Mitragynine isolated from MS is known as an indole alkaloid. Alkaloids are defined as basic nitrogenous plant products, mostly optically active and possessing nitrogen heterocycles as their structural units, with a pronounced physiological action [23]. Isolation and purification of an alkaloid from a plant is not a simple process because generally they contain a complex mixture of several alkaloids. In addition, products like glycosides, organic acids, and others present in plants may complicate this isolation further. Thus, the isolation of a pure alkaloid sometimes may become an extremely laborious procedure [23].

In the present study, MG which has been identified according to thin layer chromatography (TLC) method was then analyzed using ¹H-NMR and ¹³C-NMR to further confirm the purity and authenticity of MG. The standard ¹H-NMR and ¹³C-NMR is able to identify MG used in this study as described in previous reports by Chittakarn et al. [24], Kumarnsit et al. [25,26] and Houghton et al. [27]. From the ¹H-NMR and ¹³ C-NMR spectra, a total of 23 carbon and 30 hydrogen resonances and the molecular formula, C₂₃H₃₀N₂O₄ was derived (Figures 1–3).

The antinociceptive activities are determined by exposing animals to potentially painful stimuli such as heat or electric shock and measuring either the time it takes them to respond to the stimuli or the intensity at which they respond. If the time taken by the animal to respond to the stimulus is prolonged following compound or drug administration, then it is presumed that the compound and/or drug have altered the perception to painful stimulus. Increase in latency time until the occurrence of nociceptive responses (licking, shaking legs or jumping) reflects that there is an alteration to painful stimulus, and that the tested compounds have antinociceptive properties. Yaksh [28] indicated that the hot-plate test (HPT) was also used to study the possible involvement of supraspinal receptors whereas tail-flick test was used to investigate the involvement of spinal opioid receptors (Figure 4).

In this study, antinociceptive effect of MG was evidenced by the increase of latency time in the HPT. The increase of latency time is dose dependent with significant increase of dosage up to 35mg/kg. The effect of MG as antinociceptive agent was decreased in much higher dosage up to 60 mg/kg (data not shown). Interestingly, the increase of latency period was higher than that of morphine group. The morphine group only showed an increase in the latency period at 30 min and it is significant when compared to control groups. This finding was in line with the previous findings by Matsumoto et al. [29] which found that MG possesses an antinociceptive activity when given i.p and i.c.v in tail-pinch and hot-plate test.

In the present investigation, the effect of MG in HPT was maximal between 30 min–60 min which is slightly slower than the previous study by Matsumoto et al. [30] which indicates maximal antinociceptive effects at 15 min–45 min after injection. The differences may be due to the sources of MG, which in this study was from the Malaysian Peninsular whereas the study by the Japanese group isolated MG from a source in Thailand. It was revealed by Takayama et al. [17] and Ikram [31] that MG isolated from plants in Thailand consists of a stronger MG in biological activity and percentage.

Selective antagonists were employed in order to clarify the involvement of the receptor subtypes in the antinociceptive effect of MG. CB1 receptor is proved to have a role in pain management. The study of cannabinoids as analgesics stems from findings that compounds like Δ-9-tetrahydrocannabinol (Δ-9-THC), antinociception can produce antinociception without respiratory depression associated with opioid analgesics [32]. CB1 receptor antagonist which is AM251 was used in this study.

From the result, Δ-9-THC was completely blocked by the cannabinoid CB1 receptor antagonist AM251, as expected. This confirms that Δ-9-THC exerts its antinociceptive effect via the CB1 receptor in the HPT. However, the effect of MG was not blocked by AM251 antagonist. Previous findings by Rios et al. [33] have suggested that there is a formation of heteromeric receptor complexes between opioid specifically μ-opioid receptor and CB1 receptor which contribute to functional interaction between these two classes of agonist which may lead to antinociceptive action. In this study, it shows that antinociceptive action of MG does not directly involve the cannabinoid CB1 receptor system (Figure 6).

To investigate which opioid receptor subtypes are involved in the pharmacological effects of MG, further investigations using selective opioid antagonist have been carried out with a 35 mg/kg dosage of MG. The antinociceptive effect of i.p injection of MG was completely blocked by naloxone given i.p, a non-selective opioid receptor antagonist, indicating that opioid receptor systems are involved in the action of MG. These findings are consistent with the findings of Matsumoto et al. [30] who reported that i.p and i.c.v administration of naloxone blocked the antinociceptive effects of MG but it differs with the findings reported by Macko et al. [34] who found that i.p administration of naloxone did not block the antinociceptive actions of MG (Figure 7).

This study also supported the findings that besides μ-opioid receptors, δ-receptors are also involved in the mechanism of action of MG as antinociceptive compound. This was proved when the antinociception caused by MG was also significantly antagonized by co-administration of naltrindole, a δ receptors antagonist. With the co-administration of nalozonazine, however, a specific μ₁-receptor antagonist, even though it did inhibit the antinociceptive action of MG, it was not statistically significant. This proved the antinociceptive effects of MG, not only through μ₁-receptor, but through both μ-receptor subtypes.

The present findings showed significant difference between previous reports when MG blocked the effect of norbinaltorphimine, a selective κ-opioid receptor antagonist for 60 min, whereas in the case of Thongpradichote, [35] reported that MG blocked the effect of norbinaltorphimine, a selective κ-opioid receptor antagonist in the tail-pinch test but not in hot plate analysis (Figure 7).

Mitragyna speciosa leaves were collected from natural sources around Peninsular Malaysia and authenticated by a botanist from the Faculty of Forestry, Universiti Putra Malaysia. A voucher specimen (ATS: 001) has been kept at the herbarium for future reference. Mitragynine was isolated from leaves of Mitragyna speciosa according to the methods described by Houghton and Ikram [27], Ponglux et al. [36] and Reanmongkol et al. [37] with a slight modification.

One kilogram leaves of MS was cleaned and dried with constant temperature at 45 °C overnight before grinded into powder form. It was then macerated with absolute methanol for 72 h. The mixture was filtered to remove filtrate from insoluble particles to obtain the methanol extract. Extract was evaporated using rotary evaporator (Eyela, Japan) at a temperature below 55 °C. The methanol extract was added with 5% sulphuric acid and extensively stirred overnight. The mixture was filtered and a clear yellow solution of acidic filtrate was obtained. The acidic filtrate was then mixed with sodium carbonate and stirred until it turned to dark grey with pH 11 to become basic filtrate. To separate the alkaloids from the basic crude extract, chloroform was added to mixture in separating funnel. Component of three layers were produced which were the aqueous, salt and chloroform layer. The chloroform layer was then separated and filtered out from the mixture. The chloroform fraction was mixed with sodium sulfate anhydrous and evaporated to yield 0.73% (w/w) of crude extract. The major alkaloid isolated by silica gel chromatography and thin layer chromatography eluting with diethyl ether was identified as MG with standard nuclear magnetic resonance method (¹H NMR, ¹³C NMR). Over all, the yield of MG was approximately 0.087% (w/w) of fresh leaves (Figure 8).

Male ICR mice weighing 25 g–35 g were housed in Animal Center, Faculty of Medicine and Health Sciences (FMHS), Universiti Putra Malaysia, Malaysia. Animals were divided into different experimental groups (n = 8) in a temperature-controlled room. They were maintained under standard laboratory conditions with natural dark and light cycle and fed with standard commercial food pellets and water ad libitum. Animals were acclimatized for at least 7 days to adapt to the laboratory prior to experiment. All animal procedures were approved (UPM/FPSK/PADS/BR-UUH/00307) by Institutional Animal Care and Use Committee (IACUC), FMHS, Universiti Putra Malaysia (UPM), Malaysia.

The drugs used in this study were morphine hydrochloride (Sigma-Aldrich, USA), polyoxyetheylene sorbitan monooleate (Tween 80, Sigma-Aldrich, USA), selective opioid antagonist naloxone hydrochloride, naloxonazine, norbinaltorphimine, naltrindole (Lipomed Inc., USA), while the selective cannabinoid type 1 antagonist was; 1-(2,4-diclorophenyl)-5-(4-iodophenyl)-4-methyl-N-1-piperidinyl- 1-H-pyrazole-3-carboxamide (AM251) (Lipomed Inc., USA). Δ-9-Tetrahydrocannabinol (THC) also been used (Sigma-Aldrich, USA). Mitragynine and morphine were dissolved in 2.0% (v/v) Tween 80 and normal saline (0.9% NaCl) respectively. All reagents used in the study were of analytical grade.

Mice were randomly assigned into one normal saline (vehicle control) group and six treatment groups with eight mice per group. The pre-drug latency was tested before used where the mice with pre-drug latency below five seconds or more than 20 s were eliminated from the study. This was to avoid any bias for this experiment since the mice with extremely short or long pre-latency measurement could result in some variances in the analysis. Experimental groups assigned to: Normal saline (Vehicle control); Morphine (3 mg/kg); Mitragynine (3 mg/kg; 10 mg/kg; 15 mg/kg; 30 mg/kg and 35 mg/kg b.wt). Administration of all drugs was performed through intraperitoneal (i.p) injection at a volume of 0.5 mL/kg b.wt. All the treatments were given 15 min before perform the hot-plate test. The latency time was measured continuously after injection for 2 h at 30 min intervals. The ED₅₀ value of the compound was determined and same concentration was used for receptor antagonist study.

Mice were divided into fifteen groups (n = 8) that were pretreated with selected cannabinoid and opioid receptor antagonist. The cannabinoid receptor 1 antagonist, AM251 (3 mg/kg) was administered intraperitoneally (i.p) 30 min before MG and Δ-9-tetrahydrocannabinol (Δ-9-THC) injection. For opioid antagonists, naloxone hydrochloride (Nal) (1 mg/kg), naloxonazine (NZI) (10 mg/kg), norbinaltorphimine (norBNI) (10 mg/kg) and naltrindole (NTI) (3 mg/kg) were administered i.p. 15 min, 24 h, 1 h, 30 min, respectively before MG (35 mg/kg) or morphine injection (3 mg/kg). At the end of the experiment period, all mice were sacrificed.

In the hot-plate test, mice were placed on a stainless steel surface which was maintained at the temperature of 50 °C ± 0.2 °C (Ugo Basile, Italy, Model 7280) and the latency time was assessed. The latency time is defined as the period of time when the mice were placed in the hot surface until the occurrence of nociceptive responses such as licking, shaking or jumping was observed. The cut-off time of 50 s was used to prevent tissue damage during the experiment. 30 min prior to treatment, the nociceptive threshold was measured and the latency time had been used as the pre-drug latency for each animal. The latency time was calculated as:

The results were expressed as mean ± standard error (SEM) for eight mice in each group. Comparisons between experimental and control groups were analyzed by using one-way analysis variance (ANOVA) coupled with post hoc Tukey’s HSD test. Values with p < 0.05 were considered to indicate statistical significance.