Introduction

Mercury (Hg) is a persistent and bioaccumulative toxic substance that is widely released into the environment (Schroeder and Munthe 1998). The toxic effects of mercury on aquatic animals and mammals are unquestioned (Gentès et al. 2015; Oliveira et al. 2017; Sun et al. 2019; Zhang et al. 2020a). However, the potential environmental risks of mercury in terrestrial systems remain largely uncertain (Tang et al. 2018). In some studies, the mortality increased, and the growth rate decreased when earthworms were exposed to mercury (Zhu et al., 2011; Da Silva et al. 2016), while death or growth inhibition was not observed in others (Lock and Janssen 2001; Mahbub et al. 2017). These differences in toxicity of mercury in earthworms may be due to exposure doses and durations. Buch et al. (2017) reported that mercury at 0–25 mg/kg caused a 10% inhibition effect on reproduction (EC10) of earthworms, while another study conducted at the same doses reported that mercury increased their reproduction (Abbasi and Soni 1983). However, this element is known at a wide scale by its toxicity at high doses and its amount in the ecosystem increases day by day (Programme 2013; Li et al. 2016). Bioaccumulation may be used as a bioindicator of mercury in the terrestrial environment (Dang et al. 2016; Le Roux et al. 2016). Toxicological screening of non-target (earthworm) has been considered as a vital endpoint in ecotoxicology study (Lankadurai et al. 2015; Ponsankar et al. 2020) with a key role of earthworm in the structure and nutrient value of soil (Vasantha-Srinivasan et al. 2016).

Hydrogen is the smallest and lightest molecule in nature with colorless and odorless properties (Alwazeer 2019). Due to its small size, molecular hydrogen can rapidly pass biomembranes to the cytoplasm, mitochondria, and nucleus to exert its protective effect (Ohta 2015; Zhai et al. 2017). Numerous studies showed that molecular hydrogen can protect various organs and tissues and has potential therapeutic properties for many diseases such as diabetes (Zhang et al. 2018), obesity (Nakao et al. 2010), and pancreatitis (Chen et al. 2010). Molecular hydrogen has been proven to possess a selective antioxidant activity against hydroxyl radicals (Ohsawa et al. 2007). Moreover, recent studies in animals, humans, and in vivo have shown that molecular hydrogen possesses therapeutic benefits in improving excessive inflammation and oxidative stress (Slezák et al. 2016; Yoneda et al. 2017; Jackson et al. 2018; Lu et al. 2019; Alwazeer et al. 2021). Hydrogen-rich water (HRW) is known to alleviate ischemia–reperfusion injury (impaired cardiac functions) in mice and decrease the oxidative stress level of myocardial tissue (Li et al. 2019). HRW showed also a protective effect on radiation-induced cognitive dysfunction thanks to its antioxidative and anti-inflammatory properties and its protection of neonatal neurons by regulating the BDNF-TrkB signaling pathway (Liu et al. 2019). In toxicological studies, the toxicity effect of a substance on the earthworm body is generally measured by either its lethal concentration or by its residue level in the tissue. Metabolic disturbance after exposing earthworm to a toxic substance could be measured by analyzing either the whole-worm homogenate or the coelomic fluid. However, these methods require homogenization and extraction procedures with extra steps for further molecular level analysis (Aja et al. 2014). Little information is available about the variation in the coelomic fluid of earthworms or its chemical complement in whole-body homogenates in response to toxic stress factors such as heavy metals.

Reactive oxygen species (ROSs) are highly reactive derivatives of oxygen molecules that form as a result of oxidative reactions in cells especially in mitochondria (Sarniak et al. 2016). ROSs play critical roles in a variety of cellular processes such as immune response and cell signaling when found at moderate levels (Magherini et al. 2019). However, when their level increases due to oxidative stress and exceeds the capacity of the antioxidant defense system, ROSs react with different cellular components such as proteins, lipids, deoxyribonucleotide triphosphates (dNTP), and DNA causing alteration of their structure and disruption of their physiological functions (Sies 1993; Thannickal and Fanburg 2000; Cooke et al. 2003). 8-Oxo-7,8-dihydro-2′-deoxyguanosine (8-oxo-dG) and 8-hydroxy-2′-deoxyguanosine (8-OHdG) were typically used as critical biomarkers of oxidative stress-derived DNA damage (Ma et al. 2016; Pour Khavari and Haghdoost 2020; Wen et al. 2021). A significant increase of 8-oxo-dG in serum or urine was related to several diseases such as cancer, diabetes, and autoimmune diseases (Khavari et al. 2018; Thomas et al. 2018; Halczuk et al. 2020; Urbaniak et al. 2020). Techniques for 8-oxo-dG measurement include chromatographic approaches using GC–MS, GC–MS/MS, LC–MS/MS, and HPLC-ECD (Collins et al. 2000; Boysen et al. 2010). FTIR spectroscopy is a sensitive and highly reproducible physicochemical analytical technique that depends on measuring the wavelength and intensity of the absorption of IR radiation by the sample. The main advantages of FTIR over other techniques are that spectra can be obtained for a wide range of proteins, and direct correlations are possible between amide I band frequencies and secondary structure components (Aja et al. 2014). In another study, we proved that HRW could alleviate the nickel-induced toxic responses (inflammatory responses, oxidative stress, DNA damage) in the earthworm model (Köktürk et al. 2021). In the present study, we aim to screen the possible effects of hydrogen-rich water on mercury toxicity by interpreting biochemical changes in earthworms using ATR-FTIR and LC–MS analyses.

Materials and methods

Chemicals

Mercury(II) chloride (HgCl2) was purchased from Sigma. Stock solutions were prepared in both ultrapure water and hydrogen-rich water (Ultrapure water direct-Q®8 UV system) (HRW). Hydrogen-rich water was prepared using ultrapure water direct-Q®8 UV system (Millipore, USA). Liquid chromatography grade (LC) acetonitrile and methanol were purchased from Sigma-Aldrich. Analytical grade formic acid and ammonium formate were obtained from Sigma-Aldrich.

Preparation of hydrogen-rich water (HRW)

HRW was prepared according to Ryu et al. (Ryu et al. 2019) with some modifications. H2 gas was bubbled directly into mercury (II) chloride (HgCl2) solutions (5, 10, 20, and 40 μg/mL) for 10 min at 0.5 L/min with a hose equipped with a needle. The prepared solutions were then injected into earthworms within 5 min. The concentration of hydrogen in HRW samples and its stability were evaluated in preliminary experiments using the ORP electrode (Sensorex, USA). The levels of hydrogen were 1.6 ppm, and it did not change significantly during the assay period, i.e., 5 min (from HRW preparation moment to application one).

Animal study

Red California earthworms (Eisenia fetida) grown at the earthworm manure production unit at the College of Applied Sciences, Department of Organic Farming were used. Earthworms were kept at 20 ± 2 °C and 75% humidity and complete dark conditions until use. Healthy and adult soil worms (350–400 mg individual weight, adult earthworms) were selected for the experiment and kept under trial conditions in the laboratory 2 weeks before the experiment started.

The direct injection method was used by many researchers to investigate the toxic metabolism of chemicals in vivo in the coelomic cavity of earthworms (Nakatsugawa and Nelson 1972; Park et al. 2012; Yesudhason et al. 2018). Micro Fine Plus 0.5 mL (0.3 × 8.0 mm) needles were used. In this experiment, the earthworms were adapted to the test medium for 12 h on moist filter papers placed in a Petri dish with a vent hole. Before the direct injection, the earthworms were kept on ice for 10 min to cool down. The concentrations used in this study were chosen according to previous studies (Yesudhason et al. 2018). Twenty microliters of the HgCl2 solutions (5, 10, 20, and 40 µg/mL) was injected into the coelomic cavity of earthworms according to previous studies (Yesudhason et al. 2018; Kim et al. 2020). The application groups prepared with HRW were injected into the worms within 5 min after preparation so that the level of molecular hydrogen in water was not changed. In preliminary experiments, the hydrogen level in water was not significantly changed for few hours (data not shown). Different concentrations of mercury chloride (H1: 5 µg/mL, H2: 10 µg/mL, H3: 20 µg/mL, H4: 40 µg/mL, and C1: control) and mercury chloride prepared in hydrogen-rich water (H5: 5 µg/mL, H6: 10 µg/mL, H7: 20 µg/mL, H8: 40 µg/mL, and C2: control) were injected into earthworms. The experiments were designed with three replicates (n = 3) and 6 earthworms per group. During the experiment (48 h), the temperature was set at 20 °C and the humidity at 75%. Moist filter sheets were changed daily. When the analysis was terminated at 48 h, earthworms in each treatment group were washed with ultrapure water and placed in a Petri dish on a wet filter paper in the dark at 20 °C ± 2 °C for 12 h to remove their intestinal contents. The control and treated earthworms were directly freeze-dried using a lyophilizer (Christ Alpha 1–2 LD Plus, 2015, Germany), and samples were stored at − 80 °C until analysis.

ATR-FTIR spectroscopy

Six earthworms were taken from each test replicate, and after washing step with ultrapure water they were used for analysis. Details of FTIR-ATR techniques are available in previous studies (Rodriguez-Seijo et al. 2017; Paço et al. 2017; Rodríguez-Seijo et al. 2018). Fourier transform infrared-attenuated total reflectance (ATR-FTIR) analysis was performed using Agilent Cary 630 FTIR spectrometer. All spectra were obtained using 64 scans and a resolution of 4 cm−1 within the 4000–400 cm−1 range. Graphs of FTIR results were drawn using OriginPro 8 software. Each infrared spectrum was compared with the corresponding infrared spectra of all other different concentrations.

Since the ATR-FTIR spectrum of whole earthworm tissue is very complex and consists of various bands affiliated with the various functional groups in proteins, lipids, polysaccharides, and nucleic acids, the analyses were carried out in three different wavenumber regions, i.e., 4000–2800 cm−1 (C–H stretching region) and 1800–400 cm−1 (fingerprint region).

Table 1 shows that the (≠ 1) band 3280 cm−1 is mainly due to N–H vibrations in proteins, O–H stress vibrations in polysaccharides, and water. Since the water of the samples was completely removed, water has no contribution to this band, and only the contribution of proteins and polysaccharides could be taken into account (Çakmak et al. 2003; Cakmak et al. 2006; Cakmak-Arslan et al. 2020).

Table 1 General band assignment of ATR-FTIR spectrum of earthworm tissue based on the literature (Rodríguez-Seijo et al. 2018; Cakmak-Arslan et al. 2020)

LC–ESI–MS/MS

Earthworm sample preparation

The sample preparation method for LC–ESI–MS/MS analysis was performed according to Wang et al. (Wang et al. 2014) and Griffith et al. (Griffith et al. 2019) with little modifications. Six lyophilized earthworms were taken from each group for LC–ESI–MS/MS analysis. Earthworm tissues of C2, H4, and H8 groups were prepared with methanolic solvent (50%). The samples were then vortexed for 2 min at 2000 rpm (Bioprep-24 Homogenizer) and 4 °C followed by centrifugation at 16,000 × g (Hettich Universal 320 R, Germany) at 4 °C for 20 min. The supernatant was then left in ice for 10 min until LC–MS analysis. All the solutions were filtrated before LC–ESI–MS/MS analysis using Captiva Premium Syringe Filter, polypropylene housing, nylon membrane (0.45 µm). The product and precursor ion of molecules were analyzed by LC–ESI–MS/MS.

Mass spectrometer and chromatography conditions

A 1260 Infinity II LC System model (Agilent, USA) High-Performance Liquid Chromatograph (HPLC) coupled with a Tandem Mass Spectrometer was used to carry out a qualitative evaluation ion of some molecules. The reversed-phase HPLC was equipped with a column oven (1260 TCC), binary pumps (1260 Bin Pump), and a degasser (1260 Degasser). The chromatographic conditions were optimized to achieve optimum separation of compounds and overcome the suppression effects. Thus, the chromatographic separation was performed on a reversed-phase Agilent Poroshell 120 EC-C18 model (100 mm × 3.0 mm, 2.7 μm) analytical column. The column temperature was set to 25 °C. The elution gradient was composed of eluent A (water + 5 mM ammonium formate) and eluent B (acetonitrile + 0.1% formic acid). The applied elution was 75% (A) –25% (B), and the solvent flow rate and injection volume were settled as 0.5 mL/min and 5 μL, respectively.

The mass spectrometric detection was carried out using an Agilent 6460 Triple Quad Mass Spectrometer System Model Tandem Mass Spectrometer equipped with an electrospray ionization (LC–ESI–MS/MS) source operating in both negative and positive ionization modes. LC–ESI–MS/MS data were acquired and processed by Agilent Mass Hunter Software. The MRM (Multiple Reaction Monitoring) method was optimized to selectively detect and quantify the phytochemical compounds based on the screening of specified precursor phytochemical-to-fragment ion transitions. The collision energies (CE) were optimized to generate optimal fragmentation and maximal transmission of the desired ions. The MS operating conditions were applied as follows: drying gas as nitrogen (N2) at a flow of 15 L/min; nebulizing gas as nitrogen at a flow of 11 L/min; and capillary (V) at 4000 V and gas temperature at 350 °C.

Statistical analysis

Analyses of variance (one-way ANOVA) with Tukey’s post hoc tests were used for multiple comparisons of FTIR band area values. One-way ANOVA has been performed with GraphPad Prism 9 Software. The degree of significance was denoted as **p < 0.01 and ***p < 0.001. FTIR band area values for the heat map chart were evaluated using the IBM SPPS v21 program, and the normality of the variables was evaluated by one-sample Kolmogorov–Smirnov and Shapiro–Wilk tests. Differences between groups were tested using one-way ANOVA for parametric data and the Mann–Whitney U test for non-parametric data. The values of p < 0.05 were considered significant in the evaluation of the data.

Results and discussion

ATR-FTIR study

The present study revealed that both ultrapure water-prepared HgCl2 (5, 10, 20, and 40 µg/mL) and hydrogen-rich ultrapure water-prepared HgCl2 concentrations (5, 10, 20, and 40 µg/mL) did not exhibit mortality effect against E. fetida. However, spectroscopic data of all cell structures can show structural and/or compositional changes occurring during cellular metabolic activity due to the effects of xenobiotics and environmental changes (stress factors) (Kamnev 2008; Mechirackal Balan et al. 2018). In this study, we evaluated the metabolic changes of earthworms exposed to HgCl2 stress using FTIR spectroscopic analysis. Besides, we examined how the different concentrations of HgCl2 prepared with hydrogen-rich water can affect the cellular metabolic activity compared to the control group, i.e., earthworms prepared with ultrapure water. Figures 1A, B and 3A–D show average ATR-FTIR spectra of C1, C2, H1, H2, H3, H4, H5, H6, H7, and H8-treated groups of earthworm. The main absorption bands were numbered in these figures, and the assignments of these bands were presented in Table 1 and Fig. 2.

Fig. 1
figure 1

A) FTIR-ATR spectra of the earthworms after exposure to different concentrations of HgCl2 (prepared with ultrapure water) (5, 10, 20, and 40 µg/mL) and control (ultrapure water). B) Different concentrations of HgCl2 (prepared with hydrogen-rich water) (5, 10, 20, and 40 µg/mL) and control (hydrogen-rich water)

Fig. 2
figure 2

Changes in the band area values of the infrared bands of earthworm tissues of C1, C2, H1, H2, H3, H4, H5, H6, H7, and H8-treated earthworms. Each value presents a mean ± SD. The degree of significance was denoted as **p < 0.01, ***p < 0.001. C1: Control (ultrapure water), C2: Control (hydrogen-rich water), H1: 5 µg/mL HgCl2 (prepared with ultrapure water), H5: 5 µg/mL HgCl2 (prepared with hydrogen-rich water), H2: 10 µg/mL HgCl2 (prepared with ultrapure water), H6: 10 µg/mL HgCl2 (prepared with hydrogen-rich water), H3: 20 µg/mL HgCl2 (prepared with ultrapure water), H7: 20 µg/mL HgCl2 (prepared with hydrogen-rich water), H4: 40 µg/mL HgCl2 (prepared with ultrapure water), H8: 40 µg/mL HgCl2 (prepared with hydrogen-rich water)

Results show that there is no significant difference in some spectra (≠ 1 and ≠ 2) between the lowest concentrations of the treatment groups, i.e., H1 and H5 (Figs. 2 and 3A) (p˃0.05). However, when the H3 group is compared with the H7 application group, there is a significant difference in some spectra (≠ 1, ≠ 2, ≠ 3, ≠ 10, ≠ 11) (Figs. 2 and 3C) (p < 0.01, p < 0.001).

Fig. 3
figure 3

A) FTIR-ATR spectra of the earthworms after exposing to 5 µg/mL HgCl2 and control. C1: Control (ultrapure water), C2: Control (hydrogen-rich water), H1: 5 µg/mL HgCl2 (prepared with ultrapure water), H5: 5 µg/mL HgCl2 (prepared with hydrogen-rich water). B) 10 µg/mL HgCl2 and control. C1: Control (ultrapure water), C2: Control (hydrogen-rich water), H2: 10 µg/mL HgCl2 (prepared with ultrapure water), H6: 10 µg/mL HgCl2 (prepared with hydrogen-rich water). C) 20 µg/mL HgCl2 and control. C1: Control (Ultrapure water), C2: Control (hydrogen-rich water), H3: 20 µg/mL HgCl2 (prepared with ultrapure water), H7: 20 µg/mL HgCl2 (prepared with hydrogen-rich water). D) 40 µg/mL HgCl2 and control. C1: Control (Ultrapure water), C2: Control (hydrogen-rich water), H4: 40 µg/mL HgCl2 (prepared with ultrapure water), H8: 40 µg/mL HgCl2 (prepared with hydrogen-rich water)

In the H2 and H6 application groups, we determined changes in the bands of 3278 cm−1 and 3279 cm−1, and in the H3 and H7 application groups, the bands of 3279 cm−1 and 3280 cm−1, respectively (Fig. 3B, C). When the 3280 cm−1 (≠ 1) band was evaluated for the H3 treatment group, it showed a more significant decrease than the H7 treatment group that was prepared with hydrogen-rich water (p < 0.01) (Figs. 2 and 4). Additionally, the low decrease in the polysaccharide band in the groups prepared with hydrogen-rich water could be related to the protective effect of molecular hydrogen. The decrease of this band in the H3 group indicates a decrease in the unsaturated lipid substances. The change in the bands and the shift in the frequency values are likely due to the change in lipid metabolism induced in the H3 and H7 groups. Thus, the mercury present in hydrogen-rich water may affect proteins and polysaccharides differently than other samples of middle dose (20 µg/mL). Different concentrations of mercury show an intense peak in 3420 cm−1 characteristics of an O–H stretch caused by intra-hydrogen and intermolecular bridges in the bands between 3100 and 3650 cm−1 in living beings, and a significant decrease in band density was observed in this bandgap at different concentrations of metals (Deguchi 2014; Rocha et al. 2020). This change of polysaccharide composition was supported by the change in the band of 1047 cm−1 (≠ 11) (Fig. 2) where the decrease in H2 (p < 0.01) and H3 groups (p < 0.001) compared with controls (C1 and C2) were more significant than H6 and H7 (Figs. 2 and 4). In addition, considering the same band, there was a significant decrease in the H3 group compared with the H7 group (p < 0.001) (Fig. 2).

Fig. 4
figure 4

A heat map including hierarchical cluster analyses (dendrograms) for the samples (dataset rows) and the selected FTIR peaks (dataset columns). In the dendrogram, 11 spectra were clustered into ten groups as anticipated. In the heatmap, different colors are showing the relative abundance of the selected spectra. C1: Control (ultrapure water), C2: Control (hydrogen-rich water), H1: 5 µg/mL HgCl2 (prepared with ultrapure water), H5: 5 µg/mL HgCl2 (prepared with hydrogen-rich water), H2: 10 µg/mL HgCl2 (prepared with ultrapure water), H6: 10 µg/mL HgCl2 (prepared with hydrogen-rich water), H3: 20 µg/mL HgCl2 (prepared with ultrapure water), H7: 20 µg/mL HgCl2 (prepared with hydrogen-rich water), H4: 40 µg/mL HgCl2 (prepared with ultrapure water), H8: 40 µg/mL HgCl2 (prepared with hydrogen-rich water)

When nickel and chromium metals were applied in Escherichia coli bacteria, the band determined as 1052 cm−1 in the control sample decreased to 1049 cm−1 and 1047 cm−1 in the nickel and chromium application groups, respectively (Gupta and Karthikeyan 2016). This change of polysaccharides could be due to the change of the peptidoglycan surface density of the cell wall to better adapt to their environment under heavy metal stress (Vollmer et al. 2008). Polysaccharides inhibit fatigue and bone loss activities as well as have a protective effect against oxidative stress and injuries (Yuan et al. 2019).

The CH2 antisymmetric stretching at 2919 cm−1 and the CH2 symmetric stretching at 2851 cm−1 is associated with saturated lipid concentration of membranes. The concentration of saturated lipids is important for the detection of membrane fluidity levels (Markowicz et al., 2010; Kardas et al. 2014). Antisymmetric and symmetrical methyl (vasCH3, vsCH3) and methylene (vasCH2, vsCH2) groups in spectral 3000–2850 cm−1 regions have tensile vibrations of lipid and phospholipids, and 2850 cm−1 IR absorption bands are used to monitor lipid changes (Mendelsohn and Moore 1998; Lewis and McElhaney 2007, 2013). In the present study, the band of 2855 cm−1 (≠ 3) (Fig. 2) shows a more significant decrease in H3 treatment groups compared with the H7 group (p < 0.001) (Figs. 2 and 4). In the groups without hydrogen-rich water application, changes in the 2855 cm−1 band may lead to changes in the permeability of the membranes, the higher order of lipophilic carbon–carbon chains, and a significant increase in saturated lipid concentrations (Kepenek et al. 2019). The decrease in saturated lipid concentration may result from decreased lipid biosynthesis or degradation through lipid peroxidation (Simsek Ozek et al. 2014). However, the absolute frequency of CH stretch bands can no longer be used to predict the degree of relative lipid impairment, and a low value of these parameters also indicates the presence of a long chain of fatty acids (Staniszewska et al. 2014). Besides, the density of CH stretch bands increases with advancing age and the development of diseases (Anastassopoulou et al. 2018).

The molecular changes in organisms exposed to stressors could be used to understand the responses to stress agents, and the peak at ≈1540 cm−1 corresponds to amide II structures (G. Muthukaruppan 2015; Rodriguez-Seijo et al. 2017). Table 1 (≠ 6) shows the FTIR-ATR spectra (1536 cm−1) of the earthworms after exposure to different concentrations of HgCl2 and controls. The band 1536 cm−1 in the H3 groups prepared without hydrogen-rich water shows a significant difference compared to the controls (C1 and C2) and H7 (p˂0.05) (Fig. 4). Similarly, the effect of xenobiotics in mice and the density and frequency of amide II bands (1453 cm−1 and 1525 cm−1) was decreased due to asymmetric methyl deformation and stretching of the C = N, C = C groups compared to the control (Ashtarinezhad et al. 2014, 2015).

Comparing to the control groups (C1 and C2), there was no significant difference in H7 of the hydrogen-rich application groups in the 1230 cm−1 band (≠ 9), while a significant decrease was observed in the H3 group (p < 0.05) (Fig. 4). The intensity and frequency of the bands around 1256 cm−1 and 1219 cm−1 in treated tissue were reduced and shifted compared to untreated sample tissue, mainly owing to PO2 asymmetric (phosphate I; 1256 cm−1) and PO2 asymmetric vibrations of nucleic acids when it is highly hydrogen-bonded asymmetric hydrogen-bonded phosphate stretching mode (1219 cm−1) (Ashtarinezhad et al. 2014). The phosphate band at 1230 cm−1 can originate from the phosphate backbone of the nucleic acid (Boydston-White et al. 2006). Higher phosphate absorption in the induced fibroblast indicates low compact nucleic acid materials potentially associated with increased nucleic acid synthesis activity. For induced fibroblasts, higher absorption at 2950 cm−1 and 1230 cm−1 wavenumbers has been reported to indicate an accelerated cell growth and division (Kumar et al. 2014). It has been proven that highly compact DNA has a reduced absorbency rather than the presence of fewer DNA molecules (Whelan et al. 2013).

LC–ESI–MS/MS study

To support ATR-FTIR results, we analyzed the ion changes in earthworm tissues of the highest concentration application groups, i.e., H4 and H8 by LC–ESI–MS/MS analysis using Agilent 6460 Triple Quadrupole Mass Spectrometer with Mass Detection for a targeted method. The highest response to molecular ions was obtained. C2, H4, and H8 application groups were infused and analyzed by LC–ESI–MS/MS. When the system was operated under full scan conditions in the negative and positive ion modes, data were collected in the range of m/z 100–1300 with a scan time of 0.5 s. The ESI–MS/MS system was used with nitrogen as the collision gas and fragmentor voltage of 135 V. The results showed that ions found in the control (C2) and H8 application groups ([M + H] + ; m/z 542.3), ([M + H] + ; m/z 706.6), and ([M + H] + ; m/z 798.6) were not detected in the H4 treatment groups (Fig. 5). Differently, the ions ([M − H] + ; m / z 533.3) and ([M − H] + ; m / z 605.4) present in H4 and control (C2) were not found in the H8 application (Fig. 5). LC–MS results show that when the H4 application group was compared with control (C2) and H8, some ions ([M + H] + ; m/z 276.1), ([M + H] + ; m/z 325.2), ([M + H] + ; m/z 283.1), ([M + H] + ; m/z 233.1), and ([M + H] + ; m/z 140.1) were increased, while other ions ([M + H] + ; m/z 184.1), ([M + H] + ; m/z 132), and ([M + H] + ; m/z 138.1) were decreased (Fig. 5).

Fig. 5
figure 5

[M + H] + Heat maps of some ions detected in earthworm tissue samples. Colors represent increased (red) and decreased (blue) abundance, with the intensity reflecting the corresponding concentration, standardized by row. Control: C2 (hydrogen-rich water), Hg: H4 (40 µg/mL HgCl2, prepared with ultrapure water), Hg + H2: H8 (40 µg/mL HgCl2, prepared with hydrogen-rich water)

Mostly similar ions were detected in H4 and C2 groups in the H8 treatment group. Whereas there were ions in the H8 application group and control ([M + H] + ; m/z 542.3, m/z 706.6, and m/z 796.8), these ions were not shown in the H4 application group; however, ([M + H] + m/z 298) ion was shown (Fig. 5). Differently, the negative ions ([M − H] − ; m/z 233, m/z 533.3, and m/z 605.4) were also present in H4 and C2 but not detected in the H8 application (Fig. 6).

Fig. 6
figure 6

[M − H] − Heat maps of some ions detected in earthworm tissue samples. Colors represent increased (red) and decreased (blue) abundance, with the intensity reflecting the corresponding concentration, standardized by row. Control: C2 (hydrogen-rich water), Hg: H4 (40 µg/mL HgCl2, prepared with ultrapure water), Hg + H2: H8 (40 µg/mL HgCl2, prepared with hydrogen-rich water)

Different peak intensities of similar ions were observed in the H8 treatment group, H4, and C2 group. As seen in Fig. 5, the positive ions in the H4 treatment group ([M + H] + ; m/z 276.1, m/z 325.2, m/z 140.1, m/z 283.1, m/z 118.1, and m/z 233.1) were detected at higher levels than the H8 treatment group and the C2. Again, from negative ions ([M − H] − ; m/z 275.2 and m/z 307.2), ions were higher in group H4 (Fig. 6).

From the DNA bases, deoxyguanosine (dG) is the most easily oxidized DNA base and forms 8-oxo-2′-deoxyguanosine (8-oxo-dG) which has a very low redox potential due to its 8-position (Boiteux and Radicella 1999). The formed 8-oxo-dG can pair with adenine and lead to G:C → T:A transversion mutations if not repaired before replication (Hsu et al. 2004). The interaction of reactive oxygen species with DNA causes various modifications in 8-oxo-7,8-dihydro-2′- deoxyguanosine (8-oxodG) which has been extensively studied as a biomarker of oxidative stress (Lam et al. 2012). Oxidative DNA lesion measurements, which cause age-related macular degeneration (AMD), were measured by the LC–MS method in retinal pigment epithelial cells, and when the oxidative DNA lesion biomarker such as 8-oxo-2′-deoxyguanosine (8-oxo-dG) was examined, the ion exchange was m/z 284.0 and m/z 287.0 ions change, and the 8-oxo-dG levels were found to be increased (Ma et al. 2016). In this study, when the high-level application groups of mercury, i.e., H4 and H8 were compared with the controls, the ion exchange ([M + H] + ; m/z 283.1) representing the 8-oxo-dG level earthworms was higher in the H4 group than the H8 group (Figs. 5 and 7). However, we found that the same ion was closer to each other in C2 compared with the H8 application group (Figs. 5 and 7). Although hydrogen-rich water (HRW) has been used to prevent various oxidative stress-related diseases, the underlying mechanisms remain unclear. It has been determined that HRW alleviates mercury toxicity in plants by reducing Hg accumulation, prevents the formation of oxidative stress, and re-establishes redox homeostasis (Kosikowska et al. 2010; Cui et al. 2014). However, no studies are revealing the mitigating effect of HRW on mercury or any heavy metal toxicity in vertebrates or invertebrates. Otherwise, it has been reported that HRW consumption reduces the total ROS level and decreases the frequency of DNA damage due to ROS formation (Ohsawa et al. 2007; Suzuki et al. 2017; Zhang et al. 2017). It was also found that H2 restored the balance of the redox state and suppressed oxidative stress damage by decreasing ROS and MDA levels while increasing CAT, GSH, and SOD activity (Lu et al. 2020). Similarly, according to the findings of our study, it can be assumed that changes in the level of 8-oxo-dG, that is a marker of oxidative stress-induced DNA damage were decreased in the H8 earthworm group prepared with HRW due to the decrease of the mercury accumulation in the presence of hydrogen-rich water, and accordingly, the DNA damage was reduced. Membrane phospholipids consist of abundant polyunsaturated fatty acids. Thus, lipid peroxidation can occur in both nuclear and mitochondrial membranes. Especially in the inner mitochondrial membrane, the mitochondrial electron transfer system can easily generate superoxide anions and peroxidize the membranous lipid (Ježek and Hlavatá 2005; Henderson et al. 2009). The formed lipid peroxides are thought to easily attack mitochondrial DNA because they are located close to the inner membrane, and it produces 8-oxo-dG by induction of lipid peroxidation in mitochondria (Hruszkewycz and Bergtold 1990). It has been reported that DNA produces 8-oxo-dG when mixed with peroxidizing lipids (Park and Floyd 1992). In this case, we can assume by the support of the above-cited reports that in mercury solutions prepared with hydrogen-rich water, there is less lipid peroxidation and less formed 8-oxo-dG. Besides, when the bands (≠ 2 and ≠ 3) related to lipid materials were evaluated in FTIR results (Figs. 2 and 4), the H8 application did not differ from C2 and H4 application groups. But the decrease (1454 cm−1) in earthworm tissues of the H8 group was less than the H4 group when the band of 1454 cm−1 (≠ 7) (Fig. 4) was compared with the C1 and C2 control groups. It has been reported that the peak at 1457 cm−1 formed from the toxicity of xenobiotics in tissues showed weak DNA and RNA peaks due to vibration of methyl and methylene protein and lipid groups (Dhakshinamoorthy et al. 2017). In this case, it was thought that DNA and RNA might be less affected in earthworms injected with mercury solutions prepared with hydrogen-rich water. Molecular hydrogen has been proven to exhibit a mild but effective antioxidant activity by rapid diffusion into tissues and cells (Ohsawa et al. 2007). Although molecular hydrogen has been characterized as an effective antioxidant, it shows non-side effects, and it is mild enough to not disrupt metabolic and redox reactions nor affect the mild ROSs that play a role in the cell signaling system (Salganik 2001; Sauer et al. 2001; Liu et al. 2005). In cells stimulated with xenobiotics, mitochondrial membrane permeability can increase, and intense mitochondrial reactive oxygen species (mtROS) can be produced with potential oxidation of mitochondrial DNA (mtDNA) (Shimada et al. 2012; Guo et al. 2019; Zhang et al. 2020b). Removal of mtROS by molecular hydrogen has been reported to reduce the formation of oxidized mitochondrial DNA (Ren et al. 2016). Similarly, studies showed that molecular hydrogen can prevent the reduction of mitochondrial membrane potential, protect mitochondria from •OH radicals, prevent the decrease in the cellular level of ATP synthesized in mitochondria, and protect mitochondria and nuclear DNA (Ohsawa et al. 2007; Ohta 2011; Xin et al. 2014). Molecular hydrogen is thought to function both as a radical scavenger against oxidative stress in cells and as a mitohormetic effector in moderate mitochondrial stress (Murakami et al. 2017). In the sugar-phosphate DNA backbone region, 1300–800 cm−1, more than six bands come into focus: asymmetric and symmetric PO2 stretching mode at 1235 and 1089 cm−1, respectively, sugar-phosphate stretching band at 1070 cm−1, deoxyribose stretching mode at 966 cm−1, and A and B form markers situated at about 860 cm−1 and 835 cm−1, respectively (Tsuboi, 1970). In addition, the absorption bands 1150 cm−1 and 1020–1025 cm−1 can be assigned to the C–O bond of glycogen and other carbohydrates and are significantly overlapped by DNA (Heidarpoor Saremi et al. 2021). Some ions, which are important in the DNA structure, affect them and cause vibration and rotational movements of the above-mentioned bonds. In our study, it was determined that there was a decrease in the 1170−1 and 1047 cm−1 bands in the H3 group compared to the H7 group in the range of 1300–800 cm−1. In this case, we can assume that the changes in the bonds affect less the DNA structure of the H7 group prepared with HRW and that HRW could alleviate the negative effects of mercury on DNA. The above-cited studies explain that the ion exchange and FTIR spectra band area value changes can give information about DNA, lipid, and protein in both FTIR and LC–ESI–MS/MS results, which show that hydrogen-rich water exhibits protective effects on earthworms in the case of mercury toxication in the present study.

Fig. 7
figure 7

LC–ESI–MS/MS analysis of representative multiple reaction monitoring chromatograms of Control [M − H]:C2, Control [M + H]:C2, Hg [M − H]/[M + H]:H4, and Hg + H2 [M − H]/[M + H]:H8-earthworm tissues sample at 48 h after injection administration. Control: C2 (hydrogen-rich water), Hg: H4 (40 µg/mL HgCl2, prepared with ultrapure water), Hg + H2: H8 (40 µg/mL HgCl2, prepared with hydrogen-rich water)

The reason behind the absence of an obvious effect of HRW on the detoxification of mercury in earthworms could be attributed to the short application period of HRW and the high toxicity of mercury. In our recently published article, we have found that HRW could alleviate the effects of nickel toxicity in an earthworm model (Köktürk et al. 2021). In the previous study, a 2-week treatment with HRW was performed; however, in the present study, the HRW treatment was carried out only at the injection time of mercury. This shows that HRW would have exhibited potent and clear effects towards a heavy metal toxification if the HRW treatment time was longer like the 2-week HRW treatment performed in earthworms for the detoxification of nickel (Köktürk et al. 2021). To verify this hypothesis, it is important to extend the HRW treatment to cover a long period and perform additional analyses like immunohistochemical and immunofluorescence measurement.

Molecular hydrogen is known for its apparent inertness towards metals and nometals due to its very high dissociation energy, and it can reduce the oxides of most metals and many metallic salts to the metals only at high temperatures and pressures (William Lee Jolly 2020). As the physiological conditions of worms and assays (temperature and pressure) were not appropriate to this reaction between mercury chloride and molecular hydrogen, we assumed the absence of changes in the levels of mercury chloride during assays. However, the determination of mercury in water, HRW, and worm tissue samples forms a limitation of the study and should be also the topic of further study. The exact mechanism of the hydrogen effect on mercury and its chemical form change should form also an interesting issue of further study. Moreover, the study of the alleviation effects of HRW on DNA by exploring some additional markers in the DNA repair mechanism should also form another research subject.

Conclusion

The present work evaluated the use of hydrogen-rich water in terms of the detoxification purpose of mercury in the earthworm model, and it could show that hydrogen-rich water could mitigate the effects of toxic substances such as mercury. This study revealed that although hydrogen-rich water did not completely alleviate mercury toxicity, the results of ATR-FTIR differed from that of the normal water, i.e., without molecular hydrogen in terms of some ions and molecules. For the first time, biological monitoring of mercury-related oxidative stress could be achieved in earthworm tissues, and the formation of 8-oxodG was rapidly identified using LC–ESI–MS/MS method. These results are important for investigating innovative, friend-to-environment, and cost-effective methods for decontaminating environments contaminated with heavy metals. Additionally, this study provides an alternative approach to investigating the protective effects of hydrogen-rich water in ecotoxicological situations. These results provide a green method solution for the treatment of the problem of heavy metal toxification phenomenon in industrial regions.