Introduction

Oxidative stress has been suggested as the main culprit in the pathogenesis and progression of most neurodegenerative diseases such Alzheimer’s and Parkinson diseases (Kim et al. 2015). Oxidative stress arises when there is an increased generation of free radicals, with concomitant depletion in the ability of the body’s endogenous antioxidant system to mop them. This leads to redox imbalance. The brain’s dependence on oxygen (O2) and high consumption of glucose makes it highly susceptible to oxidative stress as leaked O2 has been implicated in the generation of free radicals such as superoxide anion (O2.-), hydrogen peroxide (H2O2) and hydroxyl (.OH) (Cobley et al. 2018; Patel 2016). The brain’s low endogenous antioxidant system and high content of polyunsaturated fatty acids, with concomitant redox-active metal load such as iron also contributes to its susceptibility to oxidative damage (Butterfield et al. 2001; Huang et al. 2004; Patel 2016).

Physiologically, iron plays an important role in the normal functioning of the brain as it acts as co-factors for most neurotransmitter synthetic enzymes such as glutamate decarboxylase and glutamate transaminase, tryptophan hydroxylase, tyrosine hydroxylase, and monoamine oxidases A and B (Hasegawa et al. 1999; Hidalgo and Núñez 2007; Li 1998). Its role in normal neurological functions as well as development of cognitive functions has also been reported (Grantham-McGregor and Ani 2001; Hidalgo and Núñez 2007). However, high concentrations of iron have been implicated in the exacerbation of neurodegeneration via excessive generation of reactive oxygen species (ROS). Iron facilitates the breakdown of H2O2 to •OH via the Fenton’s reaction. Also, the Haber-Weiss reaction readily makes iron available for the Fenton’s reaction as it involves the reduction of Fe (III) to Fe (II) in the presence of O2•− (Latunde-Dada 2017).

Plant phenolics have been reported for their neuroprotective properties as well as in the treatment and management of other diseases (Soobrattee et al. 2005). These properties have been attributed to the potent antioxidant properties of polyphenols in relations to their chemical structure (Erukainure et al. 2018; Soobrattee et al. 2005). Vanillin (4-hydroxy-3-methoxybenzaldehyde) and its acid, vanillic acid (4-hydroxy-3-methoxybenzoic acid) are amongst the common phenolics well studied for their medicinal properties. These phenolics are major components of extracts of the vanilla bean and pod (Kumar et al. 2012), and are commonly utilized as flavours in foods, cosmetics and drugs. Their presence has also been reported in other plants such as Amburana cearensis (Calixto-Campos et al. 2015).

Vanillin is a benzaldehyde with methoxy and hydroxy functional groups at positions 3 and 4 respectively. It has been reported for its potent antioxidative, anti-apoptotic, neuroprotective and anticancer activities (Bezerra et al. 2016; Dhanalakshmi et al. 2015, 2016).

Vanillic acid, an oxidized form of vanillin is a monohydroxybenzoic acid which consists of a 4-hydroxybenzoic acid substituted by a methoxy group at position 3. It has been reported for its antioxidative, antihypertensive and anti-inflammatory activities (Calixto-Campos et al. 2015; Kumar et al. 2011; Tsuda et al. 1994).

Despite these studies, there is however a dearth of information on the comparative effect of vanillin and vanillic acid on dysregulated metabolites and pathways in redox-metal aggravated neurotoxicity. This study was thus aimed at investigating the comparative and protective effect of these phenolics on oxidative imbalance, cholinergic and nucleotide-hydrolyzing enzyme activities, dysregulated metabolic pathways in Fe2+-induced neurotoxicity in brain tissues. Their cytotoxicity was investigated in hippocampal neuronal cell lines (HT22). Their molecular interactions with the studied enzymes were investigated in silico.

Materials and methods

Phenolics

Vanillin (99.1%; Mw: 152.15 g/mol) and vanillic acid (98.9%; Mw: 168.15 g/mol) were purchased from Sigma-Aldrich, Johannesburg, South Africa. Stock solutions of 1 mg/mL were prepared with each compound, from which different working concentrations consisting of 15, 30, 60, 120 and 240 μg/mL were prepared. The choice of dosage was based on previous ex vivo studies on phenolics by Salau et al. (2019).

Animals

Three male albino rats (Sprague Dawley strain; 180–200 g) obtained from the Biomedical Research Unit (BRU), University of KwaZulu-Natal, Durban, South Africa were sacrificed by euthanizing with halothane after overnight fasting. Their brains were harvested, rinsed in 0.9% NaCl solution and thereafter homogenized in 50 mM sodium phosphate buffer (pH 7.5; 10% Triton X-100). They were thereafter centrifuged for 10 mins at 15,000 rpm and 4 °C, and the supernatants collected for ex vivo studies.

The study was done under the approved guidelines of the Animal Ethics Committee of the University of KwaZulu-Natal, Durban, South Africa (Protocol approval number: AREC/020/017D).

Induction of neurotoxicity ex vivo

Two hundred microliters (200 μL) of the brain tissue supernatant was incubated with an equal volume of the different concentrations (15–240 μg/mL) of vanillin and vanillic acid, and 60 μL of 0.1 mM FeSO4 for 30 mins at 37°C in 5% CO2 incubator (Erukainure et al. 2017a, b). The negative control (untreated) consisted of a reaction mixture without any of the phenolics, while the normal control consisted of tissues without the phenolics and FeSO4.

Anti-oxidative activity

The incubated tissue samples were subjected to analysis for reduced glutathione (GSH) level (Ellman 1959), catalase and superoxide dismutase (SOD) activities (Chance and Maehly 1955; Kakkar et al. 1984) and malondialdehyde (MDA) level (Chowdhury and Soulsby 2002) in order to determine oxidative stress.

Nitric oxide level

The nitric oxide (NO) levels of the incubated tissue samples were analyzed using the Griess method as previously described (Erukainure et al. 2019; Tsikas 2005).

Cholinergic enzymes activities

The incubated tissue samples were analyzed for acetylcholinesterase (Ellman et al. 1961) and butyrylcholinesterase (Adefegha et al. 2017) activities in order to determine the cholinergic enzymes activities.

Purinergic enzymes activities

The incubated tissue samples were analyzed for purinergic enzyme activities by determining the adenosine triphosphatase (ATPase) (Adewoye et al. 2000; Erukainure et al. 2017a) and ecto-nucleoside triphosphate diphosphohydrolase (ENTPDase) (Akomolafe et al. 2017) activities.

Metabolite extraction and profiling

The tissues were subjected to incubation as described above but was incubated overnight at 37 °C. They were thereafter subjected to metabolite extraction as previously described (Chan et al. 2013; Erukainure et al. 2017b). The extracted metabolites were then subjected to LC-MS analysis. The metabolites were identified by directly comparing mass spectral data against those of the Human Metabolome Database (HMDB) (Wishart et al. 2012).

Metabolic pathway analysis

The MetaboAnalyst 4.0 online server was employed for metabolic pathway analysis of the identified metabolites (Chong et al. 2018). Metabolites that showed significant change were mapped with the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways using the HMDB numbers.

MTT assay

The cytotoxic effect of vanillin and vanillic acid on hippocampal neuronal cell lines (HT22) was evaluated using the standard MTT [3-(4, 5-dimethyl thiazole-2-yl)-2,5-diphenyltetrazolium bromide] colorimetric assay (Mosmann 1983). Briefly, HT-22 cells were provided by Prof. David Schubert from Salk Institute, San Diego, USA. The cells were incubated with the different concentrations of vanillin and vanillic acid in a humidified atmosphere with 5% CO2 for 48 h at 37°C. A 200 μL of MTT solution (5 mg/mL; in PBS) was added to each well, after replacing the medium with a fresh one. The cells were thereafter incubated at 37°C for 4 h. The medium was discarded, and replaced with 200 μL of DMSO. Absorbance was read at 570 nm.

In silico prediction of BBB permeability and oral lethal dose toxicity

The SwissADME online server was used in predicting the ability of vanillin and vanillic acid to cross the blood brain barrier (BBB) in silico (Daina et al. 2017). The PROTOX II online server was used in predicting the oral lethal dose toxicity of the compounds in silico (Banerjee et al. 2018).

Statistical analysis

Data were analyzed using one-way analysis of variance (ANOVA) and presented as mean ± Standard Deviation (SD). Significant differences between means were obtained at p < 0.05 using the Tukey’s HSD-multiple range post-hoc test. Statistical analyses were done using IBM Statistical Package for the Social Sciences (SPSS) for Windows, version 23.0 (IBM Corp., Armonk, NY, USA).

Results

As shown in Fig. 1a–d, incubation of brain tissues with FeSO4 led to significant (p < 0.05) depletion of GSH level, SOD and catalase activities while elevating MDA level. These levels and activities were significantly (p < 0.05) reversed on treatment with vanillin and vanillic acid to levels almost indistinguishable from the normal control, with a dose-dependent increase observed for the SOD activity (Fig. 1b). Based on the IC50 values (Table 1), vanillin had a higher antioxidative potency compared to vanillic acid.

Fig. 1
figure 1

Effect of vanillin and vanillic acid on a GSH level, b SOD activity, c catalase activity, and d MDA level in oxidative brain injury. Value = mean ± SD; n = 3. *Statistically significant (p < 0.05) compared to untreated tissue; #statistically significant (p < 0.05) compared to normal tissue

Full size image
Table 1 IC50 values of studied biological activities of vanillin and vanillic acid
Full size table

There was a significant (p < 0.05) elevation of NO level in brain tissues incubated with only FeSO4 as depicted in Fig. 2. Treatments with vanillin and vanillic acid significantly (p < 0.05) depleted the level dose-dependently, with vanillin showing a better activity.

Fig. 2
figure 2

Effect of vanillin and vanillic acid on NO level in oxidative brain injury. Value = mean ± SD; n = 3. *Statistically significant (p < 0.05) compared to untreated tissue; #statistically significant (p < 0.05) compared to normal tissue

Full size image

As depicted in Fig. 3a, b, incubation of brain tissues with FeSO4 significantly (p < 0.05) elevated the activities of acetylcholinesterase and butyrylcholinesterase respectively. These activities were significantly (p < 0.05) reduced in a dose-dependent manner on treatments with vanillin and vanillic acid. The low IC50 value of vanillic acid compared to vanillin for acetylcholinesterase activities (Table 1), indicates the former to be more potent. While the low IC50 value of vanillin for butyrylcholinesterase indicates a higher potency compared to vanillic acid.

Fig. 3
figure 3

Effect of vanillin and vanillic acid on a acetylcholinesterase and b butyrylcholinesterase activities in oxidative brain injury. Value = mean ± SD; n = 3. *Statistically significant (p < 0.05) compared to untreated tissue; #statistically significant (p < 0.05) compared to normal tissue

Full size image

Incubation of brain tissues with FeSO4 led to significant (p < 0.05) depletion of ATPase activity with concomitant elevation of ENTPDase activity as shown in Fig. 4a, b respectively. There was a dose-dependent increased ATPase activity on treatment with vanillin and vanillic acid. Treatments with both compounds however led to slight reduction in ENTPDase activity.

Fig. 4
figure 4

Effect of vanillin and vanillic acid on a ATPase and b ENTPDase activities in oxidative brain injury. Value = mean ± SD; n = 3. *Statistically significant (p < 0.05) compared to untreated tissue; #statistically significant (p < 0.05) compared to normal tissue

Full size image

Incubation of brain tissues with FeSO4 led to depletion of Riboflavin cyclic-4′,5′-phosphate, TG(24:1(15Z)/20:0/20:3n6), Ganglioside GD3 (d18:0/26:0), Deoxyadenosine triphosphate, Xanthosine 5-triphosphate, Adenosine tetraphosphate, Ribose 1,5-bisphosphate, Guanosine 3′,5′-bis(diphosphate), D-Fructose 2,6-bisphosphate, Ganglioside GM2 (d18:1/25:0), P1,P4-Bis(5′-uridyl) depletion of the metabolites present in the normal brain tissues (normal control). Treatment with vanillin led to the replenishment of Riboflavin cyclic-4′,5′-phosphate, Ganglioside GD3 (d18:0/26:0), Xanthosine 5-triphosphate, Adenosine tetraphosphate, Ribose 1,5-bisphosphate, Guanosine 3′,5′-bis(diphosphate), and D-Fructose 2,6-bisphosphate with concomitant addition of Thymidine 5′-triphosphate, TG(24:1(15Z)/20:0/20:3n6), 3-Phosphoadenylylselenate, Deoxyadenosine triphosphate, Ascorbic acid-2-sulfate, DOPA sulfate, Phosphatidylinositol-3,4,5-trisphosphate, dTDP-D-glucose, Phosphoribosyl-ATP, Ganglioside GD3 (d18:0/23:0), Guanosine tetraphosphate adenosine and Inosine 5′-diphosphate. Vanillic acid replenished TG(24:1(15Z)/20:0/20:3n6), Ganglioside GD3 (d18:0/26:0), Deoxyadenosine triphosphate, Adenosine tetraphosphate, Ribose 1,5-bisphosphate, Guanosine 3′,5′-bis(diphosphate), D-Fructose 2,6-bisphosphate and Ganglioside GM2 (d18:1/25:0), with concomitant addition of Adenosine triphosphate, ADP-ribose 1″-2″ cyclic phosphate, 3’-Monoiodo-L-thyronine, and Ganglioside GD2 (d18:0/18:0).

Incubation of brain tissues with FeSO4 led to inhibition of the purine metabolism and pentose phosphate pathways as compared to the normal brain tissues (Fig. 5a, b). Treatment with vanillin led to reactivation of the purine metabolism and pentose phosphate pathways, while concomitantly activating pathways for Selenoamino acid and histidine metabolisms (Fig. 5c). Treatment with vanillic acid also led to the reactivation of the purine metabolism and pentose phosphate pathways but did not activate any additional pathway (Fig. 5d). A schematic presentation of the identified pathways is shown in Fig. 6.

Fig. 5
figure 5

Pathway analysis of a normal; b untreated; c vanillin – treated; and d vanillic acid – treated brain tissues

Full size image
Fig. 6
figure 6

Schematic presentation of identified pathways. Maroon color represents inactivated pathways in untreated brain tissues; green color represents activated pathways in vanillin treated brain tissues; while black color represents pathways common to normal, untreated and treated brain tissues

Full size image

Table 3 shows the molecular docking for vanillin and vanillic acid with human SOD, ATPase, acetylcholinesterase, and catalase. The docking protocol displayed hydrogen bonding, Pi-Pi stacked, Van der Waals interactions. The O-H and Oxygen groups of vanillin and vanillic acid formed hydrogen bonding with the amino acid residues SER114, ARG112, PHE334 and HIS75. while others amino acid residues formed Pi-Pi stacked interaction with TRP84 and Van der Waals interactions. Both compounds showed the strongest binding affinity with catalase as depicted by their respective binding energy (Table 3) and presented in Fig. (Fig. 7a–d).

Fig. 7
figure 7

a 3D structure of vanillin complex with catalase (highest binding affinity); b 2-D representations displaying the interactions of vanillin with amino acid residues; c 3D structure of vanillic acid complex catalase (highest binding affinity). d 2-D representations displaying the interactions of vanillic acid with amino acid residues

Full size image

Incubation of HT-22 cells with vanillin and vanillic acid showed little or no effect on the cell’s viability as depicted in Fig. 8. The high IC50 values of the compounds further indicates their non-cytotoxic activity (Table 2).

Fig. 8
figure 8

Cytotoxic effect of vanillin and vanillic acid on HT-22 cells. Value = mean ± SD; n = 3

Full size image
Table 2 LC-MS identified metabolites in experimental brain tissues
Full size table

In silico prediction of the abilities of the compounds to cross the BBB predicted only vanillin to be permeable (Table 4). In silico toxicity study predicted both compounds to be in the toxic class of 4, with LD50 values of 1000 and 2000 for vanillin and vanillic acid respectively.

Discussion

Oxidative stress has been recognized as a common event in the etiology, pathogenesis and progression of many neurodegenerative diseases such as Parkinson’s, Alzheimer’s, Huntington’s diseases, and amyotrophic lateral sclerosis (ALS) (Belaidi and Bush 2016; Niedzielska et al. 2016). It is exacerbated in the brain owing to its high concentration of unsaturated fatty acids and redox-active metal load. Studies have reported the contribution of phenolics in the treatment and management of neurodegenerative diseases, with their potent antioxidative properties playing an important role (Bastianetto et al. 2006; Szwajgier et al. 2017). Ex vivo and in silico studies have been utilized in investigating the toxic effect of iron on brain metabolism and the protective potentials of phenolics (Salau et al. 2019). Vanillin and vanillic acid are major flavouring ingredients in the pharmaceutical, food and cosmetic industries, evaluation of their neuroprotective properties is therefore of great importance. In this study, the neuroprotective effects of vanillin and vanillic acid were investigated in Fe2+ exacerbated oxidative damage in brain tissues.

The depleted GSH level, SOD and catalase activities indicates in the untreated brain tissues (Fig. 1a–c) depicts a suppression of the tissues’ endogenous antioxidant system. Thus, indicating an occurrence of oxidative stress which can be attributed to iron toxicity via the Fenton’s and Haber Weiss reactions (Latunde-Dada 2017). This corroborates other studies on the induction of oxidative stress by Fe2+ (Oyebode et al. 2018; Sanni et al. 2019). The elevated levels of MDA in the untreated tissues (Fig. 1d) portrays an occurrence of lipid peroxidation. This can be attributed to the low catalase activity (Fig. 1c) to breakdown H2O2 to H2O and O2, leading to cellular accumulation of. OH. This accumulation coupled with the high unsaturation of the brain lipids will lead to peroxidative attack. The low SOD activity (Fig. 1b) also portrays an elevated level of O2.-. Elevated NO levels have been reported to initiate a reaction between NO and O2.- to generate peroxynitrite (ONOO−), a potent radical which has been implicated in the proinflammation. The elevated NO level (Fig. 2) in the untreated thus indicates an occurrence of proinflammation on induction of oxidative damage. The reversed activities and levels of these biomarkers in tissues treated with vanillin and vanillic acid (Figs. 1 and 2), therefore insinuate an antioxidative and anti-proinflammatory effects of the phenolics in oxidative brain injury. This further corroborates previous studies on the antioxidant and anti-inflammatory activities of both compounds (Bezerra et al. 2016; Kumar et al. 2011).

Studies have correlated the elevated activities of the cholinergic enzymes, acetylcholinesterase and butyrylcholinesterase with the pathogenesis and progression of neurodegenerative diseases (Greig et al. 2002; Maya et al. 2016). Their elevated activities on in the untreated tissues (Fig. 3a, b) thus indicates a cholinergic dysfunction which portrays an occurrence neurodegeneration. The activities of these enzymes have been reported to be enhanced in exacerbated oxidative stress (Ćupić Miladinović et al. 2018; Melo et al. 2003), thus implying the elevation in the present study may be attributed to the induced oxidative damage (Fig. 1). The decreased activities on treatment with vanillin and vanillic acid insinuates an ability of the phenolics to suppress oxidative-enhanced cholinergic dysfunction, thus indicating a neuroprotective effect which correlates with the antioxidant activities (Fig. 1). The binding energies and molecular interaction with acetylcholinesterase as revealed by molecular studies (Table 1), portrays a mechanism by which the phenolics may interact with acetylcholinesterase to inhibits its activities. This however warrants further studies.

The roles of purinergic enzymes in the mediation of several physiological functions including neuronal functions has been reported (Ademiluyi et al. 2016). These enzymes mediate the phospho-hydrolysis of Adenosine triphosphate (ATP) and Adenosine monophosphate (AMP) to produce the adenosine nucleotide (Akinyemi et al. 2017). The decreased ATPase activity (Fig. 4a) in the untreated brain tissues (Fig. 4a) portrays a decreased level of the adenosine, ADP which may be attributed to the inactivated purine metabolism pathway (Figs. 5b and 6) and its metabolites: Xanthosine 5-triphosphate, Adenosine tetraphosphate, and Guanosine tetraphosphate adenosine (Table 2). Interestingly, there was an elevation in ENTPDase activity in the untreated tissues (Fig. 4b). The elevation may be a console for the increased hydrolysis of ATP by ATPase. Alteration in the activities of the studied purinergic enzymes suggests an impaired neuronal function. Thus, the reversed enzymatic activities in tissues treated with vanillin and vanillic acid further indicates the neuroprotective effect of these phenolics.

Depletion of the metabolites, Riboflavin cyclic-4′,5′-phosphate and Ribose 1,5-bisphosphate in the untreated brain tissues (Table 3) correlates with the inactivated pentose phosphate pathway (Figs. 5b and 6). The pentose phosphate pathway has been reported for its involvement in glutathione metabolism via NADPH generated in the pathway (Deng et al. 2019; Williams and Ford 2004). Implication of this pathway has been implicated in the pathogenesis of oxidative stress, owing to impairment in the generation of GSH (Deng et al. 2019). Thus, the depleted GSH level in the untreated tissues (Fig. 1a) correlates with inactivation of the pentose phosphate pathway. The restoration of the depleted metabolite and reactivation of the pathway in brain tissues treated with vanillin and vanillic acid (Figs. 5c, d and 6) further indicates an antioxidant-protective effect of these phenolics, which correlates with the increased GSH levels. Activation of the selenoamino acid metabolism pathway by vanillin (Figs. 5c and 6) portrays a high antioxidative potency of vanillin over vanillic acid, as selenoproteins have been reported for their antioxidative and anti-inflammatory roles (Schrauzer 2000). This pathway depends on glutathione metabolism, thus dependent on the pentose phosphate pathway.

Table 3 Calculated binding energies in kcal mol−1 of compounds obtained using Autodock
Full size table

The activation of histidine metabolism pathway by vanillin (Figs. 5c and 6) further insinuates the neuroprotective effect of the phenolic, as histidine is an essential amino acid and a precursor of histamine (Yoshikawa et al. 2014). Histamine is a neurotransmitter in the brain and has been linked to several neurological functions such as the sleep-wake cycle, stress response, and regulation of appetite. Its dysfunction has been implicated in several neurological diseases such as depression, Alzheimer disease, narcolepsy, and Parkinson’s disease (Haas et al. 2008; Mehan et al. 2017; Yoshikawa et al. 2014).

In order to evaluate the safety of vanillin and vanillic acid on brain cells, their cytotoxic effect was studied in hippocampal cells. Their non-toxic effect on normal brain cells were confirmed by the little or no cytotoxic activity in HT22 cells (Fig. 8). This also correlates with the predicted LD50 values and toxicity class (Table 4), which indicates the safety of the phenolics when swallowed (Banerjee et al. 2018; Drwal et al. 2014).

Table 4 In silico prediction of BBB permeability and toxicity of vanillin and vanillic acid
Full size table

The neuroprotective effect of vanillin was further depicted by its predicted ability to cross the BBB (Table 4). However, the predicted non-permeability of vanillic acid across the BBB may portray that the neuroprotective effect of vanillic acid maybe indirective effect.

Conclusion

Taken together, vanillin and vanillic acid confers a neuroprotective effect on oxidative brain damage by exacerbating antioxidative status and ATPase activity, while inhibiting cholinergic enzymatic activity and reactivating oxidative-deactivated purine metabolism and pentose phosphate pathways. The activation of selenoamino and histidine metabolisms respective pathways by vanillin insinuates its neuroprotective potency compared to vanillic acid. The phenolics are safe and non-toxic on normal brain cells. This study however represents data from ex vivo studies which may be a limitation of the present study. Therefore, in vivo study is being recommended to further validate the neuroprotective mechanism of the studied phenolics.