Metabolism of speciociliatine, an overlooked kratom alkaloid for its potential pharmacological effects

Abstract

Speciociliatine, a diastereomer of mitragynine, is an indole-based alkaloid found in kratom (Mitragyna speciosa). Kratom has been widely used for the mitigation of pain and opioid dependence, as a mood enhancer, and/or as an energy booster. Speciociliatine is a partial μ-opioid agonist with a 3-fold higher binding affinity than mitragynine. Speciociliatine has been found to be a major circulating kratom alkaloid in humans following oral administration of a kratom product. In this report, we have characterized the metabolism of speciociliatine in human and preclinical species (mouse, rat, dog, and cynomolgus monkey) liver microsomes and hepatocytes. Speciociliatine metabolized rapidly in monkey, rat, and mouse hepatocytes (in vitro half-life was 6.6 ± 0.2, 8.3 ± 1.1, 11.2 ± 0.7 min, respectively), while a slower metabolism was observed in human and dog hepatocytes (91.7 ± 12.8 and >120 min, respectively). Speciociliatine underwent extensive metabolism, primarily through monooxidation and O-demethylation metabolic pathways in liver microsomes and hepatocytes across species. No human-specific or disproportionate metabolites of speciociliatine were found in human liver microsomes. The metabolism of speciociliatine was predominantly mediated by CYP3A4 with minor contributions by CYP2D6.

INTRODUCTION

Mitragyna speciosa (Korth.) Havil. (Rubiaceae), a medicinal plant indigenous to Southeast Asia, is more commonly known as kratom. Kratom leaves have been consumed by natives of Southeast Asia for centuries to treat a variety of ailments like fever, cough, diarrhoea, and depression. The opioid-like effects of kratom, mainly at higher doses, have made it more popular in the western world where it is being consumed for the management of pain or opioid withdrawal symptoms (1–5). Kratom has been reported to contain more than 40 alkaloids, but most studies have focused on its major alkaloid mitragynine (0.7–38.7% in kratom products) and a potent minor alkaloid 7-hydroxymitragynine (<2% in kratom products) which are considered to be major contributors to the analgesic effects of kratom through μ-opioid receptor activation (4, 6–10). Additionally, 7-hydroxymitragynine is one of the major metabolites of mitragynine, thus its overall systemic exposure is increased following kratom administration through the metabolism of mitragynine (6, 11, 12). The minor kratom alkaloids paynantheine, speciogynine (a diastereomer of mitragynine), and speciociliatine (a diastereomer of mitragynine), which are present in appreciable amounts, have not been well characterized. Speciociliatine has been reported to be present as 1% of the total alkaloidal content in kratom extract but thus far has been completely overlooked for its influence on the overall pharmacology of kratom (4, 7, 13). In recent reports where individual kratom alkaloids were quantified, speciociliatine content relative to mitragynine concentration ranged from 8–58% w/w in kratom juice or lyophilized kratom tea (10, 12, 14, 15). Further, speciociliatine is reported to be a partial agonist at the μ-opioid receptor and have a 3-fold higher binding affinity (54.5 ± 4.4 nM) than mitragynine (16). It does not recruit the β-arrestin-2 pathway (17). In the hot plate assay, speciociliatine demonstrated antinociceptive effects with a potency similar to morphine in rats (16). Speciociliatine has also been reported to reduce voluntary alcohol intake at a 30 mg/kg intraperitoneal dose in C57BL/6 mice (17). Speciociliatine has shown developmental toxicity in Zebrafish (Danio rerio) embryos at very high concentrations (> 50 μg/ml) and these concentrations may be physiologically irrelevant to humans after oral consumption (C_max, 122 ng/ml) of kratom (12, 18).

Moreover, a pharmacokinetic study of kratom alkaloids following oral administration of lyophilized kratom tea or a commercial kratom product in rats revealed that among eleven tested alkaloids, mitragynine had the highest exposure followed by speciociliatine. The area under the curve up to 24 hr (AUC_{0–24 hr}) for speciociliatine was around 4-fold lower than mitragynine but still several-fold higher than that of the other minor alkaloids detected: 7-hydroxymitragynine, corynantheidine, speciogynine, and paynantheine (14). Also, the clearance of speciociliatine (0.7 ± 0.2 L/hr/kg) was 2-fold lower than mitragynine (1.3 ± 0.1 L/hr/kg) in rats and the oral bioavailability of speciociliatine in rats was reported to be 20% (2, 19). Importantly, the pharmacokinetics of a kratom product administered orally to healthy humans revealed that speciociliatine (AUC_0-inf = 5120 nM·hr) was the major circulating kratom alkaloid with ~12-fold higher exposure than mitragynine (AUC_0-inf = 420 nM·hr). However, the dose of speciociliatine (5.12 ± 0.26 mg/g of kratom powder) was ~4 fold lower than mitragynine (19.48 ± 0.81 mg/g of kratom powder) in the kratom product (12).

Based on this data, speciociliatine is likely one of the important alkaloids contributing to the pharmacological effects of kratom. However, there is insufficient information available on the metabolism and the enzyme(s) involved in the metabolism of speciociliatine. In a previous report, Philips et. al. studied the metabolites of speciociliatine in the urine samples of human subjects who had consumed kratom products and in rat urine following a pure 40 mg/kg speciociliatine oral dose using liquid chromatography-linear ion trap mass spectrometry (20). Since these kratom products also contain mitragynine, speciogynine, and mitraciliatine, the diastereomers of speciociliatine, it would be challenging to identify metabolites of speciociliatine reliably among similar metabolites of its diastereomers. Further, interspecies comparison of metabolite profiles would help to identify the appropriate preclinical species, with metabolic pathways similar to that of humans, for future in vivo studies. Also, the identification of enzymes involved in the metabolism of speciociliatine is key to predicting potential metabolism-based drug-drug interactions due to the inhibition or induction of these enzymes by co-ingested drugs.

In the present study, we have reported the metabolic pathways of speciociliatine in cross-species liver microsomes and hepatocytes and identified the cytochrome P450 (CYP450) enzyme(s) responsible for the metabolism of speciociliatine using liver microsomes and recombinant CYP450 isoforms.

MATERIALS AND METHODS

Chemicals and reagents

Speciociliatine (Purity >98%) was purified from the alkaloid-rich extract of kratom as previously described (10). The purity and structural characterization of speciociliatine was determined by various analytical techniques including ultra-high performance liquid chromatography photodiode array detection (UHPLC-PDA), liquid chromatography high-resolution quadrupole time of flight mass spectrometry (LC-Q-TOF), proton nuclear magnetic resonance spectroscopy (¹H NMR), and carbon (¹³C) NMR. Reduced glutathione (GSH) and nicotinamide adenine dinucleotide phosphate reduced tetrasodium salt (NADPH) were obtained from MP Biomedicals (Solon, OH, USA). Sulfaphenazole, benzylnirvanol, quinidine, ketoconazole, verapamil [internal standard, (IS)], propranolol, atenolol, digoxin, trypan blue solution, and CYP3cide were procured from Sigma Aldrich (St. Louis, MO, USA). Montelukast sodium was obtained from Cayman Chemical (Waterloo, Australia). Pooled male mouse, rat, dog, cynomolgus monkey, and pooled mixed gender human liver microsomes were obtained from XenoTech, LLC (Lenexa, KS, USA). The insect cell control microsomes and rCYPs were purchased from Corning (Corning, NY, USA). Pooled male mouse, rat, dog, cynomolgus monkey, pooled mixed gender human hepatocytes, INVITROGRO^™ HT Medium, and INVITROGRO^™ KHB were procured from BioIVT (Baltimore, MD, USA). Ammonium acetate, acetic acid, formic acid, acetonitrile (ACN), potassium dihydrogen phosphate, dipotassium hydrogen phosphate, Dulbecco’s Modified Eagle Medium (DMEM), fetal bovine serum, non-essential amino acids, penicillin-streptomycin mixture, and L-glutamine were purchased from Fisher Scientific (Fair Lawn, NJ, USA). All reagents were either analytical grade or LC-MS grade. The human colorectal adenocarcinoma (Caco-2) cells and HTB-37^™, were obtained from the American Type Culture Collection (ATCC) (Manassas, VA, USA).

Instruments

Agilent 1290 Infinity series ultra high-performance liquid chromatography (UHPLC) systems (Agilent Technologies, Santa Clara, CA, USA) equipped with diode array detector (DAD) and coupled with Agilent 6540 quadrupole-time-of-flight (Q-TOF) high-resolution mass spectrometer (LC-DAD-HRMS) was used for metabolite identification of speciociliatine. Waters Acquity class I UPLC system equipped with Waters Xevo TQ-S micro mass spectrometer (Waters Corp., Milford, MA, USA) (LC-MS/MS) was used for the quantitative analysis of speciociliatine and its metabolites in metabolic stability, permeability, and reaction phenotyping (identification of the enzyme(s) responsible for metabolism) studies. The details of chromatographic conditions used for each study are provided as supplementary information.

Metabolic stability of speciociliatine in cross-species hepatocytes

The metabolic stability of speciociliatine (1 μM) was performed in pooled male mouse, rat, dog, cynomolgus monkey, and pooled mixed gender human hepatocytes (each species hepatocytes were initially thawed and cell viability was assessed as per the vendor protocol) at a concentration of 1 × 10⁶ viable cells/ml suspended in the incubation medium. Each hepatocyte suspension solution was preincubated at 37°C with 5% CO₂ for 5 min in an incubator shaker (150 rpm) followed by the addition of speciociliatine at a final concentration of 1 μM. The organic solvent concentration was kept below 0.5% v/v (21, 22). Each reaction was conducted in triplicate. Midazolam (1 μM), and umbelliferone (1 μM) were used as positive controls. The reaction was kept at 37°C with 5% CO₂ for 120 min in the incubator shaker and aliquots of 25 μL (withdrawn at 0, 10, 15, 30, 45, 60, 75, 90, and 120 min) were mixed with 125 μL of ACN containing 0.1% v/v formic acid and 25 ng/mL of IS to inactivate the cells. The samples were then vortex mixed for 5 min and filtered through a 0.45 μm membrane 96 well Millipore (Burlington, MA, USA) multiscreen Solvinert filter plate under centrifugation at 2000 ×g for 5 min at 4 °C. Filtrates were then subjected to LC-MS/MS analysis.

Metabolite identification of speciociliatine in cross-species liver microsomes and hepatocytes

A typical metabolite identification assay was performed in pooled male mouse (MsLM), rat (RLM), dog (DLM), cynomolgus monkey (MkLM), and pooled mixed gender human liver microsomes (HLM) (11, 23). Speciociliatine (at a final concentration of 30 μM) was incubated with 1 mg/ml protein of MsLM, RLM, DLM, MkLM, or HLM in 50 mM phosphate buffer (pH 7.4) containing 3 mM of GSH and 1 mM NADPH. The reaction was initiated by the addition of NADPH. The volume of the reaction was 1000 μL adjusted with 50 mM phosphate buffer, pH 7.4. The reaction was maintained at 37 °C and kept in an incubator shaker at 100 rpm for 60 min. The reaction without NADPH was used as a control. Aliquots (250 μL) were withdrawn at 0 and 60 min and protein precipitation was achieved by the addition of an equal volume of ice-cold ACN to stop the reaction.

In the case of the cross-species hepatocyte metabolite identification study, the pooled cryopreserved male mouse, rat, dog, cynomolgus monkey, and pooled mixed gender human hepatocytes were initially thawed and the percent viability was estimated as per the vendor protocol. Each species hepatocytes were suspended in the incubation medium at 1 × 10⁶ viable cells/ml and preincubated at 37°C with 5% CO₂ for 5 min in an incubator shaker (150 rpm). Following preincubation, speciociliatine, at a final concentration of 30 μM, was incubated with the hepatocytes at 37°C with 5% CO₂ for 120 min in the incubator shaker. The organic solvent concentration was kept below 0.5% v/v. Aliquots (250 μL) were withdrawn at 0 and 120 min and mixed with an equal volume of ice-cold ACN containing 0.1% v/v formic acid to inactive the cells. Both liver microsomal and hepatocyte incubation samples were processed as described in Section 2.3. Filtrates were then analyzed on the Agilent 6540 Q-TOF instrument to obtain the LC-DAD-HRMS data.

Determination of CYP enzyme contributions to speciociliatine metabolism using rCYP and HLM

The reaction phenotyping of speciociliatine was performed as previously reported (11, 23). For the rCYPs approach, speciociliatine at a concentration of 1 μM was incubated with 25 pmol/ml of rCYP1A2, rCYP2B6, rCYP2C8, rCYP2C9, rCYP2C19, rCYP2D6, rCYP3A4, or rCYP3A5 in 50 mM phosphate buffer (pH 7.4) in presence of 1 mM NADPH. The insect cell control microsomes were used to adjust the protein content of each reaction to 0.5 mg/ml. Each reaction was pre-incubated for 5 min at 37°C before the addition of NADPH as a cofactor to initiate the reaction and then the incubation was continued for another 60 min at 37°C in an incubator shaker (150 rpm). The concentration of organic solvents in the reaction was limited to < 0.5% v/v (21, 22). Each reaction was performed in duplicate. An aliquot of 50 μl was withdrawn at 0, 3, 15, and 60 min and mixed with 150 μL of ice-cold ACN containing 25 ng/ml IS. These samples were then processed as described in section 2.3.

In the case of the chemical inhibition approach, speciociliatine (1 μM) was incubated with 0.5 mg/ml HLM protein in the presence or absence of specific chemical inhibitors of CYP450 isoforms in 50 mM phosphate buffer (pH 7.4). The specific CYP450 inhibitors and their respective concentration used in this study were alpha-naphthoflavone (1 μM) for CYP1A1, 2-phenyl-2-(1-piperidinyl) propane (PPP, 15 μM) for CYP2B6, montelukast (1 μM) for CYP2C8, sulphaphenazole (10μM) for CYP2C9, N-benzylnirvanol (1μM) for CYP2C19, quinidine (1μM) for CYP2D6, CYP3cide (0.5 μM) for CYP3A4, and ketoconazole (1μM) for CYP3A4/5 (11, 23, 24). The mixture of speciociliatine, HLM protein, and respective inhibitors (except mechanism-based inhibitors) in 50 mM phosphate buffer was preincubated for 5 min at 37°C before the addition of NADPH (1 mM). The mechanism-based inhibitors, PPP and CYP3cide, were initially pre-incubated with HLM supplemented with NADPH in 50 mM phosphate buffer for 15 min, and the reaction was initiated by the addition of speciociliatine. The reaction was conducted for 60 min at 37°C and aliquots (50 μl) were withdrawn at 0, 15, and 30 min. The aliquots were quenched with 150 μL of ice-cold ACN containing 25 ng/mL of IS. These samples were then processed as described in Section 2.3 and analyzed on a UPLC coupled with Waters Acquity Xevo TQ-S triple quadrupole instrument.

The fraction metabolized (f_m) for speciociliatine was determined by comparing the elimination rate constant of speciociliatine in incubations with or without CYP450 inhibitors as previously described (25, 26). The slope of the line (elimination rate constant, k) was obtained from a graph of the natural log of the percentage of the parent compound remaining against time. The percent inhibition of metabolism of speciociliatine was calculated as a percentage ratio of the difference between the elimination rate constants of control and with inhibitor reaction relative to the control reaction. The f_m of each enzyme was calculated as the ratio of inhibition of metabolism to the sum of percent inhibition across all tested enzymes.

Assessment of permeability of speciociliatine across Caco-2 cell monolayer

Caco-2 cells were cultured in T-75 flasks in DMEM medium with 10% fetal bovine serum, 1% non-essential amino acids, 100 U/ml penicillin, 100 μg/ml streptomycin, and 2 mM L-glutamine, and incubated at 37 °C in an atmosphere with 5% CO₂. The media was changed alternate days and cells were split when more than 80% confluency was achieved. The cells used in this study were between 25 and 30 passages.

The cells (0.85×10⁵ cells/cm²) were seeded on 24-well Transwell^® cell culture inserts with 0.4 μm pore size and surface area of 0.33 cm². The media was changed on an alternate day for 15 days and every till 21–25 days. Caco-2 monolayers 21–25 days post-seeding were used for evaluating the permeability of speciociliatine at a concentration of 5 μg/mL. Before the start of the permeability assessment, DMEM was replaced with warm HBSS buffer (pH 7.4). The integrity of the Caco-2 monolayer was verified by measuring the transepithelial electrical resistance (TEER) value across the monolayer using a Millicell, ERS meter. The wells with a TEER value over 250 Ω.cm² were used for the permeability assessment. All experiments were conducted in triplicate. Propranolol (5 μM) and atenolol (10 μM) were used as high and low permeability markers while digoxin (10 μM) was used as a P-gp substrate to validate the assay. The apical (A) to basolateral (B) transport experiments across Caco-2 monolayers were conducted by adding 0.1 mL of the compound solution to the apical compartment of the inserts and 0.6 mL of blank HBSS buffer to the basolateral compartment. The basolateral (B) to apical (A) transport experiments across Caco-2 monolayers were conducted by adding 0.1 mL of blank HBSS buffer to the apical compartment of the inserts and 0.6 mL of the compound solution to the basolateral compartment. A 25 μL sample was collected from the acceptor compartment at 0 and 1 h and replaced with an equal volume of blank HBSS buffer. The apparent permeability coefficient (P_app) was calculated as

$$P_{app}=\left(\frac{V_A}{AreaXTime}\right)X\left(\frac{{\left[drug\right]}_{acceptor}}{{\left[drug\right]}_{initial,donor}}\right)$$

Where V_A is the volume (in mL) in the acceptor well, area is the surface area of the membrane (0.33 cm²), time is the total transport time in seconds, [drug]_acceptor is the concentration of the drug in acceptor compartment, and [drug]_{initial,donor} is the initial concentration of drug in the donor compartment. The efflux ratio was calculated using the formula,

$$EffluxRatio\left(ER\right)=\raisebox{1ex}{$P_{app(B\to A)}$}\!\left/ \!\raisebox{-1ex}{$P_{app(A\to B)}$}\right.$$

RESULTS

Metabolic stability of speciociliatine in cross-species hepatocytes

The in vitro metabolic stability of speciociliatine expressed as the percentage of speciociliatine remaining versus time is shown in Figure 1. speciociliatine was found to be unstable in mouse, rat, and cynomolgus monkey hepatocytes with in vitro metabolic half-life (t_1/2) values of 11.2 ± 0.7, 8.3 ± 1.1, and 6.6 ± 0.2 min, respectively. The metabolism of speciociliatine was found to be much slower in human and dog hepatocytes with t_1/2 values of 91.7 ± 12.8 and >120 min, respectively. The t_1/2 values were used to estimate the in vitro intrinsic clearance and further extrapolated to determine the hepatic clearance of speciociliatine using the well-stirred model as previously described (27, 28). The extrapolated hepatic clearance of speciociliatine was found to be 80.2, 49.2, 12.6, 39.7, and 9.5 ml·min/kg in mice, rats, dogs, cynomolgus monkeys, and humans, respectively.

Fig. 1. Metabolic stability of speciociliatine (SPC) in cross-species hepatocytes.

The percentage of speciociliatine remaining following incubation at 1 μM concentration in mouse, rat, dog, cynomolgus monkey, and human hepatocytes (1 million cells/ml) was plotted against time. The data are expressed as mean ± standard deviation (n=3).

Metabolite identification of speciociliatine in cross-species liver microsomes

The molecular ion [M+H]⁺ peak of speciociliatine (C₂₃H₃₀N₂O₄, Mol. Wt. 398.4953 Da) was found at mass to charge ratio (m/z) of 399.2280 within 5 ppm mass accuracy error and was fragmented using a 28 V collision energy. The fragmentation (MS/MS) profile of speciociliatine along with the assigned fragment structures of speciociliatine based on the ring double bond equivalent, accurate masses of the fragments, and predicted molecular formulas (within 5 ppm mass accuracy error) are shown in Figure 2.

Fig. 2.

High-resolution collision-induced dissociation (MS/MS) fragmentation spectrum of speciociliatine (m/z 399.2280).

Similar to the fragmentation profile of mitragynine, the MS/MS of speciociliatine showed characteristic fragments at m/z 110.0964, 129.0546, 174.0913, 238.1438, and 367.2016 (Figure 2) (11). The fragment ion at m/z 110.0964 corresponds to the ethylpiperidine moiety as a result of losing the methyl β-methoxyacrylate group and the C-N bond cleavage of the tertiary N-atom to the ethyl methoxyindole ion (m/z 174.0913) from speciociliatine. The fragment ion at m/z 129.0546 corresponds to the methyl β-methoxy-2-methylacrylate ion as a result of two C-C bond cleavages on the ethylhydroquinolizine group. Additionally, C-C bond cleavage on the ethylhydroquinolizine moiety corresponds to the fragment ion at m/z 238.1438 composed of the methylethylpiperidine β-methoxyacrylate moiety of speciociliatine. The fragment ion at m/z 367.2016 results from the loss of a methoxy group from the β-methoxyacrylate moiety of speciociliatine.

The metabolites of speciociliatine across species liver microsomes (60 min sample) and hepatocytes (120 min sample) were identified as new peaks in the test sample UV chromatograms (254 nm) compared to the respective control sample (0 min). Further, the base peak ion chromatograms of putative metabolites were extracted (Figure 3) and matched with the UV chromatogram retention times of new peaks of test samples. Subsequent structure elucidation of the plausible metabolites was performed by comparing fragmentation (MS/MS data) of these new peaks with the fragmentation profile of speciociliatine. Table 1 lists the details of metabolites of speciociliatine which includes retention time (based on the mass detector), modification, relative UV abundance, and exact masses of the metabolites.

Fig. 3.

Extracted ion chromatograms of speciociliatine and its metabolites upon incubation of speciociliatine in mouse, rat, dog, cynomolgus monkey, and human liver microsomes or hepatocytes.

Table 1.

Metabolite profile of speciociliatine in mouse, rat, dog, cynomolgus monkey, and human liver microsomes or hepatocytes

Metabolite/Parent	MS_RT (min)	Modification	Relative UV peak area abundance										[M+H]+ (m/z)
			MsLM	MsH	RLM	RH	DLM	DH	MkLM	MkH	HLM	HH
Speciociliatine	11.61	-	68.0	80.1	77.4	27.4	86.8	86.6	10.9	77.9	82.6	91.3	399.2278
Met 1	4.80	O-demethylation+2O+2H	ND	MS	ND	MS	ND	ND	0.1	0.5	0.1	0.2	419.2177
Met 2	5.56	O-demethylation+O	MS	MS	MS	10.5	MS	MS	MS	MS	MS	ND	401.2071
Met 3	5.73	O-demethylation-2H+Glu	NA	MS	NA	MS	NA	ND	NA	MS	NA	ND	559.2286
Met 4	5.86	O-demethylation/hydrolysis+GSH	0.1	ND	MS	MS	MS	ND	3.1	MS	0.2	ND	690.2803
Met 5	5.92	O-demethylation/hydrolysis+O	MS	ND	MS	ND	ND	ND	MS	0.5	MS	0.2	401.2071
Met 6	6.03	O-demethylation-4H+Glu	NA	MS	NA	MS	NA	ND	NA	MS	NA	ND	557.2130
Met 7	6.23	O-demethylation/hydrolysis-2H+Glu	NA	0.9	NA	MS	NA	ND	NA	0.2	NA	ND	559.2286
Met 8	6.29	O-demethylation+GSH	0.7	ND	0.2	ND	0.2	ND	4.1	ND	0.1	ND	690.2803
Met 9	6.35	O-demethylation/hydrolysis+O	ND	ND	MS	ND	ND	ND	MS	MS	0.2	0.3	401.2071
Met 10	6.67	O-demethylation+GSH	MS	MS	MS	MS	MS	ND	MS	MS	MS	ND	690.2803
Met 11	6.71	P+O	2.7	0.8	0.6	1.6	0.7	0.2	14.3	1.6	1.6	0.6	415.2227
Met 12	6.87	O-demethylation/hydrolysis+GSH	0.7	MS	0.2	9.2	MS	ND	2.3	MS	0.3	ND	690.2803
Met 13	6.88	P+O+Glu	NA	4.3	NA		NA	0.1	NA	0.5	NA	ND	591.2548
Met 14	7.07	O-demethylation/hydrolysis+Glu	NA	MS	NA	MS	NA	ND	NA	MS	NA	0.5	561.2443
Met 15	7.27	P+O	8.9	2.0	7.3	9.7	2.6	0.6	15.8	2.3	3.9	0.4	415.2227
Met 16		P+GSH											704.2960
Met 17	7.50	P+2O	3.2	0.7	2.7	2.0	0.9	0.8	5.0	2.8	1.4	MS	431.2177
Met 18		P+GSH										MS	704.2960
Met 19	7.56	P+O+SO3H	NA	ND	NA	MS	NA	MS	NA	MS	NA	0.2	495.1796
Met 20	7.61	P+O+GSH	1.3	0.5	0.1	0.9	MS	ND	5.6	0.3	0.2	ND	720.2909
Met 21	7.61	P+O+Glu	NA		NA		NA	ND	NA		NA	0.8	591.2548
Met 22	8.00	P+O	1.6	2.4	2.1	13.1	2.1	7.6	9.7	1.5	0.9	0.8	415.2227
Met 23	8.42	O-demethylation	5.1	3.7	5.1	15.5	4.6	2.2	14.5	5.5	5.3	2.5	385.2122
Met 24	8.55	P+O+SO3H	NA	ND	NA	ND	NA	ND	NA	0.2	NA	MS	495.1796
Met 25	9.02	P+O	4.2	MS	1.1	ND	1.6	1.6	3.6	1.5	1.7	0.6	415.2227
Met 26		P+2O-2H		MS		MS							429.2020
Met 27	9.13	P+O	1.8	3.1	0.4	6.5	MS	MS	1.1	0.6	0.6	0.2	415.2227
Met 28	9.33	O-demethylation-2H	MS	MS	MS	MS	MS	MS	MS	MS	MS	MS	383.1965
Met 29	9.53	O-demethylation/hydrolysis	MS	0.7	1.9	MS	MS	MS	8.1	3.1	1.1	1.5	385.2122
Met 30		O-demethylation/hydrolysis-2H	MS			MS		MS					383.1965
Met 31	9.54	P+O	MS	MS	MS	MS	0.6	0.4	MS	MS	MS	MS	415.2227
Met 32	9.89	P+O	0.6	0.4	MS	ND	MS	MS	ND	0.7	MS	MS	415.2227
Met 33	10.53	P+O-demethylation/hydrolysis-2H	1.0	0.3	0.6	2.9	MS	MS	0.9	MS	MS	MS	383.1965
Met 34	10.84	O-demethylation-4H	0.2	MS	0.2	0.6	ND	ND	ND	ND	ND	ND	381.1809

The values represent the relative UV peak area abundances of metabolites and speciociliatine in liver microsomes or hepatocytes detected at 254 nm.

MS_RT: retention time in mass chromatogram; MS: Metabolites peaks were detected in mass detector only; ND: not detected; NA: not applicable; P: parent (speciociliatine); GSH: glutathione; Glu: glucuronide; Ms: Mouse; R: Rat; D: Dog; Mk: Cynomologus; H: Human; LM: liver microsomes; H: Hepatocytes

Mono-oxidative metabolites (Met 11, 15, 22, 25, 27, 31, and 32)

Met 11, 15, 22, 25, 27, 31, and 32 showed mass addition of 15.995 Da suggesting addition of an oxygen atom, resulting in a molecular ion peak at m/z 415.2227, mono-oxidation of speciociliatine. The MS/MS product ion spectra of Met 11, 15, 22, 25, and 31 formed diagnostic product ions of m/z 110.0964, 129.0546, and 238.1438 corresponding to unmodified ethylpiperidinyl methyl β-methoxyacrylate moiety of speciociliatine (Figure S2). Further, the product ion at m/z 190.0863 represents the addition of oxygen atom (15.995 Da) to the ethyl methoxyindole ion (m/z 174.0913) implying the monooxidation on the indoloquinolizine moiety of speciociliatine (Figure S2). Additionally, Met 11, 15, and 31 (m/z 415.2227) MS/MS product ion spectra showed a major product ion at m/z 397.2122 (−18 Da) corresponding to a neutral loss of a water molecule from the metabolite, which is typical for metabolites with aliphatic hydroxylation (Figure S2) (29). In the case of the fragmentation profiles of Met 27 and 32 (m/z 415.2227), fragment ions at m/z 174.0913 and 129.0546 depicting the unchanged ethyl methoxyindole and methyl β-methoxyacrylate ions, respectively (Figure S3). A fragment ion at m/z 397.2122 as a result of water loss and fragment ion at m/z 255.1465 or 254.1387 related to ethylpiperidinyl methyl β-methoxyacrylate radical cation or cation, respectively with a hydroxy group denotes hydroxylation of the ethyl quinolizine moiety of speciociliatine (Figure S3). Among these hydroxylated metabolites, Met 11 was identified to be the 7-hydroxyspeciociliatine metabolite, confirmed by comparing the retention time and fragmentation pattern with the synthetic standard (Figure S4)

Di-oxidative metabolite (Met 17)

Met 17 showed a molecular ion peak at m/z 431.2177, a mass addition of 31.9899 Da, equivalent to the mass of two oxygen atoms representing di-hydroxylation of speciociliatine. The fragmentation profiles of Met 17 showed unchanged diagnostic fragment ions at m/z 110.0964, 129.0546, and 238.1438 implying no modification of the ethylpiperidinyl methyl β-methoxyacrylate moiety of speciociliatine (Figure S5). In addition, a fragment ion at m/z 206.0812 (+32 Da to m/z 174.0913) is indicative of the di-oxidation of the ethyl methoxyindole moiety of speciociliatine (Figure S5). The MS/MS spectrum of Met 17 also showed a fragment ion peak at m/z 178.0499 which relates to a methoxyindole with two more oxygen atoms denoting the di-oxidation of the indole moiety of speciociliatine (Figure S5).

O-Demethylated/hydrolyzed metabolites (Met 23 and 29)

Met 23 and 29 showed molecular ion peaks at m/z 385.2122 suggestive of a loss of a methylene group (−14.0157 Da) as a result of O-demethylation either at 9-methoxyindole or the β-methoxyacrylate position, or the hydrolysis of the methyl acrylate ester. The fragmentation profile of Met 23 showed unchanged fragment ions at m/z 238.1438, 226.1438, and 129.0546 meaning no change at the β-methoxyacrylate moiety of speciociliatine. Met 23 also produced a fragment ion at m/z 160.0757 (−14 Da) instead of m/z 174.0913 confirmed that O-demethylation occurred at the 9-methoxy position on the indole ring of speciociliatine (Figure S6). While the MS/MS spectrum of Met 29 showed an unchanged fragment ion at m/z 174.0913 and an absence of a fragment ion at m/z 129.0546, fragment ions at m/z 224.1281 and 212.1281 instead of m/z 238.1438 and 226.1438 (−14 Da, each) indicate the loss of a methylene group from the β-methoxyacrylate moiety (either as a result of hydrolysis or O-demethylation) of speciociliatine (Figure S6).

O-Demethylated or hydrolyzed+mono-oxidative metabolites (Met 2, 5, and 9)

Met 2, 5, and 9 showed molecular ion peaks at m/z 401.2071 as a result of loss of a methylene group (−14 Da) and addition of an oxygen atom (+16 Da) to speciociliatine. The MS/MS spectrum of Met 2 showed a fragment ion at m/z 160.0757 representing O-demethylation of the methoxy group on the ethyl methoxyindole moiety. A fragment ion at m/z 242.1387 as a result of C-C cleavage of the quinolizine moiety forming an ethylpiperidine β-methoxyacrylate daughter ion with an additional hydroxy group indicating the hydroxylation of the ethylquinolizine moiety (Figure S7). In the case of Met 5, the product ion spectrum showed a prominent fragment ion at m/z 174.0913 representing unchanged 9-methoxyindole and thus O-demethylation or hydrolysis of the methyl β-methoxyacrylate moiety. Also the fragment ion at m/z 271.1441 related to methoxyindolepyridobutyl moiety with an additional oxygen atom denoting hydroxylation on the quinolizine group of speciociliatine (Figure S7). The fragmentation of Met 9 showed fragment ions at m/z 190.0863 (174+16 Da) and 299.1754 indicating hydroxylation of the ethyl methoxyindole moiety and thus, O-demethylation or hydrolysis of the β-methoxyacrylate moiety of speciociliatine (Figure S7).

Di-oxidative+dehydrogenated metabolite (Met 26)

Met 26 showed a molecular ion peak at m/z 429.2020 as a result of the addition of two oxygen atoms (+32 Da, di-oxidation) and the loss of two hydrogen atoms (−2 Da, dehydrogenation). The MS/MS product ion spectrum of Met 26 showed unchanged fragments at m/z 238.1438, 129.0546, and 110.0964 implying no modification of the ethylpiperidinyl methyl β-methoxyacrylate moiety of speciociliatine while fragment ions at m/z 204.0655 and 313.1547 suggest di-oxidation and dehydrogenation of the ethyl methoxyindole moiety of speciociliatine (Figure S8).

O-Demethylated or hydrolyzed+dehydrogenated metabolites (Met 28, 30, and 33)

Met 28, 30, and 33 showed molecular ion peaks at m/z 383.1965 as a result of the loss of a methylene group (−14.0157 Da) and two hydrogen atoms (−2 Da, dehydrogenation). The MS/MS product ion spectrum of Met 28 showed a product ion at m/z 160.0757 meaning O-demethylation of the methoxyindolo moiety, and product ions at m/z 213.1022 and 267.1492 corresponding to the dehydrogenated ethylpyridoindole and indoloethylquinolizine moieties, respectively, suggesting dehydrogenation of the ethylindoloquinolizine moiety (Figure S9). The MS/MS product ion spectra of Met 30 and 33 showed unchanged product ions at m/z 174.0913 implying O-demethylation or hydrolysis of the methyl β-methoxyacrylate moiety and the product ion at m/z 281.1648 is associated with the loss of two hydrogen atoms from the methoxyindoloethylquinolizine moiety, referring to dehydrogenation at the ethylquinolizine moiety (Figure S9).

O-Demethylated or hydrolysis+di-dehydrogenated metabolite (Met 34)

Met 34 showed a molecular ion peak at m/z 381.1809 as a result of the loss of a methylene group (−14 Da, O-demethylation or hydrolysis) and four hydrogen atoms (−4 Da, di-dehydrogenation). The MS/MS product ion spectrum of Met 34 showed product ions at m/z 237.1022 and 279.1492 representing the methylethylpyridomethoxyindolo and methoxyindoloquinolizine moieties with loss of 4 hydrogen atoms indicating O-demethylation or hydrolysis of the methyl β-methoxyacrylate and the di-dehydrogenation of methoxyindoloquinolizine moiety of speciociliatine (Figure S10).

O-Demethylated+hydrated+mono-oxidative metabolite (Met 1)

Met 1 showed a molecular ion peak at m/z 419.2177 as a result of the loss of a methylene group (−14 Da, O-demethylation), hydration (+18 Da), and mono-oxidation of speciociliatine. The MS/MS product ion spectrum of Met 1 showed unchanged product ions at m/z 129.0546 and 238.1438 inferring to O-demethylation of the ethyl methoxyindole moiety and the product ion at m/z 194.0812 corresponds to the hydrated, mono-oxidated, and O-demethylated ethyl methoxyindole moiety of speciociliatine (Figure S11).

Direct glutathione conjugated metabolites (Met 16 and 18)

Met 16 and 18 showed molecular ion peaks at m/z 704.2960 in positive ionization mode where the m/z were greater than speciociliatine by +305 Da suggesting conjugation of a glutathione group through substitution rather than an addition to an unsaturated bond which would have increased the m/z of speciociliatine by +307 Da. The MS/MS product ion spectra of Met 16 and 18 showed prominent product ions at m/z 397.2122 due to the characteristic neutral loss of 307 Da from the glutathione conjugated metabolites (Figure S12) (30). The other product ions were not sufficient to identify the site of glutathione conjugation.

Mono-oxidative+glutathione conjugated metabolite (Met 20)

Met 20 showed a molecular ion peak at m/z 720.2909 as a result of conjugation of a glutathione group (+305 Da) through substitution to a mono-oxidative metabolite (m/z 415.2227) of speciociliatine. The MS/MS product ion spectrum of Met 20 showed a product ion of m/z 447.1948 produced by the loss of 273 Da formed by cleavage of the cysteinyl C-S bond. The product ion at m/z 222.0583 corresponds to the ethyl 9-methoxy indole product ion (174 Da) of speciociliatine with the addition of 48 Da (addition of O- and S- atom), also the unchanged product ions at m/z 238.1438 and 129.0546 suggest the glutathione conjugation and mono-oxidation on the methoxyindole group of speciociliatine (Figure S13) (30, 31).

O-Demethylated or hydrolyzed+glutathione conjugated metabolites (Met 4, 8, 10, and 12)

Met 4, 8, 10, and 12 showed molecular ion peaks at m/z 690.2803 as a result of conjugation of a glutathione group (+305 Da) through substitution to an O-demethylated or hydrolyzed metabolite (m/z 385.2122) of speciociliatine. The MS/MS product ion spectra of Met 8 and 10 showed product ions of m/z 417.1843 produced by the loss of 273 Da due to the cleavage of the cysteinyl C-S bond, additionally a product ion of m/z 238.1438 referring to unchanged ethyl piperidinyl β-methoxyacrylate moiety of speciociliatine denotes O-demethylation of the methoxyindole moiety of speciociliatine (Figure S14). Further, the product ions of Met 8 at m/z 465.1438 and of Met 10 at m/z 192.0478 correspond to O-demethylated ethylmethoxyindole moiety with glutathione or thiol group, respectively, suggestive of the glutathione conjugation on the methoxyindole moiety of speciociliatine (Figure S14). In the case of Met 4 and 12, the MS/MS product ion spectra were not captured hence the sites of modification were not established.

Mono-oxidative+glucuronide conjugated metabolites (Met 13 and 21)

Met 13 and 21 showed molecular ion peaks at m/z 591.2548 as a result of conjugation of a glucuronide group (+176 Da) to the hydroxyl group of a mono-oxidative metabolite (m/z 415.2227) of speciociliatine. The MS/MS product ion spectra of Met 13 and 21 showed characteristic product ions at m/z 415.2227 produced by loss of a glucuronide moiety (−176 Da). Additionally product ions at m/z 190.0863 and 238.1438 indicate hydroxylation and glucuronide conjugation at the ethyl methoxyindole moiety of speciociliatine (Figure S15).

O-Demethylated or hydrolyzed with or without mono or di-dehydrogenated+glucuronide conjugated metabolites (Met 3, 6, 7, and 14)

Met 3 and 7 showed molecular ion peaks at m/z 559.2286 as a result of conjugation of a glucuronide group (+176 Da) to the O-demethylated or hydrolyzed+dehydrogenated metabolite (m/z 383.1965). The MS/MS product ion spectrum of Met 3 showed a characteristic product ion at m/z 383.1965 due to the loss of a glucuronide moiety, and the product ions at m/z 213.1062 and 267.1492 belongs to O-demethylated+dehydrogenated indolomethylpyrido and indoloquinolizine moieties, respectively, suggesting O-demethylation at the 9-methoxy position+glucuronide conjugation and dehydrogenation at the quinolizine moiety of speciociliatine (Figure S16). Similarly, Met 6 showed a molecular ion peak at m/z 557.2130 as a result of conjugation of a glucuronide group (+176 Da) to the O-demethylated+di-dehydrogenated metabolite (m/z 381.1809) with MS/MS product ions at m/z 381.1809 and 265.1335 demonstrating the O-demethylation at the 9-methoxy position+glucuronide conjugation and di-dehydrogenation at the quinolizine moiety of speciociliatine (Figure S16). Met 14 showed a molecular ion peak at m/z 561.2443 as a result of glucuronide conjugation to the O-demethylated or hydrolyzed metabolite (m/z 385.2122). In the case of Met 7 and 14, MS/MS product ion spectra were not captured hence the sites of modification were not established.

Mono-oxidative+sulphate conjugated metabolites (Met 19 and 24)

Met 19 and 24 showed molecular ion peaks at m/z 495.1796 as a result of conjugation of a sulphate group (+80 Da) to the hydroxyl group of a mono-oxidative metabolite (m/z 415.2227) of speciociliatine. The MS/MS product ion spectra of Met 19 and 24 showed characteristic product ions at m/z 415.2227 produced by loss of sulphate moiety (−80 Da). Additionally, product ions at m/z 190.0863 and 238.1438 indicate hydroxylation and sulphate conjugation at the ethyl methoxyindole moiety of speciociliatine (Figure S17).

Contribution of CYP enzymes in speciociliatine metabolism using rCYP and HLMs

The incubation of speciociliatine with various human rCYP enzymes demonstrated that rCYP3A4 plays a major role in the metabolism of speciociliatine with minor contribution by rCYP2D6 while other tested rCYPs had a negligible role (Figure 4). Similarly, CYP3A4 inhibitors CYP3cide and ketoconazole markedly inhibited the metabolism of speciociliatine in HLMs, as shown in Figure 5, indicating major involvement of CYP3A4 in the metabolism of speciociliatine. Additionally, the contributions of each of these CYP enzymes in both rCYP and HLM on the formation of primary metabolites of speciociliatine (Met 11, 15, 23, 25, and 29) were also evaluated (Figure 4 and 5). The other minor monohydroxylated metabolites Met 22, 27, 31, and 32 were not detected under these incubation conditions hence enzymes involved in their formation could not be evaluated. The formation of Met 11, 15, 23, and 25 in HLM was inhibited by ketoconazole and CYP3cide and this was consistent with the data observed in rCYP incubations suggesting CYP3A4 plays a predominant role in the formation of Met 11, 15, 23, and 25 with a minor contribution of CYP2D6 (Figure 4 and 5). In contrast, Met 29 formation was markedly inhibited by quinidine (a CYP2D6 inhibitor), showing that the formation of Met 29 was catalyzed by rCYP2D6 enzyme (Figure 4 and 5). The cross-reactivity of CYP2B6 inhibitor towards the CYP2D6 enzyme resulted in inhibition in the formation of Met 29 in HLM (Figure 5). Additionally, as described in section 2.5, speciociliatine metabolism in HLM in the presence of chemical inhibitors of CYP450 enzymes was used to calculate the fraction metabolized by CYP3A4 (f_m,_CYP3A4) and it was found to be 0.55.

Fig. 4. The first order disappearance of speciociliatine (SPC) following incubation with individual recombinant cytochrome P450 (rCYPs) and contribution (%) of rCYPs to the formation of speciociliatine metabolites.

The top left plot represents the first order disappearance of speciociliatine in presence of rCYPs (A). The percent contribution of each rCYPs in the formation of Met 11 (B), 15 (C), 23 (D), 25 (E), and 29 (F) are shown in the remaining bar graphs.

Fig. 5. Metabolism and metabolite formation (%) of speciociliatine (SPC) in the presence of selective chemical inhibitors of CYP450s in human liver microsomes.

The top left graph (A) represents the contribution of CYP450s to speciociliatine metabolism in HLMs. The percent metabolite formation of Met 11 (B), 15 (C), 23 (D), 25 (E), and 29 (F) in HLM in the presence of selective chemical inhibitors of CYP450s are shown in the remaining graphs.

Assessment of permeability of speciociliatine using Caco-2 cell lines

The P_{app (A->B)} values for the positive controls propranolol and atenolol were found to be 21.79 ± 2.43 and 5.79 ± 0.53 ×10⁻⁶ cm/sec, respectively, validating the integrity of the Caco-2 monolayer throughout the experiment. The P_{app (A->B)} and P_{app (B->A)} values for digoxin (P-gp substrate) were found to be 4.28 ± 2.05 and 22.80 ± 3.15 cm/sec, respectively, with an efflux ratio of 5.33, validating the expression of P-gp in the Caco-2 monolayer. The P_{app (A->B)} and P_{app (B->A)} values for speciociliatine were found to be 14.18 ± 0.98 and 22.90 ± 1.87 cm/sec, respectively, with an efflux ratio of 1.61 suggesting that speciociliatine has high permeability with minimal to no efflux of speciociliatine through the intestine.

DISCUSSION

The metabolic stability of speciociliatine in cross-species hepatocytes revealed that speciociliatine has the highest rate of metabolism in cynomolgus monkeys and rats followed by mice while humans and dogs have the slowest rate of metabolism. The extraction ratios, calculated as the ratio of predicted hepatic clearance to the corresponding species liver blood flow rate (90, 55, 31, 44, and 21 ml/min/kg in mouse, rat, dog, cynomolgus monkey, and human, respectively), were found to be 0.9 in rodents and cynomolgus monkeys, 0.4 in dogs, and 0.5 in humans, implying that speciociliatine would be a high clearance compound in rodents and cynomolgus monkeys and a moderate clearance compound in humans and dogs (32). The stability of speciociliatine in humans is comparable to previously reported liver microsomal stability of speciociliatine (t_1/2 = 41.8 min) while there is no report on the metabolic stability of speciociliatine in other species (16). The predicted in vitro hepatic clearance of speciociliatine in rats (49.2 ml·min/kg) was found to be 4-fold greater than the reported in vivo clearance (11.6 ml·min/kg) following an intravenous administration of 2.5 mg/kg speciociliatine in male rats (2). Although this is an acceptable prediction, improvement in the prediction could be achieved by incorporating correction factors such as plasma protein binding, blood to plasma partition ratio, and non-specific binding when extrapolating the in vitro hepatic clearance using the well-stirred model. Based on the Caco-2 permeability data, speciociliatine was found to be a highly permeable compound and the efflux ratio of <2 indicates a negligible influence of efflux transporters, i.e., P-gp. Collectively, the lower metabolic rate of speciociliatine and higher permeability with no involvement of P-gp transporters could potentially result in a low first-pass effect and high oral bioavailability of speciociliatine in humans.

Further, to reliably translate preclinical animal safety and toxicity data to humans, it is imperative to know that humans would also be exposed to the same metabolites present during the safety or toxicity studies conducted in the animal models. Thus, in vitro metabolite profiling of the drug in different animal species is important to identify the animal species that most closely resembles humans in terms of metabolite profile for toxicity evaluation. As toxicity studies are generally performed in both rodents and non-rodents, the interspecies comparison of the in vitro metabolite profiles of speciociliatine were determined in mouse, rat, dog, cynomolgus monkey, and human hepatocytes. Upon incubation with liver microsomes and hepatocytes, speciociliatine was extensively metabolized to multiple metabolites across species. All human metabolites of speciociliatine were detected at appreciable amounts in both dog and cynomolgus monkey. The proposed metabolic pathways of speciociliatine across species liver microsomes and hepatocytes are shown in Figure 6.

Fig. 6. Proposed metabolic pathways of speciociliatine in mouse, rat, dog, cynomolgus monkey, human liver microsomes and hepatocytes.

The red boxes and the blue labels represent the probable site of metabolism and metabolic changes of the speciociliatine, respectively.

O-demethylation of the 9-methoxyindole group (Met 23) was found to be one of the major metabolic pathways across species contributing about 18% in mouse, 21% in rat, 16% in dog, 24% in cynomolgus monkey, and 28% in human hepatocytes to the total metabolism. The other major metabolic pathway was hydroxylation at the indoloquinolizine resulting in multiple metabolites (Met 11, 15, 22, 25, 27, 31, and 32). These hydroxylated metabolites contributed to approximately 44, 42, 77, 34, and 28% of the total metabolism in mouse, rat, dog, cynomolgus monkey, and human hepatocytes, respectively. For this estimation, the contribution of co-eluting metabolites in the case of Met 15 and 25 were assumed to be minimal. Among the hydroxylated metabolites, Met 11 was found to be 7-hydroxyspeciociliatine and its contribution to the total metabolism in hepatocytes was higher in human and cynomolgus monkey (7%) compared to rodents (<4%) and dog (<2%). Unlike 7-hydroxymitragynine, 7-hydroxyspeciociliatine has not shown potent μ-opioid activity (33). Met 29, representing the loss of methylene group on methyl β-methoxyacrylate moiety as a result of either hydrolysis or O-demethylation, was one of the major metabolites in mouse, cynomolgus monkey, and human hepatocytes contributing 4, 14, and 17% to the total metabolism, respectively, while it was found at trace levels in rat and dog hepatocytes. Several other phase I metabolites, which were the product of downstream metabolism of the primary metabolites of speciociliatine, were detected across species hepatocytes incubation at varying abundances as shown in Figure 6.

Further, multiple glutathione conjugates of speciociliatine (Met 16 and 18) and its primary metabolite (Met 4, 8, 10, 12, and 20) were detected. All of these metabolites were a product of nucleophilic substitution of glutathione on speciociliatine or its primary metabolite rather than a nucleophilic addition. Similar to mitragynine, speciociliatine also did not appear to undergo a glutathione conjugation on the β–methoxyacrylate moiety through nucleophilic addition (11). In most cases, the MS/MS spectra of these conjugated metabolites were not conclusive enough to identify the site of metabolism except for Met 8, 10, and 20 where the glutathione conjugation was found to be on the ethylmethoxyindole moiety of speciociliatine (Figure 6). As speculated for mitragynine bioactivation, in the case of speciociliatine, Met 8 and 10 could potentially have formed through either oxidation to form a quinolizidium reactive intermediate or dehydrogenation to a 3-methylindolenine-like reactive intermediate (34). While Met 20 could be a product of glutathione conjugation of quinoneimine reactive intermediates (34), the exact mechanism of bioactivation of speciociliatine is unknown. The other phase II metabolites of speciociliatine were the glucuronide conjugates of either the hydroxylated or O-demethylated metabolites of speciociliatine, while two sulfate conjugated metabolites were found to be the products of the hydroxylated metabolite of speciociliatine (Figure 6). The metabolite profiling of speciociliatine in rat and human urine, where urine samples were obtained from oral 40 mg/kg speciociliatine singly dosed rats and human samples from regular kratom users, was consistent with our current findings for both phase I and II metabolites (20). Overall, no substantial differences in the metabolite profile of speciociliatine were observed in cross-species hepatocytes incubation with no human-specific or disproportionate metabolites. The cynomolgus monkey was the non-rodent species found to have a metabolite profile most similar to humans.

Additionally, similar to mitragynine, the reaction phenotyping of speciociliatine using the rCYPs systems and the HLM incubation with specific CYP450 inhibitors show that CYP3A4 plays a predominant role in the metabolism of speciociliatine in humans (Figures 3 and 4). Thus, although the difference in the C3 configuration in speciociliatine (R) and mitragynine (S) induces a significant change in the spatial shape of the molecules, it does not affect the CYP3A4 enzyme selectivity of their metabolism (2). With the exception of Met 29, the formation of most of the primary metabolites of speciociliatine was largely catalyzed by CYP3A4 while Met 29 formation was predominantly catalyzed by CYP2D6. Based on the liver microsomal incubation in the presence of CYP450 inhibitors, the fraction of speciociliatine metabolized by CYP3A4 (f_{m, CYP3A4}) was found to be 0.55 indicating that the pharmacokinetics of speciociliatine could be substantially influenced by perpetrator drugs that can influence the expression or activity of the CYP3A4 enzyme. The increase in the systemic exposure of speciociliatine following an oral dose of speciociliatine along with reversible CYP3A4 inhibitors could be predicted by the Rowland-Matin equation (35):

$$\frac{{AUC}_{inh}}{{AUC}_{cntrl}}=\frac{f_{g,inh}}{f_{g,cntrl}}\times \frac{1}{\left(\frac{f_{m,CYP3A4}}{1+\left(\frac{{\left[I\right]}_{invivo}}{K_i}\right)}\right)+\left({1-f}_{m,CYP3A4}\right)}$$

Where AUC_inh and AUC_cntrl represent the area under the curve of speciociliatine in presence and absence of the CYP3A4 inhibitor, f_{g, inh} and f_{g, cntrl} represent the fraction of the speciociliatine passing through the intestinal wall in the presence and absence of the CYP3A4 inhibitor, and [I]_{in vivo} and K_i represent the concentration of the CYP3A4 inhibitor in vivo and dissociation constant of the inhibitor-enzyme complex, respectively. To predict the change in the exposure of speciociliatine in presence of a known CYP3A4 inhibitor, ketoconazole, the reported unbound values of [I]_{in vivo} and K_i as 0.33 and 0.015 μM, respectively, were used and the contribution of gut metabolism was assumed to be negligible. The exposure of speciociliatine would increase by 2.1-fold following coadministration of a potent CYP3A4 inhibitor suggesting moderate clinical drug-drug interaction.

CONCLUSION

Speciociliatine showed appreciable metabolic stability across species hepatocytes with high intestinal permeability. The interspecies metabolite profile comparison revealed that no human-specific or disproportionate metabolites of speciociliatine were observed in the hepatocyte incubations. Based on their metabolite profiles, cynomolgus monkey or dog could be used as non-rodent species for the toxicological evaluation of speciociliatine. The CYP450 reaction phenotyping study revealed that the clearance of speciociliatine is predominantly mediated by the CYP3A4 enzyme, with minor contributions by CYP2D6. This is the first report which has evaluated the metabolic stability and identified the metabolites across species and CYP450 enzymes responsible for the metabolism of speciociliatine. Based on the speciociliatine levels in kratom leaves, its high permeability and low clearance have resulted in higher systemic exposure of speciociliatine in humans. Additionally, considering its high affinity towards opioid receptors, speciociliatine could play a significant role in the overall pharmacology of kratom warranting further studies to evaluate these claims.