Assessment of synthetic food dye erythrosine induced cytotoxicity, genotoxicity, biochemical and molecular alterations in Allium cepa root meristematic cells: insights from in silico study

Abstract

Background

Synthetic food dyes are being exponentially used in food products and scarce studies regarding their toxicities and safety raise concern. Erythrosine is one of the synthetic food dyes being used in jams, fig, pineapple marmalades, dairy products, soft drinks, pickles, relishes, smoked fish, cheese, ketchup, maraschino cherries and a variety of other foods.

Methodology

In this study the cyto-genotoxic effect of erythrosine was evaluated, using root meristematic cells of Allium cepa for the cellular and molecular alternations at concentrations 0.1, 0.25, 0.5 and 1 mg/mL.

Results

The results revealed a significant decrease of 57.81% in the mitotic index after 96 h at the 0.1 mg/mL concentration. In biochemical analysis, the malondialdehyde content increased significantly (5.47-fold), while proline content, catalase activity and superoxide dismutase activity decreased gradually in a concentration-dependent manner showing a maximum decrease of 78.11%, 64.68% and 61.73% respectively at the highest concentration after 96 h duration. The comet assay revealed increased DNA damage with increasing concentration and attenuated total reflectance- Fourier transform infrared spectroscopy (ATR-FTIR) analysis showed significant alterations in biomolecules as indicated by multivariate analysis, i.e. Principal Component Analysis (PCA). Furthermore, molecular docking demonstrated a strong binding energy (G_best = −11.46 kcal/mol) and an inhibition constant (Ki) of 3.96 nM between erythrosine and the DNA minor groove.

Conclusion

The present study’s findings revealed the cytotoxic and genotoxic potential of erythrosine on A. cepa root cells. Further, the study also proposed the usefulness of A. cepa as a model system for studying the toxicity of food additives.

Highlights

• Erythrosine showed prominent cytotoxicity with complete inhibition of mitosis at 0.25 mg/mL conc.

• High level of genotoxicity revealed by 3.68-fold increment in chromosomal aberrations (CAs) frequency at lowest concentration (0.1 mg/mL).

• The increased MDA level, reduced CAT, SOD activity and proline content showed significant biochemical changes in antioxidant system.

• Alternations in biomolecules viz. protein, lipid and nucleic acid region showed by ATR-FITR spectrum and confirmed statistically by multivariate analysis of spectrum.

• In silico model supports the finding as the lower binding free energy of −11.46 kcal/mol between erythrosine and DNA structure was calculated.

Graphical Abstract

Graphical Abstract.

Introdution

Colour is an essential factor that influences customer preferences for the rapid identification and ultimate acceptance of food.¹ Synthetic food dyes were added to food products to improve their visual appeal. Before the 1950s, the dyes of plant and vegetable sources were used to colour food products, but the boom in the chemical industry caused the inclusion of petroleum-based synthetic compounds in food dyes.² Since then, the use of synthetic dyes has increased exponentially due to advantages such as lower prices, longer shelf lives and their high stability to light, air, pH, and microbial contamination.³ Many countries have established their own rules and regulations regarding the use of synthetic food dyes, and toxicological studies have led to the prohibition of many food dyes.⁴ Despite this, the use of synthetic dyes in foods is still controversial throughout the world because, in literature, there are limited studies that have reported the toxic effects of these compounds. Even the intake of synthetic food dyes is found to exceed their prescribed limit in children, as they are the primary consumers of coloured food.^5–7

Erythrosine (C.I. Food Red 14, C.I. 45430, 1956, FD&C Red No. 3, E127), a synthetic red dye, is widely used in food products and has been under scrutiny due to conflicting reports on its safety. It is a disodium or dipotassium salt of 2,4,5,7-tetraiodofluorescein (C₂₀H₆I₄Na₂O₅) that is synthesised from fluorescein and belongs to the xanthene class. Table 1 and Fig. 1 provide the molecular structure and general properties of erythrosine. Previous studies have shown varying degrees of cytotoxic and genotoxic effects, necessitating further investigation. Previous investigations have reported that erythrosine can induce allergic reactions in the eyes and skin, severe headaches and nausea,⁸ nephrotoxicity,⁹ hepatocellular carcinoma (HCC),¹⁰ developmental toxicity,¹¹ and teratogenic effects.¹² However, some other investigations concluded that this food dye was safe to consume since no negative effects were observed in the experimental models.^13–17

Table 1.

General properties of erythrosine.

Class	Xanthene
Molecular weight	879.86
Molecular formula	C20H6I4Na2O5
C.I. No.	45,430
IUPAC Name	disodium;2′,4′,5′,7′-tetraiodo-3-oxospiro[2-benzofuran-1,9′-xanthene]-3′,6′-diolate
Synonyms	E127 FD and C Red No.3 C.I. Food Red 14 LB-Rot-I

Fig. 1.

Chemical formula of erythrosine.

A variety of plant, animal, and microbial assays were used as bioindicators for the risk assessment. Among them, the A. cepa test¹⁸ is a widespread approach for determining the toxicity of various compounds and was recognised by the US Environmental Protection Agency (USEPA) and the World Health Organisation (WHO).^19,20 This bioassay features a high percentage of dividing cells with a lower number of large monocentric chromosomes (2n = 16) that stain well. These characteristics facilitate the detection of mitotic spindle alterations and chromosomal breakage.¹⁸ It offers advantages such as feasibility, low cost, high sensitivity and comparable results to various high-sensitivity tests. Additionally, it demonstrates a direct correlation with tests conducted in prokaryotes, lymphocytes, Ames, and Artemia salina L., as well as in vivo acute oral toxicity tests.^21,22 Yet A. cepa assay has some potential variability, particularly in terms of onion bulb sources and environmental conditions that could affect the results. This variability can be minimised by using a single onion variety for the entire study and providing stable external climate.²³ Nevertheless, the results can serve as a preliminary guide for other testing systems, highlighting the crucial need for a thorough and precise assessment of a compound’s toxicity effects.^23,24

Therefore, the A. cepa root assay was chosen as a model for toxicity assessment of synthetic food dye erythrosine. This study aims to bridge the research gap on the cytotoxic properties of erythrosine through A. cepa cells and studies explaining the structural mechanistic of erythrosine effects, investigating its binding mode and affinity to the DNA molecule using in silico approach. As well as the chemical and structural changes at the molecular level caused by erythrosine using attenuated total reflectance-Fourier transform infrared (ATR-FTIR). A logarithmic scaling approach with factor of 2–2.5× increments was used for concentration ranges because biological systems often exhibit non-linear responses to increasing chemical exposure.²⁵ For the preliminary study, a range of erythrosine concentrations (1, 2.5, 5, 7.5, and 10 mg/mL) were tested based on literature.^12,26–28 The results revealed complete inhibition of mitosis at 1 mg/mL. Therefore, the lower concentrations (0.1, 0.25, 0.5, and 1 mg/mL) were selected for further experimentation to explore sub-lethal and intermediate responses.

The cytotoxic potential of erythrosine was evidenced by a significant reduction in the mitotic index (MI), while genotoxic effects were demonstrated through increased chromosomal aberrations and DNA damage through comet assay. Oxidative stress was determined by the changes in catalase activity, superoxide dismutase activity, malonaldehyde content, and proline content. These are useful biomarkers for estimating injury due to the production of free radicals in the biological system and alterations in antioxidant defence mechanisms.^29,30 The combination of ATR-FTIR spectroscopy and molecular docking offers an efficient and complementary approach for investigating the potential genotoxic effects of synthetic food colors, such as erythrosine. ATR-FTIR spectroscopy utilizes the interaction between the chemical bonds of molecules present in a given sample with the infrared (IR) radiation and depicts an infrared spectrum which is a fingerprint of the analysed sample. This technique reveals fundamental vibrations of chemical groups in biomolecules, such as lipids, proteins, carbohydrates, and nucleic acids, allowing researchers to observe and document biochemical changes at the molecular level.³¹ On the other hand, molecular docking employs computational methods to model and visualize the interactions between molecules at the atomic level. This in silico approach predicts how molecules bind to DNA, helping to confirm their potential genotoxic effects by simulating binding modes and strengths with a high degree of reliability and low false-positive rates that can be correlated with the structural characteristics of a molecule to its in vitro and in vivo genotoxicity and cytotoxicity.^22,32,33 This synergy between ATR-FTIR and molecular docking not only validates the findings but also demonstrates the utility of combining them for a deeper understanding of toxicological effects and offer a solid foundation for further research and highlight the importance of a thorough and accurate assessment of a compound’s toxicity.

Materials and methods

Chemicals

Erythrosine (CAS No. 16423-68-0) was procured from HiMedia Laboratories Private Limited, India. The aceto-orcein (Product code. 67418) was obtained from SRL Laboratories Private Limited, India. The rest of the reagents used in this study were of analytical grade.

Test model and treatments

The A. cepa (2n = 16) (onion bulbs) of similar diameter (25–30 mm), single variety and without any treatment were procured from a local market. Before treatment, all the dry roots and scales were removed with the help of a knife without damaging the root’s primordia. The healthy onion bulbs were first grown in distilled water at room temperature until root length reached approx. 2.0 cm. Then, the 5 onion bulbs in each group were placed in different concentrations of dye solutions dissolved in distilled water as that only the roots were immersed in the solution. For the control group, some onion bulbs were kept in distilled water. In order to find out the concentrations for experimentation preliminary study was performed with different concentrations viz. 1, 2.5, 5, 7.5 and 10 mg/mL of erythrosine. The complete inhibition of mitosis was observed at 1 mg/mL, therefore, the lower concentrations i.e. the 0.1, 0.25, 0.5 and 1 mg/mL concentrations were decided for exposure experiment.

Determination of cytotoxicity and genotoxicity

For determination of mitotic index (MI) and chromosomal aberrations (CAs), after the exposure period i.e. 24 h, 48 h, 72 h and 96 h, fixation was done in Carnoy’s solution 3:1 (ethanol/ acetic acid) for 24 h. The root tips were rinsed twice with distilled water after fixation and some were preserved in 70% ethanol in the refrigerator. Subsequently, random root tips were selected and hydrolysed in 1 N HCl at 60 °C in a water bath for 8–10 min and then washed with distilled water. The aceto-orcein stain for 5 min was used for staining hydrolysed root tips and slides were prepared with squash technique. Through microscopic observation, the mitotic index (MI) and percentage of chromosomal aberrations (CAs) were determined. MI was expressed in percentage by counting cells per slide in triplicate (approx. 2,000 cells) for each control and treated group for cytotoxicity evaluation. The cells in prophase, metaphase, anaphase and telophase were considered as dividing cells. The CAs were noted as; stickiness, abnormal metaphase, vagrant and laggard chromosomes, chromosome breakage, chromosome bridges, delayed anaphase or telophase, etc. Photographs were taken for further study. The total number of cells, number of diving cells and number of aberrant cells were counted manually. The MI and percent CAs were calculated using Eq. 1 and Eq. 2, respectively.

The mitotic index (MI) was calculated as:

(1)

Percent chromosomal aberrations were calculated as:

(2)

Determination of DNA damage by alkaline comet assay

The DNA damage in root tip cells of A. cepa after treatment was determined by comet assay. The conditions were kept the same as described and three bulbs from each control and treatment group were used. The root cells nuclei were isolated by pouring 600 μL ice-cold nuclear isolation buffer (pH 7.5) consisting of 4 mM MgCl₂-6H₂O, 0.5% w/v Triton X-100, 200 mM Tris on a petri plate placed over ice. The root tips were sliced gently with a fresh razor blade. The petri plate was kept slanted for 2–4 min and runaway nuclear suspension was collected from the bottom. The nuclear suspension was centrifuged at 1,200 rpm for 7 min at 4 °C and 50 μL nuclear suspension was mixed with 0.5% LMPA in phosphate buffer saline (pH 7.4). Then the mixture was pipetted on the slides pre-coated with 1% NMPA and kept overnight at 37 °C. The 2^nd layer of gel was allowed to solidify for 8–10 min on ice. After that, coverslips were removed and all the slides were submersed in a freshly prepared lysing solution with 1% Triton X-100 and 10% DMSO added fresh at 4 °C. The slides were kept in lysing solution at least for 2 h and then immersed in a chilled and freshly prepared electrophoresis buffer (pH > 13, 1 mM EDTA, and 0.3 M NaOH) for 20 min before electrophoresis to allow unwinding of the DNA. The electrophoresis was run at 25 V and 0.3 A for 20 min followed by the neutralisation of the slides in Tris buffer (pH 7.5, 400 mM).³⁴ The slides were left undisturbed overnight in the dark for drying and stained with ethidium bromide for fluorescent observation. A fluorescent microscope (40×) with an Olympus 3× camera was used for photomicrography. CASP LAB is used for analysing random 50 comets per slide.

Biochemical parameters

The biochemical analysis was done in the A. cepa root homogenate with a spectrophotometer- Eppendorf BioSpectrometer®. Roots from 48 h and 96 h exposure groups were cut and immediately washed, weighed and homogenised in respective buffers. The icepacks were used to avoid denaturation of the enzyme during processing. The supernatant was cumulated for the assessment of biochemical studies.

Malondialdehyde (MDA) content

To determine the content of malondialdehyde (MDA), supernatant was prepared by homogenising root samples in 0.1% (w/v) trichloroacetic acid (TCA) and centrifugation at 5,000 rpm at 4 °C temperature. Then, 100 μL of supernatant and 600 μL of 20% TCA containing 0.5% thiobarbituric acid were mixed and heated at 95 °C for 30 min. The reaction was terminated on ice and allowed to cool. The centrifugation was then carried out at 10,000 rpm for 10 min. The absorbance was measured at 532 and 600 nm after the supernatant was separated.³⁵ The absorbance value at 600 nm due to unspecified turbidity was subtracted from the absorbance value at 532 nm. The extinction coefficient 155 mM cm⁻¹ was used.

Catalase (CAT)

The method of Aebi³⁶ was used to evaluate catalase activity. The roots were homogenised in potassium phosphate buffer (1:4) and centrifuged at 10,000 rpm for 10 min. The 25 μL of supernatant was mixed with 3 mL of 12.5 mM H₂O₂ prepared in 0.067 M sodium phosphate buffer (pH 7.0). The hydrogen peroxide is broken down into water and oxygen by the enzyme catalase which causes the absorption at 240 nm to decrease with time and was measured as activity of the enzyme.

Superoxide dismutase (SOD)

SOD activity was determined by using the method of Kono.³⁷ The inhibition of nitroblue tetrazolium (NBT) dye reduction by superoxide free radicals generated by hydroxylamine hydrochloride (HAH) autooxidation is the basis of this method. The reaction mixture comprised 50 mM sodium carbonate buffer (pH 10.8), 96 μm NBT, 0.6% (w/v) Triton-X-100 and 20 mM HAH and supernatant. The increase in absorbance due to the reduction of NBT was recorded at 540 nm. The activity of the enzyme was measured as 50% inhibition of NBT.

Proline content

The supernatant was prepared by homogenising roots in 3% sulfosalicylic acid and centrifugation for proline content determination. The acid–ninhydrin, glacial acetic acid and supernatant were mixed in equal volumes and kept at 100 °C for 1 h in a water bath. Then, the reaction was stopped in a cold bath and toluene was poured, agitated on vortex for 10–15 s and left undisturbed. The top layer was collected and the absorbance readings were taken at 520 nm. The standard solution of L-Proline was used to determine the proline content in milligrams per gram.³⁸

ATR-FTIR analysis

After 96 h the onion roots were collected from control and erythrosine exposed groups and rinsed with phosphate buffer saline. Then, the roots were lyophilized and analysed using Cary 630 FTIR Spectrometer (Agilent Technologies) in the 400–4,000 cm⁻¹ spectral range. The imaging system was coded for 64 scans at 16 cm⁻¹ resolution for 30 s measurement time. Before each sample measurement, the ATR diamond sensor was cleaned with acetone or ethyl alcohol.

Molecular docking

The AutoDock Tools MGL-1.5.7 (AutoDock, version 4.2) software was used to perform the molecular docking study to determine the binding affinities between the erythrosine molecule and DNA. In this study, the synthetic B-DNA dodecamer crystal structure (PDB ID: 1BNA; resolution: 1.90 Å) was downloaded from the Protein Data Bank (PDB) and selected as the receptor (target) molecule and erythrosine dye molecule as ligand. The crystal structure of erythrosine was drawn and energy minimized in ChemOffice version 22.0 and saved as pdb format. Then through Open Babel GUI, converted and saved in pdbqt format. Based on the results, the top-ranked ligand-receptor conformation was selected and visualized. The target and ligand preparation were done using AutoDockTools-1.5.7. Before docking, the polar hydrogen atoms were retained while non-polar hydrogens were merged in receptor and ligand molecules. Kollman charges and Gasteiger charges were applied to the receptor molecule (B-DNA dodecamer) and ligand molecule, respectively. In ligand, the free rotation of rotatable bonds was permitted during docking calculation. For easy interaction of receptor and ligand, the grid box of size 50 × 64 × 106 Å points with 0.375 Å grid spacing was set. In the docking parameters, the Lamarckian Genetic Algorithm along with other parameters set to default values given by AutoDock 4.2 were used. All the potential binding conformations between ligand and receptor after 50 docking runs were ranked based on binding free energy. The lowest free binding energy against B-DNA ranked first and clustered through AutoDock 4.2. The BIOVIA Discovery Studio Visualizer v21 was used to visualise and analyse the best docking conformation.

Data analysis

The software SPSS version 16.0 was used for statistical analysis. The data were represented as mean ± SE. One-way analysis of variance (ANOVA) and t-test were applied to evaluate the effect of time and concentration. For the significance of the difference between the control and treated groups for different concentrations and time intervals, the Tukey test was performed. The p-values lower than 0.05 (P ≤ 0.05) were considered statistically significant. A Pearson correlation analysis was performed between the MI, CAs and the comet assay with the different biochemical parameters using R-software. The ATR-FTIR data was analysed for Principal Component Analysis using the software MINITAB v18.

Results

Cytotoxicity and genotoxicity

In this study, synthetic food dye erythrosine was found to induce cytotoxicity and genotoxicity in root meristematic cells of A. cepa in a concentration and time-dependent manner. After erythrosine treatments, the mitotic index (MI) value decreased significantly in a time-dependent manner as well as in a concentration-dependent manner. The MI values decreased significantly by 27.85%, 38.49% and 36.16% (P ≤ 0.01) compared to control after 24 h, 48 h and 72 h, respectively at lowest concentration (0.1 mg/mL). The maximum decline of 57.81% was observed after 96 h from 9.03 ± 0.25 to 3.81 ± 0.49 (control) (Fig. 2A). Meanwhile, the cytotoxicity was so high at 0.25 mg/mL and above concentrations that no cell division was observed for all considered time duration. In the case of chromosome aberrations (CAs) observation, the treatments with erythrosine synthetic food dye showed significant increment, at lowest concentration (0.1 mg/mL) the CAs frequency increased 3.68-fold compared to control after 96 h exposure (Fig. 2B). The high levels of chromosomal aberrations (CAs) with increasing concentration and time duration indicated the genotoxic potential. The determination of CAs frequency at higher concentrations (0.25 mg/mL and above) was not possible due to complete inhibition of mitosis in roots exposed to erythrosine. The observed aberrations are well represented in the Fig. 3.

Fig. 2.

Effect of different concentrations of erythrosine on (A) mitotic index and (B) percent chromosomal aberrations in the root tip cells of A. cepa at different hours of exposure. Error bars represent standard errors (SE). Different letters a, b signify the effect of treatment at the same time interval, and p, q, r signifies the effect of duration of exposure.

Fig. 3.

CAs observed in A. cepa meristematic cells after erythrosine exposure (a and b) normal prophase; (c) normal metaphase; (d) normal anaphase; (e) normal telophase; (f–h) metaphase with chromosome break; (i) ring chromosomes; (j) diagonal metaphase; (k and l) sticky metaphase; (m) abnormal/disturbed metaphase; (n) delayed anaphase; (o) anaphase with chromosome bridge; (p) vagrant chromosomes; (q) laggard chromosome; (r–t) anaphase with chromosome break; (u and v) sticky anaphase; (w and x) abnormal/disturbed anaphase; (y) anaphase with ring chromosome; (z) C-mitosis; (aa) diagonal anaphase; (ab) anaphase polyploidy; (ac–ad) anaphase with bridge and chromosome loss; (ae) star shaped telophase; (af–ag) vagrant chromosomes telophase; (ah) telophase with chromosome bridge; (ai) disturbed telophase.

Comet assay

The extent of DNA damage in root tip cells of A. cepa after erythrosine exposure was determined using tail length, tail moment and olive tail moment as shown in Table 2. The comparisons among the treatment groups versus the control group (Tukey’s test) showed statistically significant differences between the mean values of the observed parameters. The tail length gradually escalated from 10.08 ± 0.37 (control) to 34.31 ± 1.38 at highest concentration (1 mg/mL) after 48 h exposure. After 96 h exposure, the maximum DNA damage was observed at the highest concentration where a 3.59-times increase in tail length and 4.67-times increase in tail moment was observed compared to the control. The olive tail moment also increased from 4.46 ± 0.35 to 21.27 ± 0.23 (4.77-fold) after 96 h exposure with the highest concentration (Fig. 4).

Table 2.

Effect of different concentrations of erythrosine on comet assay parameters; tail length, tail moment and olive tail moment in the root cells of A. cepa after different duration of time.

	Tail Length (μm)		Tail Moment		Olive Tail Moment
Duration →	48 h	96 h	48 h	96 h	48 h	96 h
Conc. (mg/mL)
Control	10.08 ± 0.37a	10.67 ± 0.0.23a	7.89 ± 0.06a	7.57 ± 0.8a	4.41 ± 0.59a	4.46 ± 0.35a
0.1	17.33 ± 0.75b	33.91 ± 1.08b	13.99 ± 1.85b	28.84 ± 2.07b	7.98 ± 0.49b	16.58 ± 0.43b
0.25	25.69 ± 0.89c	32.97 ± 0.91b	21.22 ± 1.95c	30.28 ± 2.53bc	12.85 ± 0.60c	17.46 ± 0.55b
0.5	34.21 ± 1.00d	37.65 ± 0.25c	30.95 ± 1.64d	33.86 ± 0.081cd	18.12 ± 0.62d	20.34 ± 0.77c
1.0	34.31 ± 1.38d	38.29 ± 0.75c	31.10 ± 0.58d	35.36 ± 1.46d	18.35 ± 0.74d	21.27 ± 0.23c
F-value	127.96**	243.84**	154.39**	134.89**	101.15**	181.39**

^*(P ≤ 0.05), ^** (P ≤ 0.01). The values are given as mean ± SE. Different letters a, b, c, d signify the effect of treatment at the same time interval., (Tukey’s test).

Fig. 4.

Photomicrographs showing level of DNA damage after exposure to different concentrations of erythrosine; (A) control, (B) 0.1, (C) 0.25, (D) 0.5 and (E) 1 mg/mL.

Biochemical parameters

Table 3 displays the effect of different concentrations of erythrosine on malondialdehyde (MDA) content in the roots of A. cepa. The results were found to be significantly different in a concentration-dependent manner compared to control for 48 h and 96 h exposure group. MDA content showed a significant (P < 0.01) increase after 48 h and 96 h which was found to be 4.59- and 5.47-fold higher than control, respectively at 1 mg/mL concentration. Whereas, after 48 h exposure duration, the catalase (CAT) activity reduced significantly from 2.35 ± 0.02 to 1.92 ± 0.06 and 0.83 ± 0.08 at 0.1 mg/mL and 1 mg/mL concentration, respectively which is 18.30% and 64.68% less compared to control. Similarly, the CAT activity reduced after 96 h exposure and a maximum reduction of 72.66% was observed at the highest concentration. The exposure of roots to erythrosine also resulted in the reduction of superoxide dismutase (SOD) activity. A significant (P < 0.01) decrease of 62.97% and 61.73% in activity was observed after 48 h and 96 h exposure respectively, as compared to the control. Similarly, the proline content of roots exposed to different concentrations of erythrosine was reduced in a concentration-dependent manner at both time durations. The proline content reduced to 0.92 ± 0.02 from 1.65 ± 0.05 (44.24%) after 48 h exposure and the maximum decline was found to be 78.11% after 96 h of exposure as compared to the control at higher concentration.

Table 3.

Effect of different concentrations of erythrosine on biochemical parameters; MDA content, CAT, SOD activity and Proline content in the root tip cells of A. cepa after different duration of time.

	MDA content (μmole g−1 FW)		Catalase (CAT) (unit min −1  g−1FW)
Duration →	48 h	96 h	48 h	96 h
Conc. (mg/mL)
Control	1.28 ± 0.04a	1.38 ± 0.07a	2.35 ± 0.02a	2.56 ± 0.11a
0.1	1.66 ± 0.04b	3.24 ± 0.13b	1.92 ± 0.06b	1.81 ± 0.16b
0.25	2.86 ± 0.05c	4.97 ± 0.20c	1.77 ± 0.06b	1.44 ± 0.12bc
0.5	4.29 ± 0.09d	6.25 ± 0.16d	1.37 ± 0.03c	1.14 ± 0.08cd
1.0	5.87 ± 0.04e	7.55 ± 0.11e	0.83 ± 0.08d	0.70 ± 0.06d
F-value	1190.00**	306.86**	114.22**	38.30**
	Superoxide dismutase (SOD) (unit min −1  g −1 FW)		Proline content (mg g −1 FW)
Duration →	48 h	96 h	48 h	96 h
Conc. (mg/mL)
Control	137.25 ± 1.17a	133.93 ± 2.57a	1.65 ± 0.05a	1.69 ± 0.05a
0.1	109.44 ± 2.57b	81.18 ± 1.22b	1.49 ± 0.02ab	1.20 ± 0.05b
0.25	80.39 ± 0.44c	67.34 ± 1.16c	1.34 ± 0.03b	0.98 ± 0.03c
0.5	58.53 ± 0.81d	58.90 ± 0.84d	1.00 ± 0.04c	0.83 ± 0.02c
1.0	50.82 ± 0.42e	51.26 ± 0.60e	0.92 ± 0.02c	0.37 ± 0.01d
F-value	720.29**	515.24**	74.14**	188.70**

^*(P ≤ 0.05), ^** (P ≤ 0.01) The values are given as mean ± SE. Different letters a, b, c, d signify the effect of treatment at the same time interval., (Tukey’s test). Here, FW represents fresh weight of A. cepa roots.

Pearson correlation analysis

Figure 5 represents the analysis performed to depict the statistical correlation between different parameters after 96 h of erythrosine exposure. The red colour showed a negative correlation and the blue colour showed a positive correlation. The results revealed that MI has a negative correlation with comet tail length, tail moment, olive tail moment and MDA content and a positive correlation with proline content, CAT activity and SOD activity. Whereas, the tail length, tail moment, olive tail moment and MDA content showed a highly negative correlation with antioxidant enzyme activities.

Fig. 5.

Pearson’s correlation between the different biomarkers evaluated after erythrosine exposure. MI (mitotic index), CAs (chromosomal aberrations), C.TL (comet tail length), C.TM (comet tail moment), C.OTM (comet olive tail moment), MDA (malondialdehyde), CAT (catalase), SOD (superoxide dismutase) and Proline content.

Attenuated total reflectance- Fourier transform infrared spectroscopy (ATR-FTIR)

The alterations in biomolecules viz. proteins, lipids, carbohydrates and nucleic acids were depicted via changes in the absorption frequency of infrared bands in different treatment samples. The general band assignment was presented in Table 4 and the spectrum (400–4,000 cm⁻¹) of control and treated group samples (0.1 and 1 mg/mL) after 96 h was shown in Fig. 6. The results showed a distinct difference in the spectrum of treated root samples compared to the control, especially at the highest concentration. In the protein region, the peaks at ~1,650 cm⁻¹ and ~1,550 cm⁻¹ corresponding to amide I (C=O stretching of proteins) and amide II (N-H bending/C-N stretching of proteins) respectively, were observed and the results revealed the reduction of 62.69% (amide I) and 66.04% (amide II) in absorption frequency in erythrosine exposed group root samples compared to control group at 1 mg/mL concentration. Similarly, significant decrease of 58.87% and 68.12% were observed at 1,260 cm⁻¹ and 3,132 cm⁻¹ peaks assigned to amide III and -NH symmetric stretching of protein spectrum, respectively. In carbohydrate region, at the peak ~1,155 cm⁻¹ corresponding to CO-O-C symmetric stretching showed significant reduction of 57.55% in absorption values upon erythrosine exposure compared to control. The absorption frequencies were also observed to be declined in lipid region, the values decreased 65.12% and 67.65% respectively at ~1,470 cm⁻¹ and ~1,750 cm⁻¹. The nucleic acid region assigned to ~961 cm⁻¹, ~1,080 cm⁻¹ and ~1,230 cm⁻¹ representing asymmetric, symmetric modes of phosphodiester group and C-N & -C stretch also revealed the similarly pattern, the erythrosine exposure caused reduction in absorption frequency in treated groups compared to control group. The decline of 59.06%, 52.22% and 57.55% at ~964 cm⁻¹, ~1,080 cm⁻¹ and~ 1,230 cm⁻¹, respectively was observed in erythrosine exposed group at highest concentration compared control. The Principal Component Analysis (PCA) was employed to identify differences between FTIR spectral data. The PCA derived scores plots along PCs 1 and 2 having eigen value greater than one was constructed for best segregation (Fig. 7). The PCs had total variability of 100% comprised of 88.9% of PC1 and 11.1% of PC2. In scores plot, nearness in space implies similarity between 0.1 mg/mL treated group and control group. While distance points of segregation represent the significant difference between control group and erythrosine exposed (1 mg/mL) group.

Table 4.

General band assignments of the FT-IR spectra.

	Wavenumber (cm −1 )
Lipids	1,750	C=O symmetric stretching
	1,470	CH2 bending
Proteins	3,132	-NH symmetric stretching
	1,650	amide I
	1,550	amide II
	1,260	amide III
Carbohydrates	1,155	CO-O-C symmetric stretching
Nucleic acid	1,225	asymmetric phosphate stretching
	1,080	symmetric phosphate stretching
	961–964	C-N & -C stretch

Fig. 6.

The representative mean FT-IR spectra of the control and erythrosine exposed roots of A. cepa in the 500–4,000 cm⁻¹ region.

Fig. 7.

3-D scores plot on PCs selected to demonstrate the best segregation of mean IR spectra derived from 96 h exposed groups of erythrosine. IR spectra were collected using ATR-FTIR spectroscopy. Each spectrum was expressed in terms of chosen PCs using MINITAB v18 software and rotated to identify the segregation of different clusters. Each symbol represents a single mean spectrum as a single point in “hyperspace”.

Molecular docking

The binding affinity of erythrosine was determined using 3D computational analysis on synthetic B-DNA dodecamer crystal structure to reveal the toxicity mode of action at the molecular level. Figure 8 shows the DNA nucleotides involved and different bond formations in the interactions of erythrosine with B-DNA. Erythrosine showed strong binding affinity into the minor groove of B-DNA surrounded by DNA bases Thy8, Cyt9, Gua10 (Chain A), Ade17, Ade18, Thy9 and Thy20 (Chain B) by non-covalent and covalent bonds. According to AutoDock, the binding energy is the sum of the intermolecular forces (van der Waals force + electrostatic energy + desolvation energy + hydrogen bond) acting upon the receptor-ligand complex. The interaction between erythrosine and B-DNA showed the strong binding affinity due to lower binding free energy of ΔG_best = −11.46 kcal/mol with inhibition constant (Ki) 3.96 nM (nanomolar).

Fig. 8.

A top-ranked conformation of the interaction between erythrosine and B-DNA structure. (A) DNA-erythrosine complex; (B) 3D ligand interaction diagram of the top-ranked docking pose. DNA nucleotides that interact with erythrosine are marked (b). Green dashed lines on the right image represent hydrogen bonds; (C) 2D ligand interaction diagram of the top-ranked docking pose.

Discussion

The cytotoxicity and genotoxicity of erythrosine synthetic food dye was evaluated using the Allium cepa test. Due to the high sensitivity of plant cells to environmental stress, A. cepa is regarded as an ideal model for assessing cytotoxicity and genotoxicity. The cytotoxicity assessment of various agents has been done based on the increase or decrease in mitotic index (MI) attributed to the number of dividing cells in total cells counted. In the present experimental work, erythrosine was found to significantly reduce MI with increasing concentration and exposure time and maximum after 96 h. The erythrosine exposure completely halted the mitosis at the concentration of 0.25 mg/mL and above at all exposure times considered. Besides cytotoxicity, the genotoxicity of erythrosine was revealed by increasing CAs in a concentration and time-dependent manner. The interference of toxicants during cell division formed CAs, by their direct action on DNA and chromosomal segregation, marked by changes in chromosomal structure or the total chromosome number.³⁹ Fiskesjo,¹⁸ was the first to propose the analysis of different types of CAs formed in all phases of the cell cycle. CAs facilitate a better analysis of the mode of action of test compounds regarding their aneugenic and clastogenic effects on the DNA allowing for a more accurate and comprehensive evaluation. Khan et al.²⁸ reported the genotoxicity in root meristematic cells of A. cepa after 24 h and 48 h exposure duration of metanil yellow and carmoisine at different concentrations (0.25%, 0.5%, 0.75% and 1.0% w/v). Their results showed a significantly decreased MI and various types of CAs observed mainly at the highest concentration and exposure duration. In the present study, the different types of CAs were observed such as chromosome stickiness, disturbed anaphase–telophase, vagrant and laggard chromosomes, c-mitosis, delayed anaphase, star-shaped telophase, chromosome bridges, chromosome breakages and formation of ring chromosomes represented in Fig. 4. The interference in spindle formation and chromosomal movement failure towards poles might cause laggard chromosomes and disturbed anaphase–telophase. Stickiness is most likely caused by DNA depolymerization, chromosomal contraction or condensation and partial breakdown of nucleoproteins. Chromosomal bridges might be formed by chromosome fusion or breakage, development of dicentric chromosomes, stickiness, uneven chromatid exchange, or crosslinking between chromosomes and proteins.⁴⁰ The reports suggest that observed significant reduction in MI in the current evaluation might be either due to mitodepressive activity of erythrosine or inhibition of DNA synthesis or a block in the G2 phase by erythrosine that prevents cell mitosis.²⁸ Moreover, the obstruction with the normal process of cell division could be the reason for the significant decrease in MI observed alongside the increased CAs with an increase in the concentration. The observed findings in this study and possible mechanism were supported by observation of Chequer et al.,⁴¹ who reported that erythrosine modulated the genes related to the DNA repair system and cell cycle.

Our results are like those attained by Prajitha and Thoppil,²⁶ who found a significantly altered cytogenetic system, inhibition of cell division and chromosomal aberrations (CAs) in A. cepa root tip cells exposed to food dyes orange red and lemon yellow. Similarly, Gomes et al.³ reported the cytotoxicity of bordeaux red, sunset yellow and tartrazine yellow in A. cepa L. after 24 h and 48 h exposure duration. Mpountoukas et al.² observed a high cytotoxicity and cytostaticity after in vitro erythrosine treatment to human peripheral blood cells. Mahfoz et al.⁴² reported carmoisine induced mito-depressive effects implying that it suppressed the DNA replication in A. cepa. Our results are also consistent with the study of the genotoxic effects of sunset yellow on root meristematic cells of Brassica campestris.⁴³ They observed significantly reduced MI and elevated CAs. Pandey et al.⁴⁴ also reported similar results in A. cepa after a genotoxicity study of five food preservatives butylated hydroxytoluene, butylated hydroxyanisole, sorbic acid, propyl gallate and sodium nitrate. However, our study contrasts with the results of Hagiwara et al.¹⁷ and Rogers et al.,⁴⁵ they found no significant erythrosine induced genotoxicity and cytotoxicity in Syrian hamster embryo (SHE) and V79 Chinese hamster lung cells, respectively. Miyachi and Tsutsui⁴⁶ reported erythrosine B had no genotoxic effect in Syrian hamster embryo (SHE) cells. Merinas-Amo et al.¹⁶ also reported non-significant toxicity of erythrosine in Drosophila melanogaster and HL-60 tumor human cell but increased tumor cell growth. While, Sarıkaya et al.¹⁵ noted inconclusive results in somatic mutation and recombination test of D. melanogaster. Likewise, Lakdawalla and Netrawali⁴⁷ observed erythrosine induced mutagenicity in the Bacillus subtilis, however, in another study they observed non-mutagenicity in Ames/Salmonella typhimurium assay.¹³ These studies highlight the importance of model- specific response.

The genotoxic potential of erythrosine was also determined by comet assay as the genetic information is carried by DNA in an organism; therefore, its integrity and stability are very important for sustainable life. In DNA, damage can occur through double and single-strand breaks due to oxidative stress and lipid peroxides. For DNA damage assessment, comet assay is one of the most suitable methods due to its easy application and authentication as a DNA damage biomarker.⁴⁸ The tail length (TL) was used to quantify the degree of DNA damage and the concept of olive tail moment (OTM) to explain DNA migration. OTM is very useful as it depicts the heterogeneity within a cell population and can also measure the DNA distribution variations within the tail. Similarly, tail length and tail intensity can be combined in a single parameter i.e. tail moment (TM) making it a very important and frequently used parameter.⁴⁹ Therefore, the tail length, tail moment and olive tail moment were included to determine the erythrosine induced DNA damage in A. cepa root cells. After 48 h and 96 h exposure, in the treated groups the values of all three parameters showed a significant increase compared to control which revealed the degree of DNA damage by erythrosine. Our results are in positive agreement with findings of Sasaki et al.,⁵⁰ who reported erythrosine induced dose-related DNA damage in the glandular stomach, colon, and urinary bladder in mouse. Chequer et al.⁴⁸ also reported that erythrosine caused DNA damage in HepG2 cells. They observed a significant increase in tail intensity and tail moment after exposure. The finding was also supported by observations of Mpountoukas et al.,² who reported DNA degradation due to strong binding of erythrosine to linear dsDNA determined by spectroscopic titration and electrophoretic mobility experiments.

Erythrosine belongs to the xanthene dyes class that contains halogen atoms. The xanthene dyes are reported to cause the production of the variable number of ¹O₂ that depends on the number of substituent halogen atoms and their atomic mass. The production of ¹O₂ may increase oxidative stress as it leads to generating peroxide products due to its highly reactive nature.⁵¹ Because in biological systems, the difference between the generation of oxidising species and cellular antioxidant defences instigates oxidative stress. The oxidative stress can damage the cellular constituents resulting in the alteration of antioxidant defence mechanisms. Therefore, to effectively evaluate the overall antioxidant status, the determination of malonaldehyde (MDA) content, proline content, catalase (CAT) and superoxide dismutase (SOD) enzyme activity has been included in the present study. The MDA content gradually increased and was found to be highest after 96 h exposure to erythrosine, compared to the control (5.47-fold). The attack of lipid peroxides on polyunsaturated fatty acids present in cell membranes might be the reason for increased MDA content after treatment. This results in altered biophysical properties of the cell membrane due to the formation of cross-links between membrane components. Consequently, it increases the permeability of the membrane, decreases the stability, and inactivates the activity of enzymes and essential fatty acids loss.⁵² A significant linear reduction in CAT activity was observed after erythrosine exposure with increasing concentration compared to control. The CAT enzyme found in most biological tissues converts hydrogen peroxide into oxygen and water in defence of oxidative damage. Hence, the decrease in the CAT activity alters the redox state of the cell.⁵³ The maximum decrease in CAT activity was observed to be 29.30% at the lowest concentration and 72.66% at the highest concentration after 96 h duration compared to control. Similarly, the SOD activity and proline content also decreased in a concentration-dependent manner at both exposure durations. The SOD breakdown the superoxide radicals into hydrogen peroxide and molecular oxygen. Hydrogen peroxide is further partitioned by catalase into water. Whereas, the proline acts as an osmo-protectant and helps in oxidative stress management.⁵⁴ The lower values of proline content, CAT and SOD activity indicate the depletion of antioxidant enzymes and increased oxidative stress. Moreover, this might be caused by inhibition of enzyme activity or the decline in enzyme synthesis caused by tissue injury.⁵² Our results are like those attained by Dharmar et al.,¹² who reported that erythrosine exposure caused decrease in CAT and SOD enzyme levels in zebrafish embryos. Wopara et al.,²⁷ also noted similar pattern of increased MDA level and decreased CAT activity in rat brain homogenates after co-exposure of erythrosine and tartrazine. However, contrasting results were observed by Gupta et al.,⁵⁵ who reported increased SOD enzyme activity in zebrafish embryo after erythrosine exposure. Demirkol et al.,⁵⁶ also noted no significant in MDA levels and CAT activity in Chinese hamster ovary cells.

In a cell homeostasis under normal physiological conditions is maintained by an endogenous system that consists of non-enzymatic antioxidants and antioxidant enzymes to keep a check on effects or damage caused by reactive oxygen species (ROS).⁵⁷ The results of the present study demonstrated that erythrosine exposure led to lipid peroxidation in root cells of A. cepa described by increased MDA content, one of the first indicators for cell membrane structural stability evaluation.⁵² Whereas, the decreasing pattern of CAT and SOD activity demonstrates erythrosine inference in the antioxidant line of defence and enzyme activity in roots. This makes cells vulnerable to oxidative stress indicating the inability of the antioxidant defence system to entirely scavenge the ROS produced that ultimately caused enhanced lipid production. This alternation in cellular redox potential resulted in oxidative damage to DNA and mitotic inhibition through either decrease or increase in the duration of the cell cycle.⁵⁷ The correlation between MI, CAs and DNA damage with biochemical parameters supported the observed results. The MI shows a highly positive correlation with antioxidant activity supporting results that decreasing antioxidant activity caused a reduction in MI. Whereas, MI showed a highly negative correlation between MDA level and DNA damage (comet assay) means observed increased DNA damage and lipid peroxidation caused a decrease in the MI. Similarly, decreasing antioxidant activity increased the DNA damage and MDA level represented by a highly positive correlation.

Furthermore, the structural changes in biomolecules induced by erythrosine were revealed by ATR-FTIR analysis. ATR-FTIR is a simple and less time-consuming technique to define the specific spectral signature associated with biological samples, and hence, used for clinical diagnosis and biomarker discovery.⁵⁸ In FTIR spectra obtained, the absorption band intensity is directly linked to the concentration of molecules helping in depicting the changes in structure and concentration in biochemical samples.⁵⁹ The results of the present study justified the effect of erythrosine on A. cepa. The peaks observed in control and treated samples were not merged and lower intensity was found indicating reduced nucleic acids, lipids and protein composition. Our results showed positive agreement with the study of cadmium stress on Padina tetrastromatica (Hauck) using FTIR and significant chemical composition alterations were observed.⁶⁰ In addition, to statistically define the difference in spectrum of different samples the principal component analysis (PCA) has been performed. PCA is a multivariate method used to lower the number of correlated variables (i.e. IR spectra) to smaller independent variables (PCs) and derive score plots along PCs 1 and 2. The PCs carry most of the variance present in the original data reducing the dimensions of data.⁵⁸ The construction of PCs score plot was done along PCs that have eigen value greater than one and rotation was allowed for the selection of view that showed the best segregation (i.e. control vs. 0.1 mg/mL group vs. 1 mg/mL erythrosine treated group). In the scores plot, the similarity is determined by closeness in space while distance denotes segregation in the computed IR spectrum of the examined biological samples. The segregation of points observed in the present study depicts the significant difference after treatment compared to control suggesting the effect of erythrosine at the molecular structural level.

To rationalize the genotoxic mode of action at the molecular level, we have investigated the binding affinity of erythrosine on crystal structure of synthetic B-DNA dodecamer (PDB code: 1BNA) utilizing a 3D computational analysis using AutoDock 4.2. The analysis molecular docking revealed the DNA minor groove recognition and intercalation ability of erythrosine to induce DNA damage. The results showed that erythrosine binds preferably to the AT-region in B-DNA structure with strong binding affinity stabilized by the non-covalent forces and lower binding free energy. In addition, the inhibition constant (Ki) was determined to be 3.93 nm which depicts the small number of molecules needed to inhibit the activity. The AT-rich region found in replication origin has been identified as the most universally conserved structural element in both prokaryotic and eukaryotic replicons.⁶¹ It may be concluded that the binding of erythrosine at that site caused the DNA replication process to stop and induced single and double-strand breaks that led to interference in genomic stability and high incidence of chromosomal aberrations and DNA damage found in our study were consistent with this notion.

Conclusion

The present study revealed that synthetic food dye erythrosine provokes cytotoxicity, genotoxicity and biochemical alterations in root meristematic cells of A. cepa. In addition, the ATR-FTIR spectrum showed the biomolecules alternation and molecular docking studies revealed the binding of erythrosine with DNA minor groove region which clearly depicted the reason of genomic instability leading to higher DNA damage observed in in vivo studies. Further, more clinical studies concerning the comprehensive risks associated with food dye consumption are necessary to enable risk managers to finalize and implement regulatory measures.

Author contributions

Mandeep Singh: Conceptualization, Methodology, Investigation, Data curation, Writing—original draft. Pooja Chadha: Conceptualization, Writing—review & editing, Supervision.

Conflict of interest statement. The authors have no relevant financial and non-financial interest to declare.