Characterizing Oxidative Metabolites of 6-Methylnicotine (6MN; aka Metatine^™): Divergent Metabolism from Nicotine and Identification of Urinary Biomarkers of Exposure

Abstract

The emergence of synthetic nicotine analogs in “tobacco-free” products, such as 6-methylnicotine (6MN; aka Metatine^™) in SPREE BAR, presents new regulatory and public health challenges. Alarmingly, little is known about metabolism of 6MN, its potential biomarkers of exposure, or its toxicity. In this study, we systematically characterized oxidized metabolites of 6MN in urine of mice exposed to 6MN (via intraperitoneal or inhalation route) using liquid chromatography–high resolution mass spectrometry (LC-HRMS). Similarly, human urine samples were analyzed for 6MN metabolites after use of SPREE BAR (Blue Razz Ice) product. Nine 6MN metabolites were identified in mouse urine, and each metabolite corresponded with a known nicotine metabolite albeit with increased mass (i.e., m/z +14 Da). Although 6MN and nicotine share oxidative routes, the metabolism of 6MN was dominated via N-oxidation (likely FMO3-mediated) than C-oxidation (likely CYP2A6-dependent) pathways whereas nicotine metabolism is vice versa. Six 6MN metabolites were detected in human urine after SPREE BAR use, demonstrating strong cross-species metabolic concordance. Among these 6MN human metabolites, 6-methylcotinine, 6-methyl-3’-hydroxycotinine, and 6-methylcotinine-N-oxide emerged as potential urinary biomarkers of exposure due to their prevalence. Importantly, 6MN, yet not an equimolar dose of nicotine, induced acute neurotoxic effects in mice, highlighting distinct toxicological risks of 6MN compared with nicotine. This research revealed a distinct metabolic profile of 6MN and established a framework for biomonitoring of 6MN exposure. Together, these findings advanced our understanding of metabolism of synthetic nicotine analogs and emphasized the importance of compound-specific profiling to support regulatory oversight of emerging nicotine-like products.

Introduction

Although nicotine (NIC, C₁₀H₁₄N₂) is the main addictive ingredient in tobacco-derived products and is relatively inexpensive to extract from tobacco, recent FDA- Center for Tobacco Products (CTP) regulatory actions have spurred the emergence of products with chemically synthesized NIC and its analogs, marketed as “tobacco-free”, to bypass established regulations set by US Congress and carried out by the FDA (Erythropel et al. 2024). For example, NIC extracted from tobacco is 98% in the S-form (Salam et al. 2023) but industry is now synthesizing S-NIC and racemic R- and S-NIC forms in order to advertise “tobacco free” products containing fewer known tobacco-derived carcinogens, e.g., nitrosamines (Emery et al. 2023), and also to claim that these products are exempted from FDA-CTP regulatory oversight (Jordt 2023). More recently, a disposable electronic cigarette known as SPREE BAR that contains Metatine^™ was introduced into the market as a nicotine-free alternative to nicotine-containing e-cigarettes (Jordt and Jabba 2024; Jordt et al. 2024). In this rapidly evolving market, other products, such as Kumi-Six and Aroma King, have also been reported to contain Metatine (Erythropel et al. 2025; Vanhee et al. 2024).

Chemically, Metatine is 6-methylnicotine (6MN, C₁₁H₁₆N₂), a synthetic analog of NIC that contains an additional methyl group at the 6-position of the pyridine ring (Jordt et al. 2024), though recent evidence suggests that trace levels of naturally formed 6MN present in tobacco leaf and in tobacco products (Pankow et al. 2025). Currently, there is limited information on 6MN including: how widespread 6MN is in non-nicotine products (e.g., SPREE BAR, Kumi-Six, and Aroma King are 6MN containing products); how addictive it is; its metabolism; and, its toxicity (Jordt et al. 2024). 6MN was synthesized in 1967 (Haglid 1967) and tested for its biological activity in the 1980s (Seeman et al. 1985; Seeman et al. 1981). In some bioassays, 6MN was 3-times more potent than NIC including for convulsions (ED₅₀) in rats in vivo (Jordt et al. 2024; Rylander 1982). However, it appears that there is no prior reported study of the metabolism of 6MN. Thus, it is critical to characterize 6MN metabolism especially if we are to assess the breadth of its presence in the marketplace and its use in the diverse and ever evolving tobacco landscape (Erythropel et al. 2024).

To address this knowledge gap, we developed a liquid chromatography-mass spectrometry (LC-MS) approach to identify and characterize the nature of 6MN metabolism. Due to its structural resemblance to NIC, we hypothesized that its metabolic pathways are likely similar and that its metabolites could be identified readily through comparison with the corresponding metabolites of NIC. To characterize 6MN metabolism, we first generated urinary samples in mice following intraperitoneal exposure to either 6MN or NIC for comparison. We then analyzed for urinary metabolites by LC-MS via an integrated library guided analysis (ILGA) that enabled us to identify 6MN metabolites based on known NIC metabolites. After that, we extended our analysis to urine samples collected from both mice after inhalation exposure to SPREE BAR-derived aerosols and humans who use SPREE BAR. This cross-species analysis allowed us to independently validate the use of 6MN metabolites as biomarkers for monitoring exposure to this novel compound. In an unplanned comparison, we found acute neurotoxic effects of 6MN in mice at a dose that was not present with NIC exposure at equal molar doses. Together, these results offer comparative insights into: 1) the utility of urinary biomarkers for detecting 6MN exposure; 2) the surprising difference in metabolism of 6MN and NIC; and 3) potential concerns regarding the neurotoxicity of 6MN-containing products.

Materials and Methods

Chemicals and reagents

(S)-6-methylnicotine (6MN; aka Metatine; CAS: 13270-56-9, 99.93% pure, 25 mg as oil) was purchased from Toronto Research Chemicals (Toronto, CAN). (−)-Nicotine di-(+)-hydrogen tartrate dihydrate (CAS: 6019-06-3) was purchased from Acros Organics. Acetonitrile, water (both UHPLC-MS grade), ammonium acetate (LC-MS grade), and Infinity^™ creatinine liquid stable reagent were purchased from ThermoFisher Scientific Inc. (Waltham MA). (−)-Cotinine (COT) and (−)-Nicotine (NIC) were purchased from Sigma-Aldrich (Burlington, MA). trans-3’-Hydroxy Cotinine (3HC) and (RS)-Nornicotine (NNIC) were purchased from LGC Standards (Manchester, NH).

Murine exposures and urine collection

For intraperitoneal (ip) injections of 6MN, healthy adult male (n=2) and female (n=8) C57BL/6J mice (20–30 weeks old; in-house colony)(Jin et al. 2024) were injected with 6MN (MW=176.26 g/mol) (ip, 1 mg/kg bodyweight (bwt) in sterile saline, male mice only; 0.1 and 0.01 mg/kg bwt in sterile saline, female mice only) or NIC ditartrate dihydrate (MW=498.44 g/mol) (ip, 0.283 and 0.0283 mg/kg bwt, female mice only). Female mice were split into 2 groups in a crossover design: Group 1 (n=4) received 6MN (ip, 0.1 mg/kg bwt) first and then 3 days later NIC (0.0287 mg/kg bwt). Group 2 (n=4) received NIC (ip, 0.283 mg/kg bwt) first and then 3 days later 6MN (0.01 mg/kg bwt). NIC dose was adjusted to 2.83x that of 6MN due to differences in compound MW (i.e., 176.26 vs 498.44 = 2.827). Notably, acute neurotoxicity (e.g., seizures) was observed immediately after administration of the highest dose of 6MN, and thus, the urine collected after the highest dose was not included in subsequent analyses. In this study, mice injected with 0.1 mg/kg bwt of 6MN or 0.283 mg/kg bwt of nicotine (NIC) were classified as the “high-dose” group, while those given 0.01 mg/kg bwt of 6MN or 0.0283 mg/kg bwt of NIC were classified as the “low-dose” group. All mice were observed carefully after administration of 6MN or NIC for potential toxic effects.

For inhalation exposures, female mice (n=6) were exposed to filtered room air for 3h before commencement of urine collection protocol as described below (Jin et al. 2024). Six days after exposure, mice were then exposed to 6MN (1% w/v) in PG:VG (30:70, v:v). Mice were placed into whole body exposure chambers (5L) and exposed to a series of 9-min sessions of 18 puffs each session for a total of 10 sessions over 3h. Each puff was 91.1 mL over 4 s at 2 puffs per min. Flexiware software (SCI-REQ) regulated e-cigarette-smoking robot (inExpose; SCI-REQ) puffing and bias air flow during non-puffing at 3 LPM as described (Jin et al. 2024). 6MN e-cigarette-like solution was loaded into a refillable Kangertech clear tank atomizer (3 mL, 1.5 Ohm; SCI-REQ) powered by a Joyetech eVIC-VTC mini battery at a power of 10W. A CEL-712 particulate matter monitor (Casella) was positioned upstream of the chamber and continuously recorded total suspended particulate mass (TSP) during the exposure. Similarly, SPREE BAR Blue Razz Ice e-liquid was removed from a SPREE BAR device and loaded into the SCI-REQ system as described above. The puffing profile was identical for both e-cigarette-like exposures.

For mouse urine collections, immediately following injection or inhalation exposures, each mouse was placed singly into a urinary collection chamber as described previously (Jin et al. 2024). Mice were provided a glucose:saccharin (3.0:0.125%) solution in water to stimulate polydipsia and diuresis. Urine was collected twice: early (0–4h) and late (4–18h; overnight) collections post-exposure. Urine was collected in chilled (4 °C) graduated cylinders to preserve metabolite integrity and retard bacterial growth. After the volume was measured, urine was centrifuged, and then urine was aliquoted for storage at −80 °C before analyses, which was performed within 2–3 months after collection.

Human Exposures to SPREE BAR and Urine Collection

Urine samples were collected under IRB-approved protocols at the University of Louisville (IRB #11.0432 and #15.0097) using two distinct exposure protocols. For the SPREE BAR exposures (IRB #11.0432), urine was collected from exclusive daily electronic cigarette users (n=2) who self-reported abstinence from tobacco and nicotine overnight. Participants used their own SPREE BAR device (Blue Razz Ice; ~0.6 wt% 6MN) ad libitum early in the morning, shortly after waking. Each participant reported taking at least 10 puffs over approximately 10 minutes. A clean-catch urine sample was then collected at home approximately two hours after waking and transported to the research clinic by the participant.

For the NIC exposure (IRB #15.0097), participants enrolled in the Reactive Aldehydes in Tobacco Study (RATS), as previously described (Lorkiewicz et al. 2019, 2022) abstained for 48 hours from all tobacco products, e-cigarettes, nicotine, marijuana, and other illicit substances, and avoided caffeine, alcohol, and grapefruit juice for 8 hours prior to the clinic visit. Upon arrival, participants provided a clean-catch baseline urine sample before using an NJOY King e-cigarette (2.4% nicotine) ad libitum for no more than 15 minutes, with a minimum of 15 puffs. Follow-up urine samples were collected in the research clinic area after exposure and after the initial void. All product use and sample collection for this protocol were conducted under supervision by trained study personnel. Urine (n=2) for this study collected 2h after use was used for comparison with the samples collected from SPREE bar users.

In contrast to the mouse studies, which used defined sampling intervals to assess time-resolved metabolite profiles, human urine samples were collected at a single time point with the primary goal of confirming the presence of 6MN metabolites following real-world exposure.

LC-MS analysis

Liquid chromatography-high resolution mass spectrometry (LC-HRMS) in data independent acquisition (DIA) mode was used for metabolite discovery and relative quantification in the mouse urine collected post injection or inhalation. Briefly, urine samples were diluted 1:10 with 15 mM ammonium acetate and then analyzed using an Acquity I-Class UPLC system with an Acquity HSS T3 1.8 um, 2.1 X 150 mm UPLC column (Waters, MA). Separation was achieved with a gradient comprising 15 mM ammonium acetate (solvent A) and acetonitrile (solvent B). The gradient profile started with holding 3% of solvent B for 0.58 min, increased to 5% of solvent B over 1.3 min, to 10% of solvent B over 0.7 min, to 30% of solvent B over 1.35 min, and to 60% of solvent B over 3.01 min. Then the gradient decreased to 15% of solvent B over 0.34 min and to 10% of solvent B over 0.3 min. After that, the gradient returned to the initial conditions over 0.36 min and re-equilibrated for 3 min before the next injection. Mass Spectrometry Elevated Energy (MS^E) data (Zhao and Lin 2014) were collected from m/z 40–600 using Waters Synapt XS MS (Waters, MA) and an electrospray ion source operated in positive mode, with low collision at 2 V (function 1) and high collision energy ramping from 10 to 40 V (function 2). To validate LC-MS performance, one quality control (QC) sample, prepared by pooling equal amount of each urine sample, was injected after every ten urine sample injections.

For structural confirmation of 6MN metabolites, tandem liquid chromatography quadrupole time of flight mass spectrometry (LC-QTOF-MS/MS) data collection was performed under conditions the same as function 2 of the LC-HRMS-DIA analysis. The resulting MS/MS spectra were averaged around their respective retention time (Rt) and examined for the mass shift to their equivalents of NIC metabolites.

Structural identification and validation

An ILGA approach was employed to identify urinary metabolites of 6MN (Xie et al. 2023). Given that 6MN is a synthetic NIC analog with a similar structure, we hypothesized that its metabolism would parallel that of NIC, producing chemically analogous metabolites. To test this, we retrieved nine major phase I NIC metabolite structures (>1% of total metabolites) from the SciFinderⁿ database (Murphy 2021) (Table S1). For each compound, we generated a corresponding 6MN analog by adding a methyl group at the 6-position of the pyridine ring (Table S2). These modified structures were then imported into a custom compound structure library. This library was subsequently used to analyze MS^E data from mouse urine samples.

Mouse urine was collected after i.p. of 6MN or NIC and analyzed by LC-MS^E. Peaks corresponding to putative 6MN metabolites were annotated using a three-level confidence system for annotation. Level 1 assignments were confirmed using authentic standards, matching both exact mass (m/z) and retention time (Rt). Level 2 identification applied when the standard for 6MN metabolite was unavailable, but its corresponding NIC metabolite standard (Table 1) was accessible. In this case, the NIC metabolite was identified based on matching m/z and Rt, and the annotation of its corresponding 6MN metabolite was based on a +14.0157 Da mass shift, representing a methyl substitution (H → CH₃). Level 3 assignments relied on a stepwise approach: NIC metabolites were first identified following NIC administration, and the corresponding 6MN analogs were inferred by the same +14.0157 Da shift.

Table 1.

6-Methylnicotine (6MN) and its metabolites identified in mouse urine after intraperitoneal injection.

Acronym	Full name	Corresponding NIC metabolite	Identification confidence level*	[M+H]+ m/z	Retention time (min)
6MN	6-Methylnicotine	Nicotine (NIC)	Level 1	177.1386	4.66
6MNNIC	6-Methylnornicotine	Nornicotine (NNIC)	Level 2	163.1230	4.02
6MCOT	6-Methylcotinine	Cotinine (COT)	Level 2	191.1179	4.74
6M3HC	6-Methyl-3’-hydroxycotinine	3’-Hydroxycotinine (3HC)	Level 2	207.1128	4.34
6MNCOT	6-Methylnorcotinine	Norcotinine (NCOT)	Level 3	177.1022	4.46
6MNNO	6-Methylnicotine-1’-N-oxide	Nicotine-1’-N-oxide (NNO)	Level 3	193.1335	3.96
6MOPBA	6-Methyl-γ-oxo-3-pyridinebutanoic acid	4-Oxo-4-(3-pyridyl) butanoic acid (OPBA)	Level 3	194.0812	3.74
6MHPBA	4-Hydroxy-4-(3-pyridyl-6-methyl)butanoic acid	4-Hydroxy-4-(3-pyridyl)-butanoic acid (HPBA)	Level 3	196.0968	2.70
6MCNO	6-Methylcotinine-N-oxide	Cotinine-N-oxide (CNO)	Level 3	207.1128	3.79

^*:Terminology describing identification was used consistently with earlier sections. Identification confidence levels were assigned based on the criteria defined in Section 2.5: Level 1 – confirmed with 6MN standards (matched m/z and Rt); Level 2 – identified by comparison with standards of the NIC analog; and Level 3 – inferred from NIC metabolites detected in urine following nicotine exposure.

Metabolite structure validation was conducted using LC-QTOF-MS/MS analysis. Urine samples of NIC- and 6MN-treated mice were analyzed under identical LC conditions. Fragmentation patterns of putative 6MN metabolites were compared with those of their NIC analogs. Structural similarity was confirmed when spectra matched directly or exhibited the expected +14.0157 Da shift in diagnostic fragments, consistent with the methyl substitution.

Additional details about the structural identification and validation of 6MN metabolites can be found in Supplementary Materials

Creatinine measurement

Frozen urine samples were thawed on ice, vortexed, and then diluted with water. The dilution ratio was 1:3 for murine urine samples and 1:20 for human urine samples. Urinary creatinine levels were measured using an Ace Axcel Clinical Chemistry System with ACE Creatinine reagent (Alfa Wassermann, West Caldwell, NJ) (Srivastava et al. 2025).

Relative Quantification and Statistical Analysis

Raw LC-MS^E data were processed using UNIFI 1.9 (Waters, MA) and queried against a compound library containing the identified 6MN and NIC metabolites. Instrument response, defined as the three-dimensional chromatographic peak volume, was used to estimate the relative abundance of each metabolite in mouse urine. To account for variation in urine dilution, all metabolite signals were normalized to urinary creatinine levels (Xie et al. 2024). Statistical analyses were conducted using GraphPad Prism (GraphPad Software, MA). For comparisons involving three or more groups, one-way ANOVA was used. Following the ANOVA analysis, Dunnett’s multiple comparisons test was used to identify significant differences relative to a single control (Saline or Air, matched by sampling time) (Lee and Lee 2018). And Tukey’s multiple comparisons test was used for to compares all experimental groups with each other (Lee and Lee 2018). Comparisons between two groups (e.g., NIC vs. 6MN) were performed using unpaired t-tests. Statistical significance was defined as p<0.05 and is indicated by asterisks in all relevant figures and tables.

Results

Identification of 6MN metabolites

Urinary metabolites of 6MN were identified following i.p. injection of the parent compound using the ILGA framework (Xie et al. 2024). Candidate LC-MS peaks for 6MN metabolites were identified based on their expected m/z values (Table S2). Further selection and structural confirmation were achieved by comparing the MS/MS spectra of these peaks to those of authentic standards or to corresponding nicotine metabolites observed in NIC-injected mice (Fig. S1–S9). The identified 6MN metabolites were 6-Methylnornicotine (6MNNIC), 6-Methylnicotine (6MN), 6-Methylcotinine (6MCOT), 6-Methyl-3’-hydroxycotinine (6M3HC), 6-Methylnorcotinine (6MNCOT), 6-Methylnicotine-1’-N-oxide (6MNNO), 6-Methyl-γ-oxo-3-pyridinebutanoic acid (6MOPBA), 4-Hydroxy-4-(3-pyridyl-6-methyl)butanoic acid (6MHPBA), and 6-Methylcotinine-N-oxide (6MCNO) (Fig. 1 and Table 1). These 6MN metabolites are corresponding to NIC metabolites nornicotine (NNIC), Nicotine (NIC), Cotinine (COT), 3’-hydroxycotinine (3HC), norcotinine (NCOT), nicotine-1’-N-oxide (NNO), 4-oxo-4-(3-pyridyl)butanoic acid (OPBA), 4-Hydroxy-4-(3-pyridyl)-butanoic acid (HPBA), and cotinine-N-oxide (CNO)respectively (Table 1).

Figure 1.

Comparison of A. proposed metabolism pathway of 6-Methylnicotine (6MN) and B. reported metabolism pathway of nicotine (NIC) (Murphy 2021). Undetected glucuronides are omitted in the figure. The percentages shown beneath each metabolite reflect relative proportions based on LC-MS instrument responses of murine urine collected at 0–4h post-exposure to high-dose of 6MN or NIC administration. (See “LC-MS analysis” and “Relative Quantification and Statistical Analysis” for details).

Following the discovery procedures, all 9 proposed 6MN metabolites were detected in the LC-MS^E result of murine urine after 6MN injection (Table 1). Each detected 6MN metabolite had a structurally analogous NIC counterpart that was also observed in mouse urine following NIC administration (Table S1). As expected, 6MN metabolites exhibited [M+H]⁺ ions with m/z values +14.0157 Da greater than their NIC analogs (Table S2), reflecting the substitution of a hydrogen atom (H) with a methyl group (CH₃). In addition, the 6MN metabolites showed retention times 0.3–1.3 min longer than their NIC analogs, likely due to reduced polarity resulting from methylation at the 6-position of the pyridine ring. More details on identification of individual 6MN metabolites are provided in the supplementary materials.

Structural confirmation of 6MN metabolites

The structure of each 6MN metabolite was confirmed via MS/MS spectral comparison with its respective NIC analog. As shown in Fig. S1–9, 6MN metabolites consistently exhibited MS/MS peaks shifted by +14.0157 Da, corresponding to fragments retaining the 6-methyl group. Other fragment ions matched those of the NIC metabolites (i.e., no mass shift), reflecting fragments without the methylation site. Mass accuracy of all observed 6MN fragment ions was within ±5 mDa of their theoretical m/z values, whether shifted or unshifted. This agreement further supports the structural validity of the proposed 6MN metabolites. We summarized representative fragment ions observed for each 6MN metabolite (Table S3). More details on structure confirmation of individual 6MN metabolites are provided in the supplementary materials.

Relative quantification of metabolites after injection

All nine of the 6MN metabolites were detected after 6MN administration, yet these were absent in both saline- and NIC-treated groups. The relative levels of 6MN metabolites in mouse urine following administration of 6MN were estimated (Table 2). However, due to the lack of authentic standards for 6MN metabolites, an absolute quantification was not feasible. Instead, the LC-MS peak intensities were normalized to urinary creatinine and then used to estimate relative metabolite abundance. Metabolite levels varied over both dose and time, with higher intensities observed in high-dose groups and at earlier time points. At the early time point (0–4h), all nine of the 6MN metabolites were significantly elevated in the high-dose group relative to saline controls. Notably, 6MNCOT and 6MNNO showed the highest normalized responses and remained elevated even at the later time point. As expected, all nine NIC metabolites were detected exclusively in NIC-treated animals (Table S4), with minimal or no signal of NIC metabolites in both saline and 6MN-treated groups. The NIC metabolite levels also varied with dose and sampling time, with the highest responses observed in the high-dose, early time point group.

Table 2.

Levels of 6MN metabolites in mouse urine after injection of 6MN or NIC.

Response (Creatinine normalized)		6MN	6MNNIC	6MCOT	6M3HC	6MNCOT	6MNNO	6MOPBA	6MHPBA	6MCNO
Early time point	Saline 0–4h (Ctl)	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 1	0 ± 0	0 ± 0	0 ± 1
	NIC high 0–4h	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0
	NIC low 0–3h	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0
	6MN high 0–4h	7 ± 6*	30 ± 8*	121 ± 23*	46 ± 16*	262 ± 71*	1099 ± 375*	25 ± 7*	36 ± 15*	4 ± 2*
	6MN low 0–3h	0 ± 0	3 ± 0	12 ± 3	0 ± 0	20 ± 13	85 ± 28	1 ± 1	2 ± 2	1 ± 1
Late time point	Saline 4–21h (Ctl)	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0
	NIC high 4–21h	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	1 ± 1*
	NIC low 3–21h	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0
	6MN high 4–21h	0 ± 0	0 ± 0	1 ± 1	0 ± 0	8 ± 6*	14 ± 7*	0 ± 0	0 ± 0	0 ± 0
	6MN low 3–21h	0 ± 0	0 ± 0	0 ± 1	0 ± 0	0 ± 0	2 ± 3	0 ± 0	0 ± 0	0 ± 0

Note: Relative quantification was based on LC-MS peak intensity normalized to urinary creatinine. The nine metabolites shown include 6-Methylnicotine (6MN), 6-Methylnornicotine (6MNNIC), 6-Methylcotinine (6MCOT), 6-Methyl-3′-hydroxycotinine (6M3HC), 6-Methylnorcotinine (6MNCOT), 6-Methylnicotine-N-oxide (6MNNO), 6-Methyl-γ-oxo-3-pyridinebutanoic acid (6MOPBA), 4-Hydroxy-4-(3-pyridyl-6-methyl)butanoic acid (6MHPBA), and 6-Methylcotinine-N-oxide (6MCNO).

^*:p < 0.05 compared with the Saline group with matching sampling time.

Proportion of 6MN metabolites and metabolic enzyme preference

As metabolites of 6MN and NIC were equivalent in number (9 for both) and were analogous to each other, we examined whether the relative proportions of the urinary metabolites were also shared. Among all the tested conditions, the high-dose, early time point group exhibited the most consistent and robust metabolite signals (Table 2, 6MN; Table S4, NIC). Thus, the normalized responses of this set were used to calculate average metabolite proportions (Fig. 1A, 6MN; Fig. 1B, NIC). To visualize their relationships, pathway diagrams were generated in which line widths were scaled according to the total response of all metabolites within each branch, with wider lines indicating greater metabolic flux. The results showed that although both compounds formed structurally analogous metabolites, there were substantial differences in relative abundances of metabolites and the overall metabolic flux through different pathways (Fig. 1AB).

The relative proportions of each 6MN metabolite and its corresponding NIC metabolite analog (Table 1) were compared side-by-side (Fig. 2A). The overall distribution of metabolites differed markedly between 6MN and NIC, with seven of the nine ‘metabolite pairs’ showing significantly different relative abundances. Only two pairs, 6MN/NIC and 6MCOT/COT, were similar in their proportions. Notably, both parent compounds (6MN and NIC) accounted for less than 0.5% of total urinary signal, consistent with extensive metabolic processing.

Figure 2.

Comparison of proportions of urinary metabolites between NIC and 6MN. A. Metabolite level; B. Pathway level. For panel A, the proportion of each metabolite was calculated by dividing its response by the total signal from all detected metabolites. In panel B, metabolite signals within the same metabolic pathway were summed to obtain pathway-level proportions. *: p < 0.05 by T-test between 6MN metabolite and its corresponding NIC metabolite.

Of the seven significantly different metabolite analog pairs, three pairs had higher proportions of the 6MN-derived metabolites compared with their NIC-derived counterpart, i.e., 6MNNIC/NNIC, 6MNCOT/NCOT, and 6MNNO/NNO (Fig. 2A). This, of course, aligns well with the higher levels of 6MNCOT and 6MNNO among the 6MN metabolites (Table 2), reflecting their roles as major 6MN metabolites in mice. The remaining four metabolite pairs had lower proportions of 6MN metabolites, with the most pronounced reduction observed for CNO/6MCNO pair, where CNO was 9.2% in NIC-treated mice and 6MCNO just 0.2% in the 6MN group – a more than 40-fold difference. Together, these differences in metabolite levels suggest a broader shift in pathway utilization between the two structurally similar compounds.

Next, we compared 6MN and NIC metabolism on pathway-level (Fig. 2B). Both 6MN and NIC underwent biotransformation through four primary oxidative routes: N-demethylation, N-oxidation, 5′-oxidation, and 2′-oxidation (Fig. 1). N-oxidation was the dominant route for 6MN metabolism, contributing over 67% of total metabolite signal compared with just 20% for NIC. The most prominent 6MN metabolite identified was 6MNNO (Fig. 2A). In contrast, the contributions of 5′-oxidation and 2′-oxidation were markedly reduced in 6MN metabolism -- decreased from over 30% in NIC metabolism to 11% and 4%, respectively. Specific enzymes are associated with metabolism of NIC (Murphy 2021). NIC metabolism is largely attributed to P450-mediated oxidation, including N-demethylation, 5′-oxidation and 2′-oxidation catalyzed by human CYP2A6 (EC:1.14.14.-) (Oscarson 2001), or mouse Cyp2a5 (Poça et al. 2017), that accounts for 77% of metabolite formation. In contrast, based on the result of this study, 6MN metabolism appeared more dependent on flavin-containing monooxygenase 3 (human FMO3; mouse Fmo3; EC:1.14.13.148) mediated N-oxidation (Perez-Paramo et al. 2019) — making up nearly 67% of the total signal (Fig. 2B). Although both 6MN and NIC undergo structurally similar oxidative transformations, their metabolism prefers different enzymatic pathways. Specifically, 6MN metabolism is biased toward FMO3-dependent N-oxidation, whereas NIC primarily undergoes CYP2A6-mediated formation of 3HC and acidic metabolites. Thus, divergence in the metabolite profiles likely reflects differential enzyme preference for the parent compounds.

6MN metabolites in mice exposed to aerosols of PG:VG+6MN and SPREE BAR

To evaluate the metabolic fate of 6MN following inhalation exposure, we measured urinary levels of 6MN-derived metabolites in mice exposed to aerosols of PG:VG+6MN (≈1%) and of SPREE BAR with 6MN (Blue Razz Ice; labeled as “5% Metatine”). Table 3 presents the creatinine-normalized LC-MS responses for each metabolite across both early and late collection time points. All nine metabolites were detected at elevated levels shortly after exposure (0–3h), with six (including 6MNNIC, 6MCOT, 6M3HC, 6MNCOT, 6MNNO, and 6MOPBA) remaining elevated at the later time point (3–18h). At the early time point, SPREE BAR exposure resulted in about twice the metabolite levels (1.6–2.3 folds) observed in the PG:VG+6MN group. The level of urinary 6MN metabolites from PG:VG+6MN exposure at the early time point reached levels comparable to those measured after the high dose (0.1 mg/kg bwt) i.p. injection (Table 2). This finding confirmed that 6MN metabolism was the same regardless of the route of exposure, and thus, 6MN urinary metabolites are likely highly robust and useful biomarkers of real-world exposures to 6MN independent to the exposure routes (e.g., oral or inhaled) or product source (e.g., SPREE BAR).

Table 3.

Levels of 6MN metabolites in mouse urine after inhalation of PG:VG 6MN and SPREE BAR.

Response (Cre		6MN	6MNNIC	6MCOT	6M3HC	6MNCOT	6MNNO	6MOPBA	6MHPBA	6MCNO
Early time point	Air 0–3h (Ctl)	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0
	PG:VG 6MN 0–3h	13 ± 5	34 ± 13*	181 ± 63*	49 ± 14*	279 ± 86*	758 ± 191*	20 ± 7*	35 ± 12*	3 ± 2*
	SPREE BAR0–3h	28 ± 16*	77 ± 24*	412 ± 112*	88 ± 13*	467 ± 61*	1708 ± 274*	32 ± 12*	58 ± 9*	5 ± 1*
Late time point	Air 3–21h (Ctl)	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0	0 ± 0
	PG:VG 6MN 3–21h	0 ± 0	2 ± 1	9 ± 3*	2 ± 3	27 ± 10*	59 ± 25*	1 ± 1*	1 ± 1	0 ± 1
	SPREE BAR3–21h	0 ± 1	2 ± 2*	14 ± 8*	5 ± 4*	37 ± 10*	84 ± 47*	1 ± 1*	1 ± 2	1 ± 0

Note: Relative quantification of 6MN metabolites was based on the peak intensity normalized to creatinine.

^*:p < 0.05 compared with the Saline group with matching sampling time.

To assess whether the route of administration influenced the proportion of 6MN metabolites, we compared the proportion of 6MN metabolites following injection or inhalation (Fig. 3). While notable differences were observed in the metabolites with the greatest abundances – specifically 6MCOT, 6MNCOT, and 6MNNO – the magnitude of these differences between i.p. and inhalation was generally modest. An exception was 6MCOT, whose proportion increased from 8% following i.p. injection to 13–14% with inhalation exposure (Fig. 3). These findings suggest that with comparable dose, the route of exposure had only a minor effect on proportional 6MN metabolism in mice.

Figure 3.

Comparison of proportions of urinary 6MN metabolites between inhalation and injection. The proportion of each 6MN metabolite was calculated by dividing its response by the total signal from all detected 6MN metabolites. *: p < 0.05 by comparison against the injection group (6MN ip high 0–4h); #: p < 0.05 by comparison against the PG:VG+6MN group.

6MN metabolites in human urine

To assess whether 6MN metabolites detected in mice may serve as biomarkers of 6MN exposure in humans, we analyzed urine samples collected from participants who reported use of SPREE BAR Blue Razz Ice – a commercial e-cigarette device containing 6MN (Erythropel et al. 2024). As shown in Fig. 4, six 6MN metabolites – 6MN, 6MNNIC, 6MCOT, 6M3HC, 6MNNO, and 6MHPBA – were detected in human urine after 6MN exposure, with retention times and m/z values matching those observed in the murine exposure studies. Notably, none of these metabolites were found in urine collected after exposure to NIC-containing e-cigarettes. Three 6MN metabolites found in mice, i.e., 6MNCOT, 6MOPBA, and 6MCNO, were undetectable in any human samples (Fig. S10). Among the 6MN metabolites detected in human urine, 6MNNO had the greatest response intensity, which mirrored its highest response in mice following either injection or inhalation. These findings suggest that 6MN metabolites – particularly 6MNNO – may serve as a robust species-independent indicator of 6MN exposure.

Figure 4.

LC-MS results showing 6MN metabolites in human urine after inhalation of SPREE BAR. A. 6MN; B. 6MNNIC; C. 6MCOT; D. 6M3HC; E. 6MNNO; F. 6MHPBA. The extracted ion chromatograms were obtained from LC-MS^E results around theoretical m/z values with a mass tolerance of ± 0.015 Da.

Neurotoxicity of 6MN compared with NIC

Following injection of 6MN (ip, 1 mg/kg bwt), two male mice (n = 2) exhibited acute neurological symptoms, including loss of righting reflex, impaired motor coordination, and seizure-like activity. One mouse experienced a mild seizure lasting 10–15 s and recovered within 20 min. The second mouse exhibited a more severe episode, with convulsions lasting at least 30 s yet making a full recovery within 1h post-injection. Despite the initial symptoms, both animals resumed normal behaviors in an overnight setting that included feeding, drinking, and urination. No additional mice were injected at this dose.

Yet three of 4 female mice given a lower dose of 6MN (ip, 0.1 mg/kg bwt) also had immediate and transient neurological effects. Observed symptoms included rapid, shallow respiration with pronounced diaphragmatic contractions, loss of balance, abnormal gait, head drooping, and brief tremors (~5 s), possibly indicative of mild seizures. One commonly observed behavior – pressing the face or head against the chamber wall (“face pressing”) – was also noted. All symptoms resolved within 1–2 min, and the affected mice displayed normal behavior and physiological functions overnight.

No overt neurological symptoms were observed in female mice injected with either 6MN (ip, 0.01 mg/kg bwt; n=4) or NIC (ip, 0.1 and 0.01 mg/kg bwt; n=4 at each dose). These animals maintained normal activity and physiological behavior post-injection with normal food and water intake and urine production overnight. No abnormal responses were observed in mice exposed to aerosols of either 6MN in PG/VG or SPREE BAR Blue Razz Ice.

Discussion

This study represents the first comprehensive investigation of 6MN metabolism by integrating in vivo exposure models with high-resolution mass spectrometry to characterize metabolite formation across species. Leveraging our ILGA workflow and a custom structure library of nine predicted 6MN metabolites, we systematically profiled urinary metabolites across two routes of administration (ip injection and inhalation) and two species that included human inhalation exposure to SPREE BAR. Using this efficient framework we made several novel findings including: 1) identification and validation of six 6MN metabolites that are candidate biomarkers for tracking human exposure to 6MN; 2) 6MN metabolism pathways are shared with NIC but are proportionally different; and 3) 6MN induced acute (resolving) neurotoxicity in mice at dosage where NIC was without effect.

Appreciating the structural similarity between 6MN and NIC, it is expected that their metabolic pathways would exhibit substantial overlap, which we observed in humans and mice. Both compounds undergo significant biotransformation through four primary oxidative pathways: N-demethylation, N-oxidation, 5′-oxidation, and 2′-oxidation (Fig. 1). These routes yield structurally analogous urinary metabolites wherein for every urinary NIC metabolite detected, a corresponding 6MN analog (m/z +14) is identified in humans and in mice (Table 1). This one-to-one match supports our hypothesis that 6MN and NIC form structurally similar metabolites with similar metabolism pathways. And we identified in mice three metabolites of 6MN (6MNCOT, 6MOPBA, and 6MCNO) that were absent in human urine (albeit we only had 2 human participants in our study).

Despite their shared oxidative pathways, the relative pathway utilization differed distinctively between 6MN and NIC. For example, NIC metabolism is more broadly distributed, with significant contributions from all four oxidative branches. In contrast, 6MN metabolism is skewed toward N-oxidation that makes up over 67% of the relative metabolite signal based on LC-MS peak intensities, compared with just ~20% of NIC-derived metabolites (Fig. 1). 6MNNO is the most abundant 6MN metabolite observed across all tested conditions and models. The underlying mechanism responsible for this shift in N-oxidation metabolism may be changes in enzyme binding or steric hindrance introduced by the methyl group at the 6-position of the pyridine ring. This is likely similar to the inhibitory effect of other NIC analogues on CYP2A6 activity (Lu et al. 2014), which redirects metabolic flux elsewhere. There is also a marked decrease in N-oxidation of the pyridine ring, as evidenced by a >40-fold decrease in the formation of 6MCNO from 6MN (0.2%, Fig. 1A) compared with CNO from NIC (9.2%, Fig. S1B). This decrease also is likely attributable to the steric hindrance introduced by the 6-methyl substitution on the pyridine ring. Together, these findings show that while 6MN follows the same overall oxidative pathways as NIC, its metabolite profile is quantitatively distinct – supporting that structural similarity does not guarantee metabolic equivalence (Lester et al. 2018).

The potential significance of the divergence in enzymatic pathways between NIC and 6MN, namely FMO3- versus CYP2A6-mediated metabolism, may have an influence on the pharmacokinetics, inter-individual variability, and toxicological profiles of 6MN. For example, the NIC metabolite, NNO, formed via FMO3-dependent N-oxidation, undergoes reduction back to NIC in vivo extending the lifetime of NIC and its activity (Dajani et al. 1975; Sepkovic et al. 1986). Thus, it is plausible that 6MNNO may also be reduced back to 6MN (given the structural similarity) potentially prolonging its systemic presence and activity. This enzymatic divergence may further impact the internal exposure to individual metabolites and modulate susceptibility to the parent compound, particularly in populations with variable FMO3 expression or activity. For example, CYP2A6 activity is influenced by genetic polymorphisms and interethnic differences (Oscarson 2001) that are linked with an increased risk of cigarette smoking and earlier smoking initiation (Pérez-Rubio et al. 2017). Likewise, genetic polymorphisms also influence FMO3 activity (Teitelbaum et al. 2018). As 6MN metabolism is highly reliant on FMO3 activity, it raises concerns about the variability in 6MN metabolism across different human populations. Furthermore, FMO3 polymorphisms are associated with the formation of reactive intermediates of certain xenobiotics (Guo et al. 2023; Rendić et al. 2022), potentially affecting their toxicological profiles. Again, these findings suggest that 6MN and NIC may present distinct kinetic and toxicological characteristics, which remains a gap in knowledge.

Just as urinary COT and 3HC are widely used to assess NIC exposures (Benowitz et al. 2010; Thomas et al. 2020), the corresponding urinary 6MN metabolites in humans (and mice) exposed to 6MN are potentially useful biomarkers of exposure. In validation, none of the 6MN metabolites are detected in urine of humans (or mice) exposed only to NIC thus reinforcing their 6MN specificity (Tables 2, 3; Fig. 4). Among these, 6MNNO is consistently the most abundant and reliably detected metabolite across all exposure models and routes (Figs. 2, 3), making it ideal for biomonitoring. Other metabolites, including 6MCOT, 6M3HC, and 6MHPBA, are good candidates because these are also observed in both species albeit at lower levels and with greater inter-individual variability. Nonetheless, these additional metabolites may serve as complementary biomarkers and enhance the accuracy of exposure assessment. For example, 6MCOT represented approximately 8% of the total response, closely matched by COT at 9%. Given its structural similarity to COT, and COT’s established role as a NIC exposure biomarker (Benowitz et al. 2009), 6MCOT also may serve as an analogous indicator for 6MN exposure. Moreover, similar to the established method for NIC (Lorkiewicz et al. 2022), it is feasible to calculate 6MN molar equivalents by summing the concentrations of multiple 6MN-derived metabolites, although this strategy requires validation.

Levels of 6MN metabolites in mouse urine after inhalation of PG:VG+6MN and SPREE BAR (Table 3) revealed important insights into the dose and formulation of the SPREE BAR product. The metabolite ratios of two 6MN sources were much lower than a predicted 1:5 ratio based on expected quantity of 6MN in the e-liquids (our in-house lab-made ≈1% vs SPREE BAR labeled as 5%). This finding was consistent with the report indicating that SPREE BAR erroneously labelled with a higher content of 6MN (Erythropel et al. 2024). It is also possible that additives in the e-liquid of SPREE BAR, such as flavoring agents or pH adjusters, may have altered the formulation (e.g., through salt formation) and affected the stability or absorption of 6MN.

Although this study was designed primarily to identify suitable biomarkers of 6MN exposure, we observed that 6MN caused acute-onset and rapidly resolving neurological symptoms (i.e., impaired motor coordination, seizures, abnormal respiration, and behaviors such as face-pressing and loss of balance) in mice that were not observed following equimolar NIC doses. These findings suggest that 6MN may have increased central nervous system (CNS) potency compared with NIC. One caveat is that we used the freebase 6MN and a NIC salt, which confounds a direct comparison of potency. Nonetheless, it is known that 6MN was synthesized in 1967 (Haglid 1967) and tested for its biological activity in the 1980s (Seeman et al. 1985; Seeman et al. 1981). Previous data including industry documents show that 6MN is either equally or up to 3-times more potent than NIC in multiple bioassays including convulsions (ED₅₀) and lethality (LD₅₀) (Jordt et al. 2024; Rylander 1982). Our data support an enhanced potency of 6MN vs NIC, yet additional studies are warranted to characterize its pharmacodynamics, receptor selectivity, and long-term toxicity effects given the increased presence of 6MN-containing products like SPREE BAR. Altogether, these observations reinforce that 6MN cannot be assumed to carry the same risk profile as NIC and further emphasizes the need for dedicated toxicological assessment of synthetic NIC analogs.

Although this study provides a foundation for future investigations into 6MN metabolism and toxicity, several limitations should be acknowledged. First, the study primarily focused on oxidative urinary metabolites that are structurally analogous to known nicotine metabolites, based on a hypothesis-driven approach to identify candidate biomarkers of exposure. As such, the investigation was not designed to comprehensively profile all potential metabolic pathways, including conjugation (e.g., glucuronidation or sulfation). In addition, untargeted metabolomic comparisons between control and exposed samples were not conducted in this study, which may have limited the discovery of unexpected or minor metabolites. Second, we did not perform in vitro enzyme incubation studies (e.g., using human or mouse liver microsomes or S9 fractions), which are typically used to identify specific enzymes involved in xenobiotic metabolism and to estimate kinetic parameters (Jia and Liu 2007; Richardson et al. 2016). While the dominance of N-oxidation over C-oxidation provides a clear indication of possible enzymatic preference (e.g., FMO3 vs. CYP2A6), definitive enzyme mapping remains a goal for future work. Third, due to the absence of authentic standards, only the relative quantification of 6MN metabolites could be performed. Consequently, the proportions of individual metabolites were estimated based on their LC-MS peak responses rather than absolute concentrations. Despite this limitation, the primary conclusion regarding enzyme preference in 6MN metabolism remains valid, assuming comparable ionization efficiencies between corresponding NIC and 6MN metabolites. Lastly, the small number of human subjects limits the generalizability of our findings. A follow-up study in a larger cohort is underway to further characterize 6MN metabolism and evaluate the reliability of the identified metabolites as exposure biomarkers.

Conclusions

To our knowledge, this is the first integrated study of 6MN metabolism across species with real world human exposures to 6MN. Using our ILGA workflow and a curated structure library, we identified six 6MN metabolites in human urine following real-world SPREE BAR product use. As these metabolites (and 3 others) were present in 6MN-exposed mice, this cross-species validation reinforces the translational relevance of our findings and highlights the utility of 6MNNO as a candidate biomarker for documenting 6MN (Metatine) exposure in humans. Equally novel is that despite high structural similarity of 6MN and NIC, their enzymatic metabolism preference clearly is divergent. The basis for this preference is likely steric hindrance in 6MN, however, it did not depend on route of administration. 6MN biotransformation is dominated by FMO3-mediated N-oxidation with reciprocal reduction in CYP2A6-linked pathways (NIC metabolism preference is vice versa). These distinctions may influence not only the metabolic kinetics and inter-individual variability but also potential toxicity. In addition to 6MNNO, several other metabolites may serve as complementary biomarkers, and integrated metrics such as 6MN molar equivalents warrant further evaluation. Importantly, 6MN caused acute neurotoxic effects in mice at doses where NIC had no observable impact – an idea supported by tobacco industry research documents. Moreover, because SPREE BAR is labeled as “5% Metatine” yet only contains (0.6%), this may increase ‘toxic encounters’ for those who DIY their e-liquids. Similarly, as novel nicotine analogs continue to emerge in consumer tobacco and non-tobacco products, rapid metabolic and toxicological profiling is essential for addressing regulatory and public health policies. Based on our current findings, we plan to establish a quantitative LC-MS/MS method for 6MN biomonitoring in humans, including synthesis of authentic standards and isotopically labeled internal standards for accurate exposure assessment in future studies. Our current study provides a robust framework for this process, combining targeted metabolic analysis, biomarker discovery, and initial safety screening.

AVAILABILITY OF DATA AND MATERIALS

Upon reasonable request, the datasets used and/or analyzed during the current study can be made available via the corresponding author.