Aminoazo dye-protein-adduct enhances inhibitory effect on digestibility and damages to Gastro-Duodenal-Hepatic axis

4-Dimethylaminoazobenzene (DAB, methyl yellow, or butter yellow), a human carcinogen, has been banned for use in foods since 1988. In 2014, DAB adulteration in Tofu occurred in Taiwan. We hypothesize that DAB can form [DAB•SBP]adduct adduct with soybean protein (SBP) which could damage Gastro-Duodenal-Hepatic axis. Sprague-Dawley rats gavage fed [DAB•SBP]adduct adduct revealed severely reduced body weight and damaged duodenum, liver, hepatic mitochondria, and spleen. Hepatic levels of glutathione and ATP were severely reduced. Serum GOT and GPT were substantially elevated. Analysis by the adsorption isotherm clearly revealed DAB formed very stable [DAB•SBP]adduct adduct at 1:1 molar ration (Phase A). The equilibrium constant of this colloidal adduct [DAB•SBP]adduct was KeqA = ∝, behaving as the most stable and toxic species. At higher protein concentration (Phase C) it formed conjugate [DAB×SBPgross]conjugate, with KeqC = 3.23×10−2 mg/mL, implicating a moderately strong adsorption. The in vitro pepsin digestibility test showed apparently reduced digestibility by 27% (by Ninhydrin assay) or 8% (by Bradford assay). Conclusively, this is the first report indicating that [DAB•SBP]adduct potentially is capable to damage the Gastro-Duodenal-Hepatic axis.


Introduction
Azo dyes are compounds characterized by their vivid colors and provide excellent coloring properties. Azo dyes account for approximately 60-70% of all dyes used in food, textile and leather industries [1] demonstrated when the azo dyes are reduced by the intestinal bacteria [2]. Due to the toxicity, carcinogenicity and potential mutagenicity of thus formed aromatic amines, the use of certain azo dyes as food, textile and leather colorants, and the exposure of consumers ingesting the colored foods or using the textile and leather colored with azo compounds causes a serious health concern [1]. Early in 1935, Sasaki and Yoshida [3] demonstrated that administration of o-aminoazotoluene provoked progressive hepatocyte hyperplasia resulting in the formation of adenomas that became malignant in later days depending on the administration history [3].
DAB, a fat-soluble industrial yellow dye formerly had been used as a food additive. In rats aminoazo dye (butter yellow) induced liver damage associated with an irreversible increase of serum immunoglobin levels. DAB induced hepatomegaly, hepatic and renal dysfunction, liver cirrhosis and cancer, bladder cancer or osteoclasia during embryogenesis in dogs, and aspermatogenesis in mice [4] contact dermatitis, cough, wheezing, dyspnea, bloody sputum, urinary frequency, haematuria, dysuria, possible occupational carcinogen. DAB is considered hazardous by the 2012 Occupational Safety and Health Administration (OSHA) Hazard Communication Standard (29 CFR 1910.1200) [5]. A diversity of mechanistic pathways has been uncovered for azo dyes. The carcinogenic activation of DAB was suggested to proceed via oxidative monodemethylation accompanied by N-oxidation, yielding the 4-hydroxy derivative of DAB, leading to the formation of a nitrenium ion which eventually attacks the position-8 of guanosine on DNA molecules [6]. Structural abnormalities of liver microsomes from rats fed 3'methyl-4-dimethylaminoazobenzene (3'Me-DAB), have been demonstrated by Arcos and Arcos [7]. Porter and Bruni [8] implicated that administration of (3'Me-DAB) could destroy organization of the rough endoplasmic reticulum and aggregation of the smooth form Porter and Bruni [8]. 3'Me-DAB forms t-RNA-azo dye complex in liver of rats, which is considered as the major target for this azo carcinogen [9]. Nonetheless, literature has indicated that the carcinogenicity of azo dyes varies with the species tested [10].
Since 1988 DAB has been banned and listed by the International Agency for Research on Cancer as a possible human carcinogen [4]. In the year 2009, European countries including Belgium and France detected the banned food colouring in Delhaize and Intermarche products [11,12]. Strikingly, on December 17, 2014, The Food and Drug Administration of Taiwan (FDAT) released a list of 36 food products adulterated with DAB soybean emulsifier supplied by Chien-Hsin Enterprise (Taiwan). The levels of DAB in these products ranged from 2.1 to 63.6 parts per billion (ppb). In reality, this DAB scandal was first set off by Hong Kong's Center for Food Safety who discovered the prohibited dye in prepackaged pepper-flavored hard Tofu curds manufactured by Te-Chang Food Co. (Taiwan).
Tacal and Ö zer [13] demonstrated the formation of dye-protein adducts when highly electrophilic cationic dyes are quantitatively trapped in biological fluids and that the primary in vivo effects (e.g. toxicity) of such dyes most likely arise from ligand-type effects on multiple protein targets [13]. Furthermore, Saeed et al. reported the digestibility of dye-protein adduct can be reduced depending on the adduct characteristics [14].
By then, no sooner several colleagues had showed severe vomiting and gastritis after having ingested the yellow dye-colored bamboo shoot jam (Yi-Lan brand, Taiwan) for 12 h, than the motivation that inspired us to start with this experimentation. It was hypothesized that the adulteration of DAB into Tofu in vitro could form dye-protein adducts which may retard the digestibility of protein and simultaneously damage the gastroenteric system when ingested per os and it was supposed that 2-3 week-chronic ingestion should be able to induce the same event in the animal model. To verify this, we conducted the in vitro pepsin digestibility study as well as the in vivo animal experiment using the DAB-soybean protein model.

Chemical and reagents
Pepsin was purchased from Sigma-Aldrich. The proteolytic activity of pepsin was ! 400U/mg protein. Methyl yellow (DAB) (95.0%) was purchased from Sigma-Aldrich (St. Louis, MO, USA). GOT and GPT ELISA kits were provided by Novatein Bioscience (Woburn, MA. USA).
Preparation of reference DAB curve: methyl yellow (Sigma Aldrich, St. Louis. MO, USA) was accurately measured and dissolved in 2 mL 0.1N HCl to make the stock solution. PBS (pH 3.2) was used to make dilutions to 10,20,30,40,50, and 60 μg/mL, respectively. The solutions were filtered through 0.45 μm Micropore and the absorbance was read at 430 nm to establish the standard curve.
Preparation of DAB digestibility test solution: DAB (11.265 mg) was accurately weighed, dissolved in 2mL 0.1 N HCl, and made volume to 10 mL with PBS (pH 3.2).
Preparation of soybean protein (SBP) solutions: SBP (ranging from 10 to 1000 mg, as indicated) were accurately weighed and transferred into 10 mL centrifuge tubes. To each tube 5 mL of PBS (pH 7.2) was added and agitated to facilitate the dissolution.
Preparation of DAB-SBP hydrocolloids DAB (11.265 mg) was accurately measured and dissolved in 10 mL PBS (pH 6.5) to make a stock solution of DAB (1.1265 mg/mL). For preparation of DAB-SBP hydrocolloid feed, to 5 mL of SBP solutions (40 mg SBP/mL) 300 μL of experimental DAB solution was added, agitated at 25˚C for 30 min and centrifuged for 3 h at 12,000×g. The supernatant was collected for use. For adsorption experiment, to 5 mL of SBP solutions at different concentrations (0 to 220 mg SBP/mL) 300 μL of experimental DAB solution (1.1265 mg/mL) was added, agitated at 25˚C for 30 min and centrifuged for 3 h at 12,000×g, the supernatant was read at 430 nm.

Animals
This proposal for animal experiment had been approved by the Ethic Committee For Care of Animals and Animal Experiment of Hungkuang University (The affidavit of approval of animal use protocol has been attached herewith) (Supplement S-1). Sprague Dawley rats, age 4 weeks and weighing 140-150g, were purchased from LASCO Corporation (Taiwan). These animals were admitted and acclimated for the first two weeks in the animal room with good ventilation, controlled at RH 70-75%, 22±2˚C, and 12/12h day/night cycle. The access to the water and the regular chow was ad libitum.

Animal treatment
To mimic the toxicity of DAB enroute Tofu-ingestion, a dose of DAB at 281.6 ppb was estimated for gavage-fed rats having body weight around 400 g, which approximately corresponded to 4.5 folds of the maximum level of DAB (63.6 ppb) ever adulterated in Tofu. In brief, at the beginning of week 3, the rats were divided into three groups, 8 in each, and 2 rats per cage. Group 1 serving the control group was gavage fed 100 μL PBS daily. Group 2 was gavage-fed 100 μL DAB-SBP hydrocolloid (DAB 1.1265 mg/mL% 281.6 ppb) once only at the beginning. And group 3 was gavage-fed same amount of DAB-SBP hydrocolloid daily for a period of 14 days. The animals were monitored by their appetite, running activity, response to light and dark, and body weight change during the experimental period. It seemed that there was no apparent adverse effects when treated with DAB. On day 14, the body weight of animal was taken, and then the animals were fasted 12 h before euthanized. After the animals were CO 2 -euthenized, the organs hearts, kidneys, livers and duodenum were excised and rinsed with cold PBS buffer several times. The adhering water drops were soaked off with the kitchen tissue. The organs were weighed and the ratio of organ to body weight was accessed.

Pathological change by Hematoxylin-Eosin staining
Part of the organs was subject to Hematoxylin-Eosin staining for histopathological examination at the Dr. J.-W. Liaw's Lab at the Department of Pathology, College of Veterans of Chung-Hsing University (Taichung, Taiwan).

Transmittance electron micrograph
After the organs were excised, rinsed and the adhering water was soaked off, part of the organs were immediately immersed in 20% formalin. The samples were delivered to the Electron Microscope Division of Chung-Shing University (Taichung, Taiwan) to carry out the TEM observation.

Hepatic glutathione analysis
Part of hepatic lobe from individual rats were analyzed for reduced and total glutathione. The method was based on the determination of thiols by high-performance liquid chromatography with fluorescence detection [15]. Briefly, one mg of liver was placed in a test tube with 1 ml of 5% trichloroacetic acid containing 5mM EDTA, homogenized, vortexed, and centrifuged. Next, 50 μL of homogenate was transferred to a test tube with 950 μL of 5 mM EDTA containing 0.1 M borax (pH 9.3). Then, 100 μL aliquots were placed in duplicate tubes for each liver sample. Fluorobenzofurazan-4-sulfonic acid ammonium salt was added for reduced glutathione and tri-n-butylphosphine in acetonitrile was added to the other tube for total glutathione. High performance liquid chromatography with fluorescence detection was used for determining total and reduced glutathione and compared against calibration standards that were processed with each assay.

Measurement of liver ATP by spectrophotometry
Liver ATP concentration was assayed by spectrophotometry using an assay kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China). The absorbance at 636nm was recorded and the concentration of ATP was expressed as millimoles per gram tissue (mmol/g tissue) [16].

GOT and GPT assays
GOT and GPT were assayed according to the protocol instructed by the manufacturer (Novatein Bioscience, Woburn, MA. USA).

In vitro pepsin digestibility test with DAB-SBP adduct
In vitro pepsin digestibility test was conducted by following Malinauskyte et al. [17], Sarkar et al. [18], and Sánchez-Chiang et al. [19]. As stomach is the front line digestive organ, only the pepsin digestibility test was performed to explore the effect of DAB on the digestibility of the soybean protein. The composition of the simulated gastric fluid (SGF) was as follows: 2.0 g NaCl, 6.9 g pepsin (601 U/mg), 7 mL conc. HCl, diluted to 1 L and pH adjusted to 1.2 using 1.0 M HCl. SGF was prepared according to procedure of Sarkar et al. [18]. The proteolytic activity of SGF was 36.6±0.8 U/mL [measured according to Sánchez-Chiang et al. [19].
In brief, judging from the adsorption isotherm, DAB digestibility test solution was added and thoroughly agitated for 30 min. The mixture was centrifuged at 12000×g for 3 h. The supernatant was discarded and the sediment was dissolved in 1 mL 2% KOH with agitation. The pH was adjusted with 1 M HCl to 1.2. SGF (2 mL) was added and incubated at 37˚C for 30, 60, 90, 120, and 150 min respectively. The digestion reaction was terminated immediately at the end point by addition of 100 μL 2% KOH. The solution was centrifuged at 6000×g for 20 min. An aliquot of 100 μL of the supernatant was measured and subjected to amino acid ninhydrin analysis and Bradford protein assay. A control test was similarly performed using the same amount of SBP without DAB.

Statistical analysis
Data obtained in the same group were analyzed using the SPSS 10.0 statistical software (SPSS, Chicago, IL, USA) and expressed in mean±S.D. An analysis of variance (ANOVAs) and Tukey's test were used to analyze the variance and identify the significant differences between paired means at p < 0.05.

Body/Organ weight affected by DAB-SBP hydrocolloids
No apparent morphological and behavioral changes were found during the experimental period. However prominent body weight change was observed. The body weight of group 2 rats (DAB-treated for one day) revealed to be the heaviest, reaching a body weight of 444.3 ±6.0g (p<0.05). The group 3 (DAB-treated for 2 weeks) rats was the lightest (395.8±5.0g), compared with 427.7±7.0g of the group 1 (the control) (p<0.05). A maximum deviation of 13.2% in body weight was found. The long term DAB-treated rats exhibited the heaviest livers (16.2 ±0.7g) comparing to 13.8±0.8g and 12.5±0.6g for the group 1 (control) and group 2 ( Table 1). Long term DAB-treated group also characteristically revealed the heaviest kidneys and the lightest heart, however, the difference was insignificant. Nonetheless, the group 3 (DAB-treated for 2 weeks) rats exhibited the largest ratio of liver to body weight (0.041) and kidney to body weight (0.0088) (p<0.05), implicating DAB could have damaged these two organs (Table 1). In contrast, the heart weight and the ratio of heart to body weight did not show any significant changes (Table 1). Evidence has accumulated since the initial observations of Miller and Miller [20] suggesting that covalent bonding of aminoazo dyes to rat liver protein is essential for this group of carcinogens to manifest their activity [20]. Suggestively, DAB at the very early stage of feeding tended to form various conjugates with different hepatic enzymes, resulting in reduced body weight and damaged organs.

DAB-protein adduct severely damaged the duodenum, but only moderately injured livers
Normal duodenum appeared denser and exhibited brilliant transparency (Fig 1). The one-day DAB-treated duodenum seemed normal or only slightly damaged (Fig 1). As contrast, the duodenum of rats treated with DAB for two weeks was apparently smaller in circumference with pale outer look. The lumen of duodenum was totally destroyed leaving a larger cavity (Fig 1).

Hematoxylin-Eosin staining revealed injured duodenum and liver tissues
The duodenum of control group with normal elongated villi can be clearly observed. The villous to crypt ratio was around ! 3:1, which is the regular size of normal duodenum (Fig 1). As contrast, the duodenum of DAB-administered rats were noted with blunted and discrete villi, atrophy of central lacteal, rupture of columnar epithelium and much of oligomucous cells full of viscous fluid, and some of the villi were affiliated with pathological increase of intraepithelial lymphocytes two weeks after DAB treatment, implicating that the mitotic capability was disappearing and the cells were transforming into goblet cells (Fig 2B) (H&E, 400×). Previously, similar finding had been reported by Orr et al. [21].
Histopathological examination by HE stain indicated that the livers of control group were all normal (Fig 2C). DAB enlarged and congested central venules and caused a bundle of swollen sinusoids and the interhepatic plate spaces. The cavities fully filled up with plasma after damaged by DAB implicated a state of mild inflammatory (Fig 2D).
Literature previously in detail cited by Opie [22] demonstrated similar pathogenesis in animals having received DAB, the earliest changes suggested cirrhosis occurring about hepatic veins and their smallest branches. Disintegration and disappearance of liver cell columns is associated with the accumulation of mononuclear cells [22].

TEM micrograph
The TEM image showed DAB after administered for two weeks severely damaged the mitochondria. As seen, the inner membrane structure, in particular, the cristae were all apparently destroyed (Fig 3).

Hepatic glutathione level
Hepatic level of reduced glutathione (GSH) was severely reduced due by DAB administration for two weeks ( Table 2). The level of GSH was reduced from 5.30±0.80 to 1.65±0.61 μmol/g (p<0.01). Substantial reduction of GSSG was also observed (p<0.05). On the other hands, levels of GOT and GPT were highly upregulated, reaching 188±26 and 56±14 μmole/μL (p<0.01), respectively comparing to the control ( Table 2). Glutathione has multiple functions ranging from anti-oxidant defense to modulation of a diverse immune function and many physiological and biochemical conditions are related to low glutathione levels [23,24]. Hepatic ATP level ATP was severely reduced due to DAB administration, correspondingly the ATP level was reduced from 2.85±0.40 to 0.41±0.22 μmol/g (p<0.01) ( Table 2), compared to the cited normal (fed) value 3.21±0.2 μmol/g [25].
From the results shown in Fig 3 and Table 2, it is evident that mitochondrial structure was severely damaged by ingesting DAB for two weeks, in particular, the inner membrane cristae as well as the intermembrane space (lower panel in Fig 3B, 30,000×). The structure of inner membrane is highly complex, including all of the complexes of the electron transport system for oxidative phosphorylation, the ATP synthetase complex and transport proteins, and is freely permeable only to oxygen, carbon dioxide, and water [26]. As found, the intact structure of outer membrane totally disappeared when treated with DAB ( Fig 3B, lower panel), suggesting that DAB could have completely destroyed both the transport and oxidative phosphorylation systems. Complex IV contains 13 different subunits encoded by both mitochondrial DNA and nuclear DNA. It receives an electron from each of the four cytochrome c molecules, and transfers it to one oxygen molecule, converting it into two molecules of water. In this process, it also binds to four proton molecules and translocates them across the membrane to establish electrochemical gradient, which is utilized for the synthesis of ATP [27].

Hepatic GOT GPT levels
On the other hands, levels of GOT and GPT were highly upregulated after treating with DAB, reaching 188±26 and 56±14 μmole/μL (p<0.01) comparing to the controls 75±15 and 25  ±11 μmole/μL, respectively (Table 2). Obviously these levels may be higher for humans, but it corresponds well to data for rats [28].
[DAB×SBP gross ] conjugate severely reduced pepsin digestibility The pepsin digestibility on soybean protein was substantially suppressed by the adduct [DAB• SBP] adduct and the DAB-SBP adsorption conjugate, [DAB×SBP gross ] conjugate . As seen, the ninhydrin assay revealed apparently the suppressed amino acid release from SBP before 90 min, but finally leveled off after 90 min and a 27.0% reduction of pepsin digestibility was observed at 120 min (Fig 4A). On the other hand, the Bradford protein assay also revealed a delayed digestibility by 8.0% at the initial phase at 60 min (Fig 4B), implicating the cleavage site on the SBP to produce larger peptide fragments had been completely blocked by DAB due to the formation of the [DAB×SBP gross ] conjugate conjugate. Results have clearly implied a solid evidence supporting strongly our hypothesis that the adduct [DAB•SBP] adduct and the DAB-SBP adsorption conjugate, [DAB×SBP gross ] conjugate , tend to enhance the injury to the gastroduodenal system. As already mentioned, the digestibility of dye-protein adduct can be reduced depending on the adduct characteristics [14].
We suggest firstly that DAB being a cationic dyestuff forms ion-pairs between the cationic dyestuff and an negatively-charged surface ions on biomacromolecules like SBP resulting in formation of an electroactive liquid membrane which retards the transport of ions and signals. Similar theoretical basis has been described for ion selective electrode fabrication [29]. Moreover, the surface of pepsin is more negatively charged [30] which would facilitate the adduct formation. Secondly, a coordination complex could occur between DAB and pepsin active site. We suggest that the well oriented [DAB•SBP] adduct adduct would fit better on the active site of pepsin (Fig 4C) than the DAB-SBP conjugates that are usually formed by random conjugation. Saeed et al. [14] reported the digestibility of dye-protein adduct can be reduced depending on the adduct characteristics [14].

Correlation of the atypical adsorption isotherm with the toxicity
The experimentally acquired physicochemical parameters for each phase are listed in Table 3. (As for the mathematical deduction, please be referred to Section A1 in S1 Appendix).
As already mentioned, cationic dyes forms dye-protein adducts in biological fluids. Its primary in vivo toxicity arises from adduct or complex type associated with protein targets [13,14], while the digestibility of dye-protein adduct can be reduced depending on the adduct characteristics [14].

Conclusions
The toxicity of DAB can be enhanced when forming adduct with the soybean protein (SBP) (i.e. DAB-SBP). The adsorption behavior of DAB onto SBP reveals a three phase adsorption isotherm. In phase A, the soluble hydrocolloid [DAB-SBP] produced by 1:1 molar ratio is suggested to be the most toxic as evidenced by the in vitro pepsin digestibility test and the in vivo animal experiment with Sprague-Dawley rats. The main organs affected by [DAB-SBP] are duodenum, liver, kidneys, and mitochondria. The typical Langmuir adsorption can only be perceived in Phase C (SBP !117mg/mL). The equilibrium constant of Phase A is infinitely large, implicating formation of extremely stable [DAB-SBP] adduct, while Phase C reveals to be only moderately strong adsorption. To our knowledge, this is the first report demonstrating the augmented Gastro-Duodenal-Hepatic toxicity of DAB when forming adduct with soybean protein.