Raman spectroscopic comparison of zearalenone and its derivatives for non-destructive rapid detection

Abstract

Raman spectroscopy is used to investigate the absorption and dispersion of zearalenone (ZEN) and its derivatives. The C=C stretching vibration modes of ZEN and zearalenol (ZENOL) were appeared at 975–990 cm⁻¹. The C=O vibration mode was present at 1680–1690 cm⁻¹ for ZEN and zearalanone (ZAN), but it was absent for ZENOL and zearalanol (ZANOL) which have –OH group instead of –C=O group in ZEN and ZAN molecular structures. On the basis of this characterization, Raman spectra of specific chemical groups and linkages corresponding to the structural difference of ZEN and its derivatives were identified. These results indicated that Raman spectroscopy can apply for the identification of ZEN and its derivatives and has a potential for the non-destructive rapid detection of these compounds in food.

Introduction

Mycotoxins produced by Fusarium spp. are generally classified into two types: ones are the nonestrogenic trichotecenes including deoxynivalenone(DON), nivalenol, T-2 toxin, and diacetoxyscripenol, and the others are the mycoestrogens including Zearalenone(ZEN) and its derivatives, such as Zearalanone(ZAN), Zearalenol(ZENOL), and Zearalanol(ZANOL) (Bennett and Klich 2003). ZEN and its derivatives are known as one of strong mycotoxins with potent estrogenic activities, and frequent contaminants of cereal crops, such as maize, barley, oats, wheat and other grains throughout the world (Darra et al. 2019). Consequence of animal feedings of those crops, many cases have been observed estrogenic effects from diverse animals, such as bird, pig, cattle, and other domestic animals. Although the poultry is known to show relatively less sensitive reaction to these mycotoxins, the swine reacts the most severely among all domestic species. In case of the cattle, the infertility, reduced milk production, and hyperestrogenism have been reported due to association with these mycotoxins. While increasing consumption of processed foods such as cereals and ready to consume grains, their estrogenic activity is becoming a potential risk to human as well. Because of the potential toxic effects of ZEN and its derivatives for human health, the foods for human and animal feeds are closely monitored and tested for the presence of these toxins. Furthermore, many countries have established guidelines for maximum tolerance range of these mycotoxins from 30 to 1000 μg/kg in grains depends on the type of foods (Appell et al. 2022).

Recently the potential toxicity of ZEN and its derivatives on the reproductive structural and functional parameters have been intensely studied by many groups (Minervini et al. 2001; Zinedine et al. 2007). The estrogenic activities of ZEN and related compounds have been reviewed (Takemura et al. 2007). In the report, the authors claimed that sensitivity to the mycotoxins is related to species-dependent biotransformation pathways and these reactions show similarities to process in steroid metabolism. There are two metabolic pathways including hepatocytes and intestinal cells. Through these pathways, ZEN and its derivatives are metabolized to ZENOL with hydroxyl group by hydroxysteroid dehydrogenase (HSD). The high sensitivity to ZEN-mediated oestrogenic effects shown in some animals is related to this metabolic transformation of hydroxyl group. It has been proved that the higher oestrogenicity of ZENOL in pig and rat is related to its strong binding affinity to the cytoplasmic oestrogenic receptor (Minervini et al. 2006). As many studies showed, the small difference in structures of ZEN and its derivatives is one of the important factors for the species-dependent sensitivity. It also induces in the different binding affinity to the cytoplasmic receptor, resulting in difference symptom and sensitivity. In the previous studies, it was reported that the diverse analysis techniques based on chromatography and immunoassay are applied to detect these mycotoxins (Josephs et al. 2001; Krska et al. 2005). Particularly, chromatographic techniques including thin-layer chromatography (TLC) (Josephs et al. 2001; Schaafsma et al. 1998), gas–liquid chromatography (Josephs et al. 2001), gas chromatography-mass spectrometry (Tanaka et al. 2000; Zhang et al. 2006), HPLC (Saeger et al. 2003), and HPLC-mass spectrometry (Pallaroni and Holst 2003) are considered as a strong candidate method for the detection and determination of these mycotoxins. In addition, recently immunosensors and surface plasmon resonance (SPR) sensor have been employed to detect food contaminants with high sensitivity and fast analytical time (Abouzied and Pestka 1994; Choi et al. 2009). However, these methods often involve expensive, time-consuming pretreatment steps, including solid-phase extraction, separation, detection, and sample clean-up. Furthermore, they cannot provide detailed structural information of mycotoxins which are closely related in their structures. It is well known that spectroscopic method is non-destructive, rapid, and reliable experimental technique with minimal sample treatment. In previous studies, optical spectroscopic methods were proved as highly sensitive analytical techniques to obtain the structural information of target mycotoxins without complicated extraction steps (Gordon et al. 1998; Dowell et al. 1999; Pettersson and Åberg 2003; Delwiche and Hareland 2004; Delwiche 2008). Among the optical spectroscopic methods, Raman spectroscopy is suitable to study the structural differences at the molecular level. Because Raman spectra contain the detailed information of molecular structure, it has been employed as quite effective probes to investigate process-induced changes and to monitor quality of food (Li-Chan 1996; Belton 1993; Scotter 1997).

In this study, Raman spectroscopic technique was used to characterize the mycotoxin ZEN and its derivatives. Comprehensive characterization of Raman spectra of four mycotoxins (ZEN, ZAN, ZENOL and ZANOL) was conducted. From full spectral range of the spectra of those molecules, specific chemical groups and linkages corresponding to the subtle differences in their structures were identified. The main objectives of this research are to provide detailed spectroscopic data of the mycotoxins of ZEN and its derivatives.

Experiment

Zearalenone, β-zearalenol, 4earalenone, and β-zearalanol lenone, were purchased from Sigma (St. Louis, MO, USA). The molecular structures of ZEN and its derivatives are shown in Fig. 1. To determine the three-dimensional structure of ZEN and its derivatives, their theoretical calculation was performed by mean of computer simulation with MM2 (ChemBioOffice, ver.11.0). The molecule was set at a minimum energy position of 0 K for MM2. And then, molecular dynamics was carried out with a step size of 1 fs at a constant temperature of 298 K. And data analysis was performed for 300 ps.

Fig. 1

Molecular structure of ZEN and its derivatives

The Raman spectrum was obtained from the powder sample at room temperature (25 ± 2 ºC). Raman spectra were collected using a triple monochromator coupled with a CCD array detector (Triplemate 1877, Spex Industries, Edison, NJ, USA). The laser power used for all experiments was 15 mW at 514.5 nm (Innova 70 Ar + laser, Coherent Co., Santa Clara, CA., USA). The laser light was passed through a premonochromator to remove plasma emission lines. The scattered light was collected in a 45° angle geometry. Spectra obtained with 514.5 nm excitation were recorded using a 1800 grooves/mm holographic grating. All spectra were obtained in the Raman shift range between 300 and 3700 cm⁻¹ at room temperature (298 K).

Results and discussion

The theoretical calculation of molecular dynamics for the ZEN and its derivatives was performed by MM2 at a constant temperature of 298 K and the determined three-dimensional structures of them are shown in Fig. 2. As shown in Fig. 2, optimized geometry of ZEN and its derivatives well represented their three-dimensional characteristics due to the structural similarity and difference. The similarity of molecular structure between ZEN and its derivatives is not only having a benzene ring with two hydroxyl groups but also, having a ring-shaped circular carbon chain structure. This structural similarity is well represented in their optimized geometry in Fig. 2. Moreover, it was demonstrated well through the peak assignment of v(C–H) aromatic ring, v(C=C) aromatic ring, and v(C–C) aromatic ring in Table 1. While, the main difference between ZEN and its derivatives with estrogenic activity is the existence of C=C double bond and C=O residue in a circular carbon chain structure. Firstly, ZEN and ZENOL have a C=C double bond at near the benzene ring and this C=C double bond shows more rigid properties than that of C–C single bond. Therefore, it is expected that these structural rigidity due to the C=C double bond decrease the degree of freedom of circular carbon chain structure and result in the difference of three-dimensional structure against ZAN and ZANOL with C–C single bond at the same position. Their optimized geometries in Fig. 2 clearly represent these structural differences between ZEN/ZENOL and ZAN/ZANOL. The Raman spectra were measured in the range from 200 cm⁻¹ to 3700 cm⁻¹. Figure 3 shows the Raman spectra of ZEN and its derivatives; while the peak positions along with the appropriate vibrational assignments are listed in Table 1.

Fig. 2

The optimized geometry of ZEN and its derivatives, calculated by means of MM2

Table 1 Spectrum band position and assignment of zearalenone and its derivatives

Fig. 3

Raman spectra of zearalenone and its derivatives (zearalenone (a), zearalanone (b), zearalenol (c), zearalanol (d)). The spectral region between 1800 and 2800 cm⁻¹ was omitted due to the lack of spectra information

As shown in Fig. 3 and Table 1, the CH₂ stretching vibration modes and CH₃ stretching vibration modes of ZEN and its derivatives appeared at 2848–3028 cm⁻¹. The peak at 3075–3085 cm⁻¹ could be attributed to the =C–H stretching vibration mode. Raman peak at 3228–3230 cm⁻¹ were C–H stretching vibration mode in aromatic ring (Duley and Williams 1983; Jones et al. 1983; Barrett 1981; Vasko et al. 1972). The peak at 3483–3485 cm⁻¹ could be attributed to the –OH stretching vibration mode of aromatic ring in the ortho-position forming (Shaw et al. 1990). As shown in Fig. 4, the C–C stretching vibration modes of ZEN and its derivatives appeared at 875–889 cm⁻¹ (Korolevich et al. 1990). Meanwhile, the C=C stretching vibration modes of ZEN and ZENOL appeared at 976–980 cm⁻¹ and the peak at 1621–1624 cm⁻¹ could be attributed to the C=C stretching mode (Table 1) (Potts and Nyquist 1959; Wexler 1967). Because the v(C=C) aryl peaks at 976–980 cm⁻¹ and 1621–1624 cm⁻¹ are the structure specific Raman peaks which can distinguish ZEN and ZENOL from ZAN and ZANOL, these peaks can be used as one of spectroscopic fingerprint for the determination of ZEN and ZENOL among the related mycotoxin species. Raman peak at 1019–1025 cm⁻¹ were C–C stretching vibration mode in aromatic ring (Korolevich et al. 1990; Wexler 1967; Kacuráková and Mathlouthi 1996), and the peak at 1327–1332 cm⁻¹ could be attributed to the C–OH stretching vibration mode of aromatic ring in the para- position forming (Barrett 1981). The C–OH stretching vibration modes of ZENOL and ZANOL were appeared at 1126–1128 cm⁻¹ (Green et al. 1971). The CH₃ bending mode of ZEN and its derivatives were appeared at 1362–1370 cm⁻¹, and the peaks at 1460–1473 cm⁻¹ were attributed to contribution of two main vibrational modes, such as CH₂ bending and CH₃ bending (Nava et al. 1996; Bourée et al. 1996). Raman peak at 1644–1653 cm⁻¹ corresponds to C=C stretching vibration mode in aromatic ring (Potts and Nyquist 1959; Wexler 1967). ZEN and ZAN have a C=O residue in the middle of circular carbon chain structure. The existence of C=O residue is one of main reason for the difference of three-dimensional structure between ZEN and it derivatives, along with the C=C double bond. As shown in Fig. 4 and Table 1, the C=O vibration mode appeared at 1688–1693 cm⁻¹ (Schrader et al. 1997) for ZEN and ZAN, but did not appear for ZANOL and ZENOL which have –OH group instead of –C=O group in ZEN and ZAN molecular structures. Because the v(C=O) peaks at 1688–1693 cm⁻¹ are the structure specific Raman peaks which can distinguish ZEN and ZAN from ZENOL and ZANOL, therefore, we expect that these peaks can be used as one of spectroscopic fingerprint for the determination of ZEN and ZAN among the related mycotoxin species.

Fig. 4

Raman spectra of zearalenone and its derivatives in the 800–1800 cm⁻¹ region (zearalenone (a), zearalanone (b), zearalenol (c), zearalanol (d))

The optimized geometrical images of ZEN and its derivatives in Fig. 2 well represented their structural differences. These differences due to the three-dimensional structure of molecules frequently bring different inter- or intra-molecular interaction. The widths and intensities of the spectral bands marked small changes depending on the difference of molecular structures. Firstly, the C=O band in ZAN was shifted to lower wavenumber (1688 cm⁻¹) compared to that of ZEN (1693 cm⁻¹). This seems to be mainly due to the inter-molecular interactions between the C=O group and C–H group of neighboring molecules (Yu et al. 2002). When the CO group of ZEN and ZAN belong to the inter-molecular interaction, they can form hydrogen bonding with C–H group of neighboring molecules. At that time, there are two C–H candidate groups which can participate in inter-molecular hydrogen bonding. One is aromatic C–H in benzene ring and the other is aryl C–H in circular carbon chain. In case of aryl C–H group, the Raman peak appears as symmetric and asymmetric CH₂ stretching mode. If the C–H group belongs to the inter-molecular hydrogen bond, it will be expected that the symmetric CH₂ mode will decrease while the asymmetric CH₂ mode will increase because of the interference of CO group of neighboring molecules. In our results, the CH₂ sym peaks of ZEN and ZAN showed very lower intensity than that of ZENOL and ZANOL. The restriction of CH₂ symmetric vibration mode is due to the inter-molecular interaction between C=O group (Raman peak at 2848–3028 cm⁻¹) and C–H group in the circular carbon chain. Meanwhile, a previous report explained that aromatic (C–H) peak would shift to a lower frequency by hydrogen bond formation due to the weakening of C–H bond and it would shift to a higher frequency upon breaking down the hydrogen bonding since the weakened C–H bond would be reinforced (Yu et al. 2002). In our results, there is no remarkable shift of aromatic (C–H) peak to the lower frequency. These results indicate that C–H group of aromatic ring was weakly participating to the intra-molecular interaction between other adjacent molecules. Moreover, R–OH⋯O = C peak can be affected by intra-molecular interaction between aromatic R–OH in benzene ring and C=O group in circular carbon chain, appeared at 3483–3485 cm⁻¹ and did not show frequency shifts. These results indicate that the changes due to the C=C double bonding and C=O/C–OH group exchanging, give just very small effect to the intra-molecular interaction between R–OH in benzene ring and C=O group in circular carbon.

Conclusion

In this study, we characterized some specific Raman peaks of ZEN and its derivatives based on their structural differences. The C=C stretching vibration modes of ZEN and ZENOL were appeared at 975–990 cm⁻¹. The C=O vibration mode was present at 1680–1690 cm⁻¹ for ZEN and ZAN, but it was absent for ZANOL and ZENOL which have –OH group instead of –C=O group in ZEN and ZAN molecular structures. On the basis of this characterization, Raman spectra of specific chemical groups and linkages corresponding to the structural difference of ZEN and its derivatives were identified. These results indicated that Raman can apply for the identification of ZEN and its derivatives and has a potential for the non-destructive rapid detection of ZEN and its derivatives in food. Our studies are in progress to deepen understanding the spectroscopic difference between ZEN and its derivatives in low frequency range under 200 cm⁻¹. Moreover, we will study the Raman spectroscopy of mixed sample with corn and wheat for the non-destructive rapid detection of mycotoxin, as a future work.