Phytochemical study
The dichloromethane extract from aerial parts of S. melissiflora furnished four known compounds (1–4), while the ethanol extract yielded 5 as main constituent (Fig. 1). These compounds were identified by analyses of their NMR data and comparison with literature.
Compounds 1 and 2 were isolated in mixture. Their NMR data revealed typical signals of triterpenes. The basic skeleton was determined by analyzing the chemical shifts of the carbons of the double bond, which indicate the presence of triterpenes belonging to the oleanane (δC 143.8, C-13 and 122.2, C-12) and ursane types (δC 138.3, C-13 and 125.7, C-12) (Olea and Roque, 1990). Analyses of HSQC and HMBC spectra led to identification of oleanolic acid (1) and ursolic acid (2) (Dais et al. 2017).
Compound 3 exhibited characteristic signals of a clerodane diterpenoid in its NMR data. Signals for a monosubstituted furan ring (δH 7.45, H-16; 7.39, H-15; 6.40, H-14), a γ-lactone (δΗ 4.37 and 4.30, H-19; δC 170.5, C-18; 71.6, C-19), a δ-lactone (δΗ 5.65, H-12; δC 174.8, C-17; 71.7, C-12), a methyl group (δΗ 1.28, H-20), a trisubstituted double bond (δΗ 6.90, H-3; δC 136.3, C-3; 132.0, C-4), and an oxygenated quaternary carbon (δC 73.8, C-10) were observed. Careful analyses of HSQC and HMBC correlations confirmed that 3 is a pentacyclic cis-clerodane known as ent-(5R,9R)-15,16-epoxy-10S-hydroxycerodan-3,13(16),14-triene-17,12S;18,19-diolide (Almanza et al. 1997), for the which we suggested the common name melissiflorine.
The NMR data of compound 4 were similar to those of 3, indicating a clerodane diterpenoid. However, the 1H and 13C NMR spectra of 4 exhibited additional signals that could be attributed to an α,β-unsaturated ketone (δC 197.5, C-10) and an acetyl group (δH 2.02; δC 169.5). Additionally, the methyl group at C-20 (δH 2.09) appeared to be more deshielded in comparison with the same methyl group in 3. These findings suggested the presence of a double bond between C-8 and C-9 and a cleavage of the bond between C-9 and C-10, resulting in a tetracyclic rearranged seco-clerodane. HSQC and HMBC correlations confirmed the structure of 4, a compound known as 7-epi-salvianduline A (Ortega et al. 1991).
The 1H NMR spectrum of 5 showed signals characteristic of two trisubstituted aromatic rings (δH 7.02–6.25), and an olefinic bond (δH 7.50 and 6.25, J = 15.9 Hz, H-8, H-7; δC 169.3, C-9), suggesting a caffeoyl-derivative. The remaining data supported the identification of rosmarinic acid (Kuhnt et al. 1994).
Antinociceptive and anti-inflammatory activity studies
Formalin-induced nociception is characterized by two phases: an initial phase (phase I), which is believed to be neurogenic, and a second phase (phase II), which starts approximately 15 min after the injection of formalin and lasts till the end of the experiment and is considered to be an inflammatory phase. The treatment of the mice with 10, 30 or 100 mg kg− 1 EESM did not change the first phase of formalin-induced nociception but significantly reduced the second phase (Fig. 2A and B). Similarly, the non-selective, non-steroidal anti-inflammatory drug IND, also reduced only the second phase of formalin-induced nociceptive behavior. Additionally, EESM dose-dependently reduced the LPS-induced mechanical hyperalgesia. The lowest dose used (10 mg kg− 1) did not change the LPS-induced mechanical hyperalgesia while higher doses, 18 mg kg− 1 and 30 mg kg− 1, reduced and completely abolished LPS-induced hyperalgesia (Fig. 3A). As expected, IND also reduced this response (Fig. 3A). This extract also showed anti-inflammatory activity. EESM 100 mg kg− 1 significantly reduced the carrageenan-induced edema (Fig. 3B), particularly 5 h after Cg injection. The steroidal anti-inflammatory drug DEX significantly reduced edema formation at time points 3 and 5 h after Cg injection (Fig. 3B). As mentioned above, the main compound found in EESM was rosmarinic acid (5).
The analgesic and anti-inflammatory activity of rosmarinic acid (5) is well documented. Using similar models, Boonyarikpunchai et al. (2014) showed that rosmarinic acid (5), isolated from Thunbergia laurifolia, also significantly reduced acetic acid-induced writhing, while 100 mg kg− 1 additionally inhibited the first and second phases of formalin-induced nociception and carrageenan-induced edema. This study also showed that rosmarinic acid (5) reduced heat-induced nociception, an effect that was blocked by naloxone, suggesting that rosmarinic acid (5) is effective against inflammatory pain and has an effect dependent of opioid release. However, other mechanisms may be involved since the isolated compound was effective in reducing carrageenan-induced edema, which is not reduced by opioid drugs. Our results show that EESM, which contains rosmarinic acid (5), was effective mostly in the inflammatory pain and edema. We did not observe a reduction in the first phase of formalin-induced pain, suggesting that the extract, at the doses used, is not acting through opioid release. The highest dose of EESM we tested was 100 mg kg− 1 and, therefore, the amount of rosmarinic acid (5) in the extract is certainly lower than that used by Boonyarikpunchai et al. (2014) to reduce formalin-induced nociception. Other mechanisms for rosmarinic acid (5) to reduce inflammatory pain and edema have been proposed involving the cholinergic systems, the L-arginine-nitric oxide pathway, the reduction in the release of inflammatory cytokines particularly in neuropathic pain (Guginski et al. 2009; Ma et al. 2020; Rahbardar et al. 2017; 2018; Areti et al. 2018; El Gabbas et al. 2019, Borgonetti and Galeotti, 2022). The effect of rosmarinic acid (5) on formalin-induced nociception and LPS-induced mechanical hyperalgesia cannot be attributed to motor impairment since even high doses of this compound did not alter the motor performance of the animals, as showed in previous studies (Yu et al. 2022). Therefore, the anti-inflammatory and antinociceptive effect of EESM was expected and confirmed by these studies and is related, at least in part, to the presence of rosmarinic acid (5).
Differently from EESM, EDSM was only tested in the inflammatory hyperalgesia since the evidences in the literature suggested that oleanolic acid (1) and ursolic acid (2) are mainly related to the blockage of the release of inflammatory mediators. EDSM also significantly reduced the LPS-induced mechanical hyperalgesia observed at 3 h and 5 h after the injection of LPS in the hind paw (Fig. 4A) without interfering with the motor performance of the animals (Fig. 4B). As mentioned before, oleanolic acid (1) and ursolic acid (2) were the main triterpenes found in EDSM and are also commonly found in other Salvia species (Jassbi et al. 2016; Ortiz-Mendoza et al. 2022).
Several studies have shown that extracts from different plants that contain oleanolic acid (1) and ursolic acid (2) possess anti-inflammatory and antinociceptive activity, including the reduction of mechanical hyperalgesia (Azevedo et al. 2016; Kuraoka-Oliveira et al. 2020). Therefore, the presence of these compounds as main compounds in EDSM justifies the antinociceptive effect of EDSM. Ursolic acid (2) was also effective in neuropathic pain reducing mechanical and thermal hyperalgesia (Bhat et al. 2016), thermal pain, carrageenan-induced edema formation acting as, COX-2 inhibitors and/or glutamate receptors antagonist (Silva et al. 2017; Rodrigues et al. 2012). Additionally, Qasaymeh et al (2023) also showed that both oleanolic acid (1) and ursolic acid (2) inhibited the release of TNF-α and IL-12 in vitro. Due to this possible effect in glutamate receptors, we investigated if EDSM, at the doses used in the present study, could affect the motor performance of the animals, a common effect observed in drugs that affect the glutamatergic system. We did not observe any effect on the motor performance of the animals, suggesting that the antinociceptive effect observed in the mechanical threshold evaluation is not a bias due to a motor impairment, but does not exclude the possibility that the EDSM could be acting through both, blocking inflammatory mediators release and as NMDA antagonists. However, it is important to mention that previous studies also have shown that oleanolic acid (1) and ursolic acid (2) modulate the activities of several cytochrome P450 (CYP) enzymes suggesting that the consumption of herbal medicines containing these triterpenes can cause important pharmacological interactions (Kim et al. 2004).
Antioxidant assay
The antioxidant activity of the extracts from S. melissiflora was evaluated using the ORAC-FL method and measured in µmol Trolox equivalent (TE) per gram of dry extract (µmol TE g− 1). Extracts with TE > 800.0 µmol TE g− 1 are considered to have antioxidant activity (Prior et al. 2003). Among the extracts tested, only EESM showed antioxidant activity (Table 1). This result aligns with the findings of the phytochemical study, which identified rosmarinic acid (5) as the major constituent in this extract. In our assay 5 showed antioxidant capacity comparable to that of caffeic acid (Table 1). Rosmarinic acid (5) is a well-known compound with powerful antioxidant properties, including scavenging free radicals, activating enzymes, inhibiting lipid peroxidation, and protecting DNA. These properties enable it to effectively counteract oxidative stress-induced damage (Guan et al. 2022; Huang et al. 2019), and contribute to the anti-inflammatory and antinociceptive properties of this extract.
Table 1
Antioxidant capacity of Salvia melissiflora extracts
Sample | ORAC assay (µmol TE g− 1) |
EDSM (CH2Cl2 extract) | 202.6 ± 38.10 |
EESM (EtOH extract) | 2845.2 ± 2.70 |
Caffeic acida | 2.85 ± 0.03b |
Chlorogenic acida | 2.65 ± 0.03b |
Isoquercetina | 5.15 ± 0.09b |
Quercetina | 5.60 ± 0.08b |
Rosmarinic acida | 3.00 ± 0.05b |
a positive experimental control; b data of pure compounds are expressed as relative Trolox equivalent. The values are average of triplicate assays ± standard deviation. |