadin produced HG with a 26% degree of peptide-bond hydrolysis and a 0% degree of side-chain deamidation, while cation-exchange-resin treatment produced DG with a <1% degree of peptide-bond hydrolysis and a 26% degree of side-chain deamidation. These results indicate that deamidated-only gliadin, hydrolyzed-only gliadin, and hydrolyzed and deamidated gliadin could be prepared.
Using each gliadin, we evaluated the allergenicity after cutaneous immunization followed by intraperitoneal injection of gliadin. The anaphylactic score in the HDG group was about 3, which was significantly higher than that in the other groups (p < 0.05). Although the UG, DG, and HG groups showed slight anaphylactic symptoms after intraperitoneal injection of each gliadin, scoring about 1, no significant difference was found to the DW group. In addition, the rectal temperature of the DW, UG, DG, and HG groups was retained at almost the same level for 30 min, while that of the HDG group dropped by about 4.2 • C after 30 min, showing significant difference from the other groups (p < 0.01). The severe allergic responses observed in the HDG group were in good accordance with those reported in the literatures using HCl-treated wheat protein [37]. Therefore, the sharp contrast between the HDG and other groups regarding allergic reactions demonstrated that both peptide-bond hydrolysis and side-chain deamidation are crucial for cutaneous sensitization. In other words, our results indicate that DG and HG have a lower risk of inducing allergic reactions after cutaneous immunization compared with HDG.
The severe allergic reactions observed only in the HDG group could be attributed to the high sensitizing potency of HDG. The plasma histamine, gliadin-specific IgE, and IgG1 levels in the HDG group were significantly higher than those in the other groups (p < 0.05), indicating that mice in the HDG group were sensitized with HDG and induced a Th2-immune response. However, the IgE and IgG1 levels specific to native gliadin in the UG, DG and HG groups were significantly lower than those in the HDG group and were at similar levels to those in the DW group. These results suggest that UG, DG and HG have low sensitizing abilities.
The difference in sensitizing abilities among the gliadins can be explained by the activities of antigen-presenting cells. The immune process for cutaneous sensitization begins with antigen presentation by cells such as dendritic cells and basophils. Dendritic cells such as Langerhans cells and basophils play important roles for Th2 induction and IgE production upon epicutaneous protein exposure [23]. Basophils lead to Th2 skewing, supporting dendritic-cell induction of the Th2 response upon antigen-protein exposure by trogocytosis of peptide-major histocompatibility class II complexes [25,38]. In addition, basophils directly stimulate T cells and promote IL-4 production in CD4 + T cells upon exposure to peptide antigens [25]. The enhancement of basophils in HWP-sensitized patients was reported to be induced by HWP administration [39]. Moreover, HCl-treated gliadin increased degradation by bone marrow-derived dendritic cells [40]. We evaluated the sensitizing property of each gliadin by measuring the activities of basophils and dendritic cells. The mRNA expression level of Mcpt8 under the skin was higher in the HG and HDG groups than in the DW, UG and DG groups immediately after intraperitoneal administration of each gliadin. This finding indicates that HG and HDG were recognized as antigens by basophils, but UG and DG were not recognized and the sensitizing abilities of UG and DG were lower than those of HG and HDG. Migration of basophils is not sufficient to explain the sensitization potency of HDG because HG and HDG induced migration of basophils. However, it is difficult to isolate dendritic cells from mouse skin without activating them. Therefore, we evaluated the activity of dendritic cells in vitro after the addition of HDG or HG to THP-1 cells by the h-CLAT method [41], measuring the mRNA levels of CD54 and 86. The mRNA levels of CD54 and 86 in the HDG group increased, but little or no increase was observed in the HG group. These results indicate that dual activation of dendritic cells and basophils is required for the highly sensitizing and allergic potency of HDG.
This study clarified that both peptide-bond hydrolysis and deamidation of gliadin are necessary to induce high cutaneous sensitization. Peptide-bond hydrolysis reduces the molecular weight of the allergen protein, increasing its permeability through the skin barrier. The invaded antigen is then captured by antigen-presenting cells. Different from HDG, HG activated basophils but rarely induced allergic reactions after cutaneous immunization followed by intraperitoneal injection of HG. Therefore, the glutamic acid residues in gliadin may play an important role in activating dendritic cells and inducing allergic reactions after permeation through skin.
Although we did not examine the crucial limit of the molecular size of gliadin for skin permeability using gliadins with different degrees of hydrolysis, we showed that gliadin has to be hydrolyzed to permeate through skin. Deamidation using cation-exchange resins does not hydrolyze peptide bonds, which means DG cannot permeate through skin. In addition, as shown in our previous paper, DG has high foaming capacity, the value of which is higher than that of egg white protein [15]. Therefore, DG could be used as a safe foaming agent both in foods and cosmetics.

Conclusions
We have revealed that both peptide-bond hydrolysis and side-chain deamidation of gliadin are necessary for cutaneous sensitization, indicating that both the reduction in molecular size and increase in glutamic-acid residues are crucial for activating antigen-presenting cells and inducing allergic reactions. Our study has also shown that deamidated gliadin with no peptide-bond hydrolysis (DG) is promising as an alternative material for cosmetics because deamidation improves physicochemical properties such as foamability and emulsifiability of gliadin and DG scarcely activates antigen-presenting cells and shows no allergic response after transdermal exposure.