Background
Oxidative stress is defined as the imbalance between the levels of reactive oxygen species (ROS), reactive nitrogen species and endogenous antioxidants. Any alteration of the homeostasis leads to an increased production of free radicals that cannot be counteracted by detoxifying mechanisms [
1]. Free radicals interact with proteins and genes, which leads to tissue damage. Cardiovascular diseases, neuronal degeneration, and cancer are the result of such imbalance. For example, it is known that the hyperglycaemia observed in diabetes mellitus is responsible for the generation of oxidative stress [
2]. Oxidative stress is also involved in the aetiology and pathogenesis of several oral diseases, including the development of dental caries and periodontitis [
3]. For example, the expression of antioxidant genes has been observed in patients with periodontitis [
4]. Moreover, oral pre-cancerous lesions such as lichen planus and leukoplakia and oral squamous cell carcinoma have been found to be associated with oxidative stress and high levels of malondialdehyde (MDA) in serum, which is a marker of oxidative stress [
5,
6].
The antioxidant enzymes secreted in the saliva play a major role in the maintenance of oral health through the scavenging of reactive species such as hydrogen peroxide and superoxide anion and nitrogen reactive species, some of which are produced by the oral microflora [
7]. The main antioxidant enzymes present in saliva are superoxide dismutase, catalase, and glutathione peroxidase, which are mainly secreted by the salivary glands such as the submandibular ones [
8]. Submandibular glands produce approximately 65–70% of saliva, thus being the main source of the antioxidant enzymes of the oral cavity. Although the parotid glands are larger than submandibular glands, they only produce 20% of the saliva secreted. Therefore, any change in the secretory capacity of submandibular glands will lead to the development of oxidative stress-induced oral diseases. It is known that the structure and function of salivary glands is altered in diabetic patients [
9]. The amount of microorganisms in the oral cavity is controlled through appropriate oral hygiene, particularly in diabetic patients who are prone to infections. In this sense, mouthwashes can prevent and/or alleviate oral pathologies, acting as deodorant, antiseptic, disinfectant, analgesic, astringent and antioxidant.
It has previously been demonstrated that streptozotocin (STZ), a drug that induces the development of type 1 diabetes mellitus, generates an imbalance between oxidative and antioxidant systems by decreasing the secretion of peroxidase in submandibular glands [
10,
11] This is then, a suitable model to test the antioxidant activity of drugs.
Unlike synthetic antioxidants, which are known to cause severe adverse effects, natural antioxidants, when administered locally or systemically, would protect tissues of the oral cavity from the oxidative stress, acting as adjuvants in the treatment of oral diseases of oxidative origin.
Larrea divaricata Cav. (Zygophyllaceae) is a bush that grows in South America and is widely distributed in Argentina. This plant is used in folk medicine for its anti-inflammatory properties and is also known to have antitumoral and immunomodulatory activities and antimicrobial properties [
12‐
15]. The antioxidant compound nordihydroguaiaretic acid (NDGA) has already been reported to be present in this plant [
16]. It has previously been demonstrated that the aqueous extract of
Larrea divaricata (AE) stimulates the secretion of peroxidase in normal rat submandibular glands [
17]. The effect of the AE on the pro-oxidant/antioxidant homeostasis in salivary glands of normal rats has been studied elsewhere [
18]. However, the effect of the AE on the pro-oxidant/antioxidant homeostasis in submandibular glands subjected to the effects of an oxidantive stressor such as STZ remains to be studied.
The aim of this work was to determine the antioxidant activity of the AE prepared with the leaves of L. divaricata in a model of oxidative stress induced by STZ in submandibular glands. The levels of ROS, the degree of lipid peroxidation and protein oxidation, and the activity of antioxidant and pro-oxidant enzymes related to the metabolism of H2O2 and NO were determined in submandibular glands obtained from STZ-treated rats. The participation of the majority compound of the AE, namely NDGA, was also studied. It is hypothesized that the local administration of the AE as either a mouthwash or as a systemic antioxidant could control the damage induced by oxidative stress in soft oral tissues, particularly in diseases that feature oxidative stress such as diabetes.
Discussion
In this work the effect of Larrea divaricata AE was studied in a model of oxidative stress induced in submandibular glands 10 days after the administration of a single dose of STZ to female rats. The administration of the AE restored the pro-oxidant/antioxidant balance in the glands, thus demonstrating its antioxidant activity.
The NDGA content was determined in the AE by HPLC in order to calculate the concentration of this compound in the AE used in the experiments and to study the participation of NDGA in the overall effects exerted by the extract. The individual effect of NDGA was studied because it is the majority compound of the extract. It is noteworthy that other compounds such as flavonoids have been reported in L. divaricata; however, they are in lower quantities than NDGA. The peak observed at RT = 25 min in the HPLC chromatogram is a complex constituted by compounds such as vicenin, rutin and flavonoids measured as quercitrin.
The AE concentration used in the experiments had previously been determined in normal glands [
18,
29].
Concentration-response experiments were done with 100, 500 and 1000 μg/ml of AE. The AE was employed at 500 μg/ml, which corresponded to 1.5 μg/ml NDGA. These concentrations produced a maximum effect. The presence of NDGA in this AE has also been described elsewhere [
16].
The diabetic status induced by the administration of STZ was corroborated by the high levels of serum glucose, NO and MDA. It is known that the administration of a single dose of 60 mg/kg of streptozotocin triggers an autoimmune process that results in the destruction of the Langerhans’s islet beta cells, thus resulting in the development of clinical diabetes in a short time (2–4 days), and in the induction of oxidative stress [
10,
30]. In this context, the submandibular glands are also affected by the treatment with STZ. In fact, a weight decrease of 25%, as compared to control rats was observed in relation to the decrease in protein content. Moreover, signals of oxidative stress were observed in those glands having high levels of MDA, protein carbonylation and reduced glutathione. It is known that MDA reflects the effects of hydroxyl radical damage to cell lipids (lipid peroxidation) under oxidative stress conditions [
31].
The increase in the number of carbonyl groups (aldehydes and ketones), which are produced on protein side chains (especially of Pro, Arg, Lys, and Thr), has been observed not only in diabetes but also in Alzheimer’s disease, inflammatory bowel disease, and arthritis [
32].
Both the extract and NDGA were capable of reversing the oxidative stress generated in glands by decreasing the levels of MDA and the number of protein carbonyl groups but not those of gluthatione, which remained high, due to an increased synthesis rate. NDGA and the AE inhibited the generation of MDA by 95 and 84%, respectively. These results indicate that the effect of the extract was highly related to the presence of NDGA. On the other hand, the inhibition on protein carbonylation achieved by NDGA was stronger than that achieved by the AE (84% vs. 25%), thus suggesting that the AE has compounds that might counteract the effect exerted by NDGA.
The lipid peroxidation and protein carbonylation induced by the treatment with STZ could be related to the increase of H
2O
2 and NO induced by STZ. In fact, the treatment with STZ caused an increase in the levels of H
2O
2 and a decrease in O2
.- (Fig.
4 d and e). It is known that H
2O
2 is generated upon the SOD-catalysed dismutation of O
2 (Fig.
5b). STZ could be accelerating the transformation of O
2.- into H
2O
2 through the activation of SOD, thus explaining the high levels of H
2O
2 and the low levels of O
2.-[
33]. On the other hand, STZ deceased the activity of Px and down-regulated its expression (Fig.
5 a and c, Table
1). This finding would also explain the high levels of H
2O
2 found in STZ-treated animals.
Moreover, Klotz et al. (2005) have demonstrated that under oxidative stress conditions, ROS such as H
2O
2 can activate the insulin-like growth factor 1 receptor (IGF1-R)- phosphatidylinositol 3-kinase (PI3K)-AKT pathway, which inhibits the synthesis of forkhead box (FoxO) protein, which is a transcriptional inducer of antioxidant enzymes [
34]. The inhibition of this factor could also explain the down-regulation of Px and SOD exerted by STZ (Fig.
5c and d). It has also been demonstrated that ROS, e.g. H
2O
2 activates Cu-Zn-SOD in a concentration-dependent manner by oxidizing the thiol groups of the enzyme during post-transcriptional regulation. The oxidation of its critical thiol group is necessary for the activation of SOD [
35]. Taking into account these data, it could be hypothesised that the SOD activation is the result of the increase in the H
2O
2 levels induced by STZ (Fig.
5b).
Both the AE and NDGA exerted an antioxidant activity by decreasing not only the levels of H
2O
2 but also those of O
2.-. (Fig.
4d and e). The decrease in the H
2O
2 levels was achieved by the induction and activation of Px (Fig.
5a and c). Furthermore, the decrease in O
2.- levels could be related to the NDGA-promoted inhibition of the NADPH oxidase complex, which is involved in the synthesis of O
2.-[
36]. However, the latter hypothesis remains to be tested. Beside NADPH-oxidases, cyclooxygenases/lipoxygenases (5-LOX) are generally recognized as the principal physiological sources of O
2.-, which, in turn, dismutates into H
2O
2 [
37]. It is known that 5-LOX catalyses the production of leukotrienes and ROS from arachidonic acid [
38]. The inhibitory capacity of NDGA on LOX would explain the decrease in the levels of O
2.-. The decrease of H
2O
2 induced by both the extract and NDGA could, in turn, allow FoxO protein to induce the expression of PX and SOD (Fig.
5c and d Table
1). However, the reduction of H
2O
2 levels exerted by the extract and NDGA could not be sufficient to activate Cu-Zn-SOD, whose activity was decreased (Fig.
5b). The same results were obtained with the extract and NDGA in rats not treated with STZ. Both the extract and NDGA were capable of decreasing the levels of H
2O
2 and O
2.- under basal conditions through the activation of Px and the decrease of SOD activity. In line with these results, So Yong Kim et al. (2008) have demonstrated that, as a LOX inhibitor, NDGA scavenges intracellular ROS, inhibits the tumor necrosis factor α (TNF-α)-induced ROS accumulation, blocks the TNF-α-induced NF-κB activation, and inhibits LPS-induced TNF-α production and NF-κB activation in ovalbumin-induced asthma in mice [
39].
On the other hand, STZ was capable of increasing the NO levels related to the up-regulation of iNOS. It is known that iNOS generates both O
2.- and NO, leading to peroxynitrite-mediated cell injury. Therefore iNOS is involved in many diseases associated with inflammation and oxidative stress. It is known that H
2O
2 can regulate the expression of iNOS. In fact, it has been demonstrated that H
2O
2 stimulates the expression of iNOS in a concentration-response relationship and that H
2O
2 can enhance the expression of iNOS induced by cytokines through NFκ-B activation [
40,
41]. Therefore, by increasing H
2O
2 levels (Fig.
4d), STZ could induce the expression of iNOS (Table
2 and Fig.
6); such effect was counteracted by the extract and NDGA, which decreased H
2O
2 levels (Fig.
6 and Table
2). The probable mechanism of action of the extract and NDGA is summarized in Fig.
7.
Even though glands obtained from STZ-treated animals presented a lower weight, as compared to controls, histologic changes were observed in the architecture of neither acini nor ducts, probably due to the short time of the STZ treatment. In a previous study Anderson et al. (1994) only observed a reduction in the acinar cell size after a long treatment (4–6 months) with STZ [
42]. Neither degenerative changes in the parenchymal, nor vacuolation, focal loss of salivary architecture or signals of local necrosis or pyknotic nuclei were observed after any treatment (STZ, AE or NDGA), thus demonstrating that neither the extract nor NDGA were cytotoxic at the concentrations assayed.
Conclusion
The extract of L. divaricata was capable of reversing the oxidative stress in the submandibular glands induced by the administration of STZ to female rats. As none of the compounds used in this work had cytotoxic effects on the gland cells, the antioxidant sources remained unaltered, thus being able to prevent the development of oral oxidative diseases.
Although NDGA was demonstrated to be involved in this reversal, the activity of other compounds cannot be ruled out.
The molecular mechanism by which the extract would modulate the oxidative stress would include the modulation of both the activity and the expression of antioxidant and pro-oxidant enzymes, which modulate the levels of oxygen and nitrogen reactive species, preventing MDA formation and protein carbonylation.
Therefore, the extract could be used as a local or systemic preventive agent against oral diseases caused by oxidative stress in diabetic subjects. Further studies are needed to confirm this hypothesis.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.