Qualitative and Quantitative Analysis of Nicotine, Nicotine Derivatives, and Nicotine-Related Alkaloid Optical Isomers: A Review

The purpose of this review is to provide an in-depth coverage of both current and historical literature pertaining to the qualitative and quantitative analysis of nicotine and secondary alkaloid optical isomers in a variety of matrices. Interest in nicotine, related “secondary alkaloids” (e.g., nor-nicotine, anatabine, anabasine, myosmine) and more specifically their optical isomers, has recently garnered intense interest. Synthetic nicotine optical isomers are being produced in the laboratory and used in e-cigarettes with pure and mixtures of the L(−)- and D(+)-nicotine optical isomers being employed. The United States Food and Drug Administration (FDA) has begun regulating the use of synthetic nicotine, termed tobacco free nicotine (TFN), as it has done previously with tobacco derived nicotine (TDN) products. In a similar vein, the FDA has stated that all combustible tobacco products should have lower tobacco nicotine values than those currently in the United States market. This mandate would most likely require that the tobacco industry develop methodologies to reduce the tobacco nicotine content of current cured tobaccos or begin employing Nicotiana tabacum cultivars containing notably lower amounts of nicotine and its secondary alkaloids. At this time, L/D nicotine ratio data for nicotine and related secondary alkaloids do not sufficiently establish as it currently stands, data surrounding the D/L ratio of nicotine and its secondary alkaloids for the lower nicotine tobaccos, does not sufficiently document a well described ratio across different tobacco cultivars. Hence, robust and relatively rapid analytical methodologies directed at qualitative and quantitative analysis of nicotine enantiomers and secondary alkaloid enantiomers will be essential in providing reliable data.

Nicotine (Pyridine, 3-[(2S)-1-methyl-2-pyrrolidinyl]-) is a naturally occurring alkaloid existing as a tertiary amine with one chiral carbon located at the 2’ position of the N-methylpyrrolidine ring. There are two optically active isomers of nicotine, the naturally-occurring (S)(−) form, and the (R)(+) form. By 1904 (2) both optical isomers had been successfully identified and fully characterized by wet (classical) chemistry methods (1, 2).

Although there are well over 50 species of plants containing nicotine, Nicotiana tabacum (N. tabacum) and N. rustica (3) are the two predominant species containing nicotine, N. tabacum being the predominant species used in tobacco products (3).

Nicotine in N. tabacum exists almost entirely as the L(−) (or (S)(−) form) optical isomer with the D(+) (or (R)(+) form) optical isomer playing only a very small part in the total tobacco-derived alkaloid composition.

In 1999, a treatise on the numerous analytical determinations of nicotine, related compounds, and their metabolites without emphasis on optical isomer analyses was published (4). There are literally hundreds of publications on nicotine and nicotine-related topics that have used classical and modern analytical determinations of nicotine in various matrices (e.g., tobacco, smoke, biological fluids, etc.) and detection methods (electrochemical, polarimetry, flame ionization, etc.) (4). Today, the standard method for the determination of nicotine in tobacco and tobacco products is by gas chromatographic analysis connected to a flame ionization detector (FID) (5). The method is applicable to ground tobacco, cigarette filler, ground cigar filler, and smokeless tobacco products (e.g., snus, moist snuff, dry snuff, and chewing tobacco) and nicotine pouches. This standard method is applicable for the determination of nicotine in traditional and very low nicotine (VLN) content tobacco and tobacco products.

However, and most importantly for this discussion, the current standard method for the determination of nicotine is not applicable for the separation, resolution, and identification of the optical isomers of nicotine, i.e., D(+) ((R)(+)) and L(−) ((S)(−)).

The L(−) isomer of nicotine is generally believed to be the more pharmacologically active of the two nicotine enantiomers (6, 7, 8). LD₅₀ values for a variety of species have been measured (6, 7, 8). The test for LD₅₀ is dependent on the chemical agent, the racemate, the species being tested, the sex of the animal, the route of exposure, the phase (liquid or gas) of the agent along with many other variables can affect the LD₅₀. In mice, for example, the intravenous administration LD₅₀ for L(−)-nicotine has been measured to be 0.33 mg/kg while the LD₅₀ for the D enantiomer has been measured to be 6.15 mg/kg (7).

The chiral separation of nicotine enantiomers and minor alkaloid enantiomers from tobacco sources is, therefore, of interest because the D(+) and L(−) isomers of these compounds have notable differences in their physiological activity (6, 7, 8). This difference is also reflected in the physiological properties of tobacco-specific N’-nitrosamines (TSNAs), in particular that of N’-nitrosonornicotine (NNN) (9, 10). Thus, knowledge of the total nicotine content of any sample does not in and of itself provide a complete assessment of the potential pharmacological and physiological activity of tobacco samples. Such a determination requires that the distribution of the optical isomers within the sample be accurately defined.

In general terms, the literature can be segregated into three broad main categories regarding the qualitative and quantitative analysis of nicotine, secondary alkaloids, and their optical isomers: 1) chiral/gas chromatography with information-rich detectors such as mass selective detection coupled with selected ion monitoring (Chiral/GC/MSD/SIM), 2) chiral/liquid chromatography with information-rich detectors like mass spectrometry (Chiral/LC/MS), and 3) chiral supercritical fluid chromatography (Chiral SFC), with many of the same wide array of detectors used for GC and LC approaches. The structures of nicotine and related tobacco-derived alkaloids are shown in Figure 1 (11).

Figure 1

Evidence of the interest in the optical isomer distribution of nicotine and tobacco-related alkaloids appeared well before the turn of the 21^st century (1, 2, 12, 13, 14, 15, 18, 19, 20, 21, 22). There are numerous reports of successful attempts at enantiomeric separations using both liquid and gas chromatographic techniques. For example, Seeman et al. (12) reported on the room temperature, chiral, high performance liquid chromatography (HPLC) using β-cyclodextrin bonded phases for the examination of the enantiomeric separation of racemic nicotine and 19 racemic nicotine analogues. The impact of mobile phase pH, mobile phase composition, and structural features of the substrates were addressed. A host-guest complexation interaction was advanced as a primary mechanism of separation (12).

Likewise, Armstrong et al. (13) reported on the determination of the enantiomeric composition of nicotine in 18 smokeless tobaccos, 3 strains of tobacco leaf, 8 pharmaceutical products, and 4 commercial reagents smokeless tobacco, medicinal products, and commercial reagents had been made (13). The relative amount of the minor enantiomeric component, (D(+)- or (R)(+)-nicotine, ranged from −0.1% to −1.2% of the total nicotine in all samples. The highest level of (R)(+)-nicotine was found in a commercial transdermal patch. The extraction and purification processes used in obtaining commercial L(−)- or (S)(−)-nicotine supplies from tobacco do not appear to decrease the amount of (R)(+)-nicotine present. The detection of relatively minor amounts of tobacco-derived secondary alkaloids in commercial nicotine reagents with nicotine purities in excess of 99%, led to the conclusion that the presence of relatively minor amounts of secondary alkaloids may point to the origin of the nicotine as arising from natural sources (13).

In work by Perfetti and Swadesh (14) a high-performance poly(styrenedivinylbenzene) reversed-phase column was used to separate the enantiomers of nicotine from other components. On-line polarimetric detection with a commercial laser polarimeter permitted sensitive, reproducible quantitation of the relative enantiomeric purity of nicotine. Detection limits of about 12 μg/mL were established, with the range of linearity extending to about 200 μg/mL. It was possible to assign the relative purity of mixtures of nicotine to about +/− 0.5%. The precision of the on-line polarimeter was comparable to that of a static polarimeter, but the sample requirement was approximately 1000 times less. Optically inactive components were separated, making online polarimetry intrinsically more accurate than static polarimetry, and readily adaptable to the analysis of complex mixtures (14). Somewhat more recently, chiral gas chromatography/mass spectrometry/selected ion monitoring (Chiral GC/MS/SIM) was successfully employed to quantitate the presence of the two nicotine enantiomers in selected samples (15, 16, 17, 18). The approaches employed readily attainable GC and MS conditions and a capillary column containing a β-cyclodextrin optically active stationary phase to separate the nicotine optical isomers. The optical isomer detection limits were typically in the single ng/mL range due in part to the SIM capability of the mass spectrometer. Under the conditions employed, the retention times of the two isomers were approximately 159.0 and 159.5 min. Thus, this GC-based methodology could be directly applicable to the determination of the qualitative and quantitative distribution of nicotine optical isomers, providing capability to both authenticate the presence of a specific nicotine optical isomer as well as the amount of the isomer present in the sample of interest. With a minimal variation of column oven temperature programming and employing SIM for the following secondary alkaloids, retention times of the optical isomers of the tobacco-related minor secondary alkaloids, nornicotine, anabasine, and anatabine were in the range of 29.5, 33 and 38 min, respectively. Table 1 lists the ions associated with the electron impact mass spectrum of selected alkaloids and serves to illustrate the uniqueness of the molecules, more specifically, their even numbered molecular weights and the presence of relatively abundant molecular ions, both of which contribute significantly to the detection and quantitation of these molecules at relatively low concentrations, ng/μL, regardless of the separation protocol employed. These molecular and electron impact mass spectral characteristics attributable to these alkaloids facilitate their being detectable and quantifiable at very low concentrations (15, 16, 17, 18).

Hence, gas chromatography/mass spectrometry (Chiral GC/MS/SIM) has been successfully employed to quantitate the presence of the two nicotine isomers in selected samples (15, 16, 17, 18). The approaches employed readily attainable GC and MS conditions and a capillary column containing a β-cyclodextrin optically active stationary phase to separate the nicotine enantiomers. However, a significant disadvantage of the GC approach is that related to the nicotine enantiomer retention times of greater than 150 min. This fact essentially renders this approach best suited to research efforts rather than situations demanding rapid sample analysis turnaround times.

In addition to the papers discussed above, additional late 1980s and 1990s publications have reported the determination of alkaloid enantiomers in tobacco.

These alkaloid enantiomer analysis approaches have employed normal phase high-performance liquid chromatography (HPLC) coupled with a diode-array or UV detector and mass spectrometry (19, 20, 21, 22). A 1987 publication described the chiral separation of nicotine-related alkaloid enantiomers using a β-cyclodextrin bonded phase with retention times on the order of 15 min (20). An HPLC-based separation of (±) nicotine and (±) nornicotine by using two derivatized cellulose-based chiral stationary phases operated in the normal phase mode. It was found that different chiral substituents linked to the cellulose backbone significantly influenced the chiral selectivity of the derivatized chiral stationary phases. The results showed that, in general, the tris(4-methylbenzoyl) cellulose (Chiralcel OJ) surpassed tris(3,5-dimethylphenyl carbamoyl) cellulose (Chiralcel OD) in its capacity for enantiomer resolution. On the former column, the resolution of (±) nicotine and (±) nornicotine enantiomers depended largely on mobile phase compositions. For the separation of the nicotine enantiomers, the addition of trifluoroacetic acid to a 95:5 hexane/alcohol mobile phase greatly improved the enantiomer resolution. For (±) nornicotine separation, a reduction in the concentration of alcohol in the mobile phase was more effective than the addition of trifluoroacetic acid, and stationary phase interactions were discussed to explain how different additives in the mobile phase and different substituents on the cellulose glucose units of the chiral stationary phases impacted the separation of both pairs of enantiomers.

In the mid-1990s a manuscript appeared describing the use of macrocyclic antibiotics, at the time, a new class of chiral selectors for liquid chromatography (23). With this new technology, nicotine enantiomers were separated employing either a normal or reverse mobile phase in approximately two minutes using solvent systems that were compatible with electrospray ionization mass spectrometry (ESI-MS). The concept of chiral stationary phases, employing antibiotics, for HPLC has been extended to cover a wider range of chiral phases, including teicoplanin, avoparcin, and ristocetin (24, 25, 26), all three demonstrating the capability of enantiomer resolutions. A review of the use of these types of stationary phases appeared in 2001 (27). Continued further developments in these antibiotics technologies will appear in the upcoming discussions, vide infra.

The chiral determination of nornicotine, anatabine, and anabasine in tobacco by achiral gas chromatography with (1S)(−)-camphanic chloride derivatization with an emphasis on application of the method to the enantiomeric profiling of N. tabacum cultivars and curing processes has been published (28). An improved method for simultaneous and high-precision determination of the individual enantiomers of nornicotine, anatabine, and anabasine in four tobacco matrices, based on an achiral gas chromatography-nitrogen phosphorus detector (GC single bond NPD) with commonly available Rtx-200 column using (1S)(−)-camphanic chloride derivatization was described. The method development consisted of the optimization of both extraction and derivatization protocols and the screening of achiral columns. Under the optimized experimental conditions, this method exhibited excellent detection capability for the alkaloid enantiomers, with coefficients of determination (r ²) > 0.9989. The limit of detection (LOD) and limit of quantitation (LOQ) ranged from 0.087 to 0.24 μg/g and 0.29 to 0.81 μg/g, respectively. The recoveries and within-laboratory relative standard deviations (RSDR) were 94.3% – 104.2% and 0.51% – 3.89%, respectively. The tobacco cultivars were found to have had a significant impact on the nornicotine, anatabine, anabasine concentration and enantiomeric fraction (EF) of (R)(+)-nornicotine, whereas the only significant change induced by the curing processes was an increase in the EF of (R)(+)-anabasine.

A fully automated multi-dimensional gas chromatography (MDGC) system with a megabore precolumn and chiral cyclodextrin-based analytical column was developed to analyze the enantiomeric compositions of anatabine, nornicotine, and anabasine in commercial tobacco (29). The multi-dimensional GC system provides for the transfer of a selected portion of a pre-column (megabore) eluant onto a second analytical column having a superior and unique separation capacity. The enantiomer abundances of anatabine and nornicotine were found to vary among different tobaccos. (S)(−)-anatabine, as a proportion of total anatabine, was 86.6% for flue-cured, 86.0% for Burley and 77.5% for Oriental tobacco. (S)(−)-nornicotine, as a proportion of total nornicotine, was 90.8% in Oriental tobacco and higher than in Burley (69.4%) and flue-cured (58.7%) tobacco. (S)(−)-anabasine, as a proportion of total anabasine, was relatively constant for flue-cured (60.1%), Burley (65.1%) and Oriental (61.7%) tobacco. A simple solvent extraction with dichloromethane followed by derivatization with trifluoroacetic anhydride gave relative standard deviations of less than 1.5% for the determination of the (S)(−) isomers of all three alkaloids.

The enantiomeric compositions of nornicotine, anatabine, and anabasine have been measured using chiral gas chromatography/mass spectrometry (Chiral GC/MS) in three types of tobacco leaf (Burley, Turkish, and Virginia); three types of smokeless tobacco (loose-leaf, dry snuff, and moist snuff); and four types of cigarettes (30). Regardless of the tobacco type or product, anabasine always had the highest relative percentage of the minor (R)(+) enantiomeric component (between 40 and 46% vs. 54–60% of the (S)(−) enantiomer). Of the four common tobacco alkaloids, nicotine was found to have had the highest enantiomeric excess (e.e.) while anabasine had the lowest (in the tobacco leaf and tobacco products analyzed). Nornicotine and anatabine had intermediate e.e. values.

Evaluating the source of nicotine in e-liquid samples is arguably a complex challenge. A portion of the data necessary to help answer this question is the determination of the nicotine optical isomer distribution in samples of this type. Tobacco-derived nicotine contains predominantly (S)(−)-nicotine, whereas, possibly potentially tobacco-free nicotine products such as e-liquids may not. In an effort to address these issues, a normal phase high-performance liquid chromatography method to determinate the enantiomeric composition of nicotine in 10 kinds of flue-cured tobacco, 3 kinds of Burley, 1 kind of cigar tobacco, 2 kinds of Oriental tobacco, 5 kinds of Virginia cigarette, 5 kinds of blend cigarette, 10 kinds of e-liquid, and 4 kinds of smokeless tobacco (31) has been developed. Across the range of samples, the amount of (R)(+)-nicotine ranged from ~0.02% to ~0.76% of total nicotine. An e-liquid sample was found to contain the highest level of (R)(+)-nicotine. Thus, these recent LC-based publications (31 and references found within) directed at the separation of nicotine enantiomers would seem to be poised to provide meaningful, accurate, and precise data assisting in the determination of nicotine sources.

The tobacco-specific nitrosamines (TSNAs), N-nitrosonornicotine (NNN) and N-nitrosoanatabine (NAT), are found in unburned, cured tobacco. In 2000, no data were available on the enantiomeric composition of these nitrosamines in commercial unburned tobacco. Following sample fraction collection from LC-based separations, a chiral stationary phase GC approach with nitrosamine-selective detection (e.g., Thermo Energy Analyser (TEA)) and selected ion monitoring mass spectrometry (SIM-MS) to determine the enantiomeric composition of NNN and NAT in moist snuff, chewing tobacco, and cigarette tobacco was developed. Optimized GC retention times ranged between 33.8 and 45.6 min for baseline resolution of the four pairs of enantiomers. (S)(−)-NNN was found to comprise 75.0 ± 8.83% (SD) (n = 12) of total NNN while (S)(−)-NAT comprised 82.6 ± 1.44% (n = 12) of total NAT (32). Levels of the (S)(−) enantiomers of NNN and NAT were generally similar to those of the corresponding secondary amines, nornicotine and anatabine, suggesting a possible precursor to product relationship. Thus, a combination of the strengths of LC and chiral GC-based technologies would seem to be a viable approach to the detection and quantitation of TSNA enantiomers.

N-nitrosonornicotine, a carcinogen present in tobacco products is optically active and thus has two enantiomers with the (S)(−)NNN exhibiting higher tumorigenic potency than (R)(+)-NNN. An early report described the levels of (S)(−)-NNN in various tobacco products currently marketed in the United States (U.S.) at the time (32). In 2013, Stepanov et al. (9) conducted a more inclusive study. A chiral gas chromatography-based approach was employed to determine (S)(−)-NNN levels in 37 tobacco products at the time currently marketed in the U.S. These included conventional smokeless tobacco, novel smokeless tobacco products, and cigarette tobacco filler. Among all products analyzed, the (S)(−)-NNN enantiomer averaged 62.9 ± 6.3% (SD) of the total NNN. The absolute amount of (S)(−)-NNN in conventional moist snuff averaged 1.26 ± 0.5 μg/g tobacco; in novel smokeless products 0.70 ± 0.2 μg/g tobacco; and in cigarette filler 1.36 ± 0.6 μg/g tobacco. For each cigarette brand, the enantiomeric composition of NNN in cigarette smoke was similar to that of the corresponding tobacco filler. The results clearly demonstrated that (S)(−)-NNN was the predominant NNN enantiomer in moist snuff, novel smokeless tobacco products, and cigarettes marketed in the U.S. at that time.

A new method for the analysis of nornicotine, anabasine, and anatabine has been developed, based on an original derivatization employing isobutyl chloroformate derivatives and a relatively simple chiral gas chromatography/mass spectrometry (Chiral GC/MS) analysis employing a Rtβ-Dex chiral GC column (33). The method allows separate quantitation of (S)(−)-nornicotine and (R)(+)-nor-nicotine, and the analysis of anabasine and anatabine (without isomer separation). It was found that the proportion of (S)(−)-nornicotine in the total nornicotine present in tobacco varies, depending on the tobacco type, between 52.6% for a flue-cured tobacco to 91.4% for a Burley. Green tobaccos (freeze dried) showed lower levels of minor alkaloids and (S)(+)-nornicotine accounted for between 31.6% to 43.8% of the total nornicotine. It is understood today that enantioselective demethylation of nicotine is the reason for the high level of (R)(+)-nornicotine in tobacco leaf (59). Studies by CAI et al. (59) suggest that an enantio-selective mechanism facilitates the maintenance of a reduced (R)-nicotine pool and, depending on the relative abundances of the three nicotine demethylase enzymes (CYP82E4, CYP82E5v2, and CYP82E10), can confer a high (R)-enantiomer percentage within the nornicotine fraction of the leaf. Under optimized conditions, the retention times for the optical isomers of nornicotine were 46.63 and 46.96 min. Thus, both GC and LC approaches now join together in representing powerful protocols for the detection and quantitation of nicotine and related tobacco alkaloids.

The analysis of (S)(−)- and (R)(+)-nicotine enantiomers in various nicotine samples and in e-liquids, has been reported (34). Both chiral GC and chiral LC-based methods were employed producing comparable results. The GC approach involved chiral GC/MSD employing two columns, RTGammaDEXsa linked in series. The LC approach involved UV detection using a thermostated column compartment and a Chiracel OJ-3 chiral LC column. All samples, with the exception of a synthetic nicotine sample, were found to predominately (> 99%) have the S(−) optical isomer. The dominance of the S(−) enantiomer led the authors to conclude that most likely the source of nicotine for the examined samples was natural. Contained within this report were retention times for the nicotine enantiomers for both the GC and LC approaches. For the GC approach, the S(−) and R(+) retention times were 70.56 and 71.40 min, respectively. For the LC approach the S(−) and R(+) retention times were noticeably and significantly lower at 5.87 and 8.78 min, respectively. From a sample turn around perspective, the shorter nicotine enantiomer separation time for the LC approach is very attractive when compared with the much longer GC retention times. The relatively long nicotine enantiomers retention times when compared with the retention times attainable with LC approaches will be shown to be a consistent theme for most all of the reported GC-based studies, vide supra, vide infra.

Method development and validation for the determination of nicotine enantiomers in electronic cigarette liquids using reversed-phase and chiral high performance liquid chromatography (HPLC) has been described (35). Separation was achieved using a Daicel Chiralpak AD-H column. In selected cases discrepancies between product quoted nicotine levels and measured levels were noted.

A recent study compared the efficacy of chiral GC and chiral HPLC for the analysis of nicotine (36). To develop a suitable dispersive liquid-liquid microextraction (DLLME) method, the following parameters were optimized: pH, extraction solvent, dispersive solvent, type and quantity of salt, and laboratory temperature. The validation of the method was carried out by employing an optimized HPLC method.

The LODs were 0.11 μg/mL and 0.17 μg/mL for the (S)(−) and (R)(+) enantiomers, respectively. The LOQs were 0.30 μg/mL and 0.44 μg/mL, respectively. The optimal calibration range was between 0.30–18 μg/mL and 0.44–4.40 μg/mL, respectively, and the correlation coefficient (r ²) was 0.9978–0.9996. The intra-day accuracy was 79.9– 110.6%, and the intra-day precision was 1.3–12.0%. The inter-day accuracy was 87.8–108.0%, and the inter-day precision was 4.0–12.8%. E-liquid and, very importantly, biological fluids (urine and saliva) were analyzed for enantiomers using the optimized method. The chiral GC enantiomer retention times were ~148.5 and 149 min, while the LC retention times were ~5 and ~6 min.

While all of the publications cited above have their strengths, there are a few aspects that should be viewed as less than desirable from the perspectives of analytical protocol simplicity and sample analysis time. In a number of cases, chemical derivatization has been necessary to render the nicotine and related minor secondary alkaloid enantiomers compatible with GC analysis conditions. For a number of GC separations, the retention times required for enantiomer resolution were close to or greater than to 1 h, which is inconsistent with typical high throughput sample analysis times. While most all of the LC reports site very short, resolved enantiomer retention times, often less than 10 min, several of them employ mobile phase components that have some degree of exposure toxicity of their own.

A relatively recent publication detailing the resolution of nicotine and minor secondary alkaloid enantiomers has addressed a number of the previously noted undesirable analytical protocol attributes (11). Specifically, a reverse phase ultra-performance liquid chromatography/mass spectroscopy/mass spectroscopy (UPLC/MS/MS) method for determination of alkaloid enantiomers in tobacco matrices has been developed. The reported advantages of this overall protocol included relatively simple sample (tobacco cultivars and products) preparation with the optimized extraction of the alkaloid enantiomers using methanol and water, performing the UPLC analysis with reasonably acceptable mobile phases, and a notably low limit of detection for the alkaloid enantiomers due in part to the MS/MS approach, multiple reaction monitoring (MRM). Through an optimization of chiral column types, mobile phase compositions (ammonium formate, ammonium hydroxide, methanol, and acetonitrile), mobile phase pH, gradient elution profiles, and MRM sequences, very effective resolution and detection of nicotine, nor-nicotine, anabasine, and anatabine enantiomers were attained, with relatively short overall retention times of less than 17 min. The LOD and LOQ ranges for the enantiomers were found to be between 1–8 and 4–28 ng/mL, respectively. As a function of tobacco type this study revealed that the % R isomer for nicotine, nornicotine, anatabine, and anabasine varied between 0.07–0.4%, 15–49%, 14–16%, and 37–42%, respectively.

Nornicotine, a well-documented minor alkaloid constituent of tobacco, is a known precursor to the carcinogen N-nitrosonornicotine, NNN (9, 10). NNN is produced during the curing and processing of tobacco (37). Recent evidence has revealed that nornicotine enantiomers have different pharmacological properties. In a HPLC-based study, an accurate and relatively rapid method was developed for the enantioseparation of (R)(+)-nornicotine and (S)(−)-nornicotine enantiomers in tobacco by ultra-performance convergence liquid chromatography (UPLC) with tandem mass spectrometry (MS/MS). Chromatographic conditions were optimized to achieve the optimal resolution of the two nornicotine enantiomers. Results indicated that (R)(+)-nornicotine and (S)(−)-nornicotine could be separated within 5 min when ammonium hydroxide was added into the co-solvent, and a best resolution (R_s = 4.76) was achieved on a immobilized cellulose tris-(3,5-dichlorophenylcarbamate) chiral stationary phase. The method was applied to analyze the compositions of (R)(+)-nornicotine and (S)(−)-nornicotine in three typical types of tobaccos (flue-cured, Burley, and Oriental). It was found that the enantiomer fraction of nornicotine (the proportion of (S)(−)-nornicotine in the nornicotine pool) in Burley tobacco samples was relatively high and constant compared with flue-cured and Oriental tobaccos.

Employing a chiral HPLC column and an optimized mixed solution mobile phase of hexane/isopropanol (95/5: vv) and diethylamine as the mobile phase additive in the normal phase chromatography mode, coupled with detection via a diode array detector (HPLC-DAD) a method was developed for the separation and quantitative analysis of nicotine optical isomers in tobacco and tobacco products (38). The characterization and test results of this method revealed the detection limit (LOD) and limit of quantification (LOQ) were 0.2 mg/g and 0.05 mg/g, respectively, with accompanying precision of intra- and inter-day measurements of nicotine of less than 0.85% and 4.00%, respectively. The recovery rates of S(−)-nicotine and R(+)- nicotine were close to 100%. Using this protocol, the distribution of nicotine optical isomers of 26 different types of tobacco and tobacco products was determined and revealed that nicotine enantiomer from tobacco was mainly S(−)-nicotine, and 6 samples were detected having relatively low amounts of R(+)-nicotine. The proportion of R(+)- nicotine in total cigarette smoke was higher than that of the precursor cigarette tobacco. Taken collectively, the methods discussed above strongly suggest that isolated TDN and its secondary alkaloids are overwhelmingly dominated by the S(−) enantiomer.

A wide variety of new TFN products have been commercialized as e-liquids. The TFN designation means that the nicotine contained within the TFN products was not derived from tobacco. Tobacco-derived nicotine contains predominantly (S)(−)-nicotine, whereas TFN products may or may not, vide supra. A variety of commercial tobacco and TFN products have been analyzed to identify the presence and composition of each nicotine enantiomer (39). A rapid and effective enantiomeric separation of the nicotine enantiomers was developed employing a modified macrocyclic glycopeptide (chiral phase) bonded to superficially porous particles (SPP) with accompanying detection using diode array, circular dichroism systems, and electrospray ionization mass spectrometry (ESI-MS). The nicotine enantiomeric assay was rapid and could be completed in ~2 min with high resolution and accuracy using chiral HPLC with electrospray ionization mass spectrometry, the retention times of the S(−) and R(+) enantiomers being 1.5 and 2.0 min, respectively. All of the TFN products were found to have a 50/50 ratio of nicotine enantiomers, while the TDN (tobacco derived nicotine) products were observed to have in excess of 99% of the S(−) enantiomer. Using the same analytical hardware and HPLC protocols, separation of the nornicotine enantiomers was attained with retention times for the S(−) and R(+) enantiomers being ~4.2 and 4.8 min, respectively. For the nicotine enantiomers the %RSD was excellent at +/− 0.2%. The presence of a 50/50 mixture of nicotine enantiomers in selected samples would possibly implicate a synthetic source of nicotine, TFN.

Employing very similar hardware as described above (39) with HPLC columns prepared with superficially porous particles (SPP, 2.7 μm) bonded with modified chiral stationary phases, effective resolution of the nicotine enantiomers was attained in less than 20 s (40). The retention times for the well resolved S(−)- and R(+)-nicotine enantiomers were ~11 and ~16 s, respectively, under optimized conditions. Through the optimization of mobile phase protocols and column selection, the effective resolution of enantiomers associated with ~40 nicotine analogues was attained. With a wide range of functionality represented across this wide range of nicotine analogues, retention times never exceeded 15 min for the vast majority of the compounds investigated. Of note, diastereomeric interconversion during the chromatographic separation process of selected tobacco-derived nitrosamines was discovered even employing optimized separation conditions. Varying the column temperature was found to influence the degree of the diastereomeric interconversion. Molecular chirality has been confirmed as a major thrust in the pharmaceutical industry; therefore, there is a continuous demand to extend the available analytical methods for enantiomeric separations and enhance their efficiency, due in part to the well-established observation that in an overwhelmingly large number of cases enantiomers display significantly different pharmacological responses (41). Most recently, liquid chromatography methods based on the application of chiral stationary phases have become a very sophisticated field of chiral HPLC enantiomeric separations through the use of unique chiral bonded stationary phases (42 and references therein). Hundreds of such chiral stationary phases for chiral liquid chromatographic separation have been commercialized over the past few years. Among these are complex bonded macrocyclic glycopeptide-based chiral supports that have proved to be an exceptionally useful class of chiral molecules for the separation of enantiomers. Detailed information on glycopeptide-based chiral separation applications is discussed in the review, including both the glycopeptide structure, the column size, the support particle size, the mobile phase composition, and gradient profiles. More specifically, the macrocyclic glycopeptides, teicoplanin, modified teicoplanin, and vancomycin bonded to sub-2 μm superficially porous particles and fully porous particles have been effectively employed in chiral supercritical fluid chromatographic (SFC) and HPLC enantiomeric separations of nicotine, nicotine metabolites, other secondary minor tobacco alkaloids, and tobacco specific nitrosamines with relatively rapid retentions times and excellent enantiomer resolution.

A multicolumn ultra high-performance liquid chromatography (UHPLC) screening workflow capable of combining 14 columns (packed with sub-2 μm fully porous and sub-3 μm superficially porous particles (SPP) bonded to chiral stationary phase materials) with nine mobile phase eluent choices has been described (43). The technology is based in part on the influence on retention times produced with ultrafine particles covalently bonded to chiral stationary phases (42). The automated system operates with vast selection of reversed-phase liquid chromatography, hydrophilic interaction liquid chromatography, polar-organic mode, and polar-ionic mode conditions with minimal manual intervention. Computer-assisted modeling of all aspects of the separation process for the highly efficient enantioseparations is described. The computer-derived retention models were found to be very accurate for chiral resolution of single and multicomponent mixtures of enantiomeric species across different types of chiral stationary phases, with differences between experimental and simulated retention times of less than 0.5%. Illustrations were provided clearly establishing a robust foundation for very effective UHPLC-based enantioseparations with retention times routinely described in seconds. Thus, ultrafine particles (< 2 μm) bonded to chiral stationary phases has a well-documented positive impact on the separation of alkaloid enantiomers.

A review article discussing results of enantiomer resolution techniques for a variety of plants including N. tabacum (44) has been published. References to and applications of techniques for chiral alkaloid analysis, HPLC, capillary zone electrophoresis (CZE), sample preparation for chiral alkaloids analysis, ultrasound assisted extraction (UAE), microwave assisted extraction (MAE), alkaloid purification approaches, liquid-liquid extraction (LLE), solid phase extraction (SPE), enantioselective chromatographic methods, and enantioselective electrophoretic methods were discussed in detail.

The robustness of the LC approach for the separation and quantitation of nicotine enantiomers has been clearly documented such that a relatively large number of commercially available chiral LC columns have appeared in the marketplace (45) with illustrated enantiomeric separations. Enantiomer retention times of ~14.5 and 16 min can be easily attained with resolution factors around 1.52, with accompanying detection using relatively inexpensive UV-VIS systems.

Applications to the analysis of additional biologically-derived samples employing chiral LC techniques have recently appeared (46, 47). A high-performance liquid chromatography-tandem mass spectrometry (HPLC-MS/MS) method for simultaneous quantification of nicotine and cotinine enantiomers in rat plasma was developed and applied in a stereoselective pharmacokinetic study. The analytes in the rat plasma were extracted by one-step protein precipitation with methanol, then separated on a Chiralpak IG-3 column (250 mm × 4.6 mm × 3 μm) using a mobile phase of 0.2% ammonium formate in methanol at a flow rate of 1.2 mL/min, and subsequently quantified with an isotope internal standard. The enantiomer detection was performed on a triple quadrupole tandem mass spectrometer by multiple reaction monitoring mode (MRM) at positive electrospray ionization interface. The results showed that nicotine and cotinine enantiomers were baseline separated with the limit of quantitation (LOQ) of 0.5 μg/L. Retention times for the well resolved nicotine and cotinine enantiomers under optimized conditions were all less than 15 min. In yet another biological matrix, a chiral stationary phase liquid chromatography-nanoelectrospray ionization-high resolution tandem mass spectrometry (LC–NSI–HRMS/MS) method to investigate the enantiomeric composition of low parts per trillion amounts of the carcinogen N’-nitrosonornicotine (NNN) in the urine of cigarette smokers and smokeless tobacco users has recently appeared (47). (S)(−)-NNN is the major enantiomer in tobacco and has been shown to be more carcinogenic than (R)(+)-NNN in rats (10). The results demonstrated that the more highly carcinogenic (S)(−)-NNN was the major enantiomer in human urine, and that the enantiomeric composition of NNN in human urine was similar to that found in cigarette smoke and smokeless tobacco. This report was the first such study to combine chiral stationary phase LC-based separations with nanoelectrospray ionization and high-resolution tandem mass spectrometry to quantify trace levels of enantiomeric metabolites in human urine.

While all of the above references and discussions related to the chiral separation of nicotine and nicotine analogue enantiomers have focused on methodologies dominated by gas chromatography and liquid chromatography, the application of supercritical fluid chiral chromatography (Chiral SFC) can be expected to have applications to these enantiomers as well (48). For example, it is shown that in recently published manuscripts on SFC the range of successfully assessed compounds was very wide, including compounds of variable polarity (lipids, flavonoids, saponins, terpenes) or acidity/alkalinity (phenolic acids, alkaloids), and importantly chiral separations required for a meaningful characterization of, e.g., cannabinoids (49). The versatility in SFC target analytes is mostly possible because of the innate flexibility of SFC, being compatible with different stationary phase chemistries (e.g., silica, reversed phase or chiral materials, hydrophilic interaction chromatography (HILIC)) and multiple detection modes (e.g., DAD (Diode Array Detector), MS, ELSD (Evaporative Light Scattering Detector), ECD (Electrochemical Detector)). For example, the enantioseparation of nornicotine in tobacco by SFC-MS has been accomplished. In order to determine this precursor of the carcinogenic N-nitrosonornicotine (NNN) the authors first screened different chiral stationary phases (Trefoil CEL1 was advantageous over Trefoil CEL2, Trefoil AMY1 and Chiralpak IC-3) and searched for the optimum cosolvent, 0.05% NH₃ in MeOH. When applied in gradient mode, the separation of (S)(−)-nornicotine and (R)(+)-nornicotine was possible in 5 min with excellent resolution (Rs = 4.76). In a comparison of the separation of nicotine enantiomers, the optimized retention times for chiral SFC were 1.8 and 2.5 min, while for an ultrafast chiral LC approach the retention times were 11.8 and 13.5 s (50).

Chiral SFC has been applied to the separation, isolation and purification of various pharmaceutical enantiomers (51). A detailed discussion of chiral SFC and HPLC was published in terms of productivity, stacked injections, sample solubility, particle size, green incentive, organic solvent, recycle solvents, and economical aspects. In general, while separations from an analytical perspective compared favorably between the two methodologies, in selected cases, SFC has been found to be a better technique than HPLC for chiral separations on a preparative scale.

In a recent review, chiral supercritical fluid chromatography was found to be highly efficient for enantioseparations, analytical-scale separations were fast and rapidly developed, and preparative-scale methods provided high productivity. SFC separations were demonstrated to have both economical and ecological advantages (52). Two dimensional (2D) SFC with achiral and chiral column configurations was reported applicable to enantiomeric separations, as well. The use of superficially porous particles (SPP) and fully porous particles (FPP) with bonded chiral stationary phases in SFC columns was shown to provide for the separation of enantiomers on the order of seconds.

To illustrate the maturity of chiral SFC, a review published in 2016 describes the details surrounding the technology related to the preparative separation of enantiomers employing preparative scale chiral SFC (53). At the time preparative chiral SFC was becoming the choice in the pharmaceutical industry for the separation of optically active drug compounds into their respective enantiomers. Thus, examples of the SFC chiral separations on a preparative scale have been clearly demonstrated for a wide array of racemates.

Furthermore, the online coupling of chiral SFC with a precursor step of supercritical fluid extraction (SFE), allowing simultaneous extraction and analysis of a sample, could be a relatively new technical option that may possibly have application to the extraction and analysis of enantiomers from a variety of natural materials and other sample matrices.

Today there exist a number of analytical challenges for the tobacco industry and nicotine testing communities (54, 55, 56, 57, 58). A GC-based method has been validated by CORESTA and subsequently re-validated by ISO. This GC method is for the analysis of nicotine in pure and formulated nicotine products. Other CORESTA-approved options for the detection of nicotine in this limited class of products have demonstrated advantages. Importantly, however, there are no approved methods (54, 55, 56, 57, 58) specifically for the analysis of TFN or synthetic nicotine.

In contrast to TDN which is overwhelmingly dominated by the S(−)-nicotine isomer (> 99%), synthetic nicotine (or TFN) does not come from the tobacco plant. As a result, it contains no other tobacco by-products or tobacco alkaloids. It is synthesized using either a chemical or an enzymatic process. The synthesis is then often followed by a chemical or enantiomeric purification. There are multiple forms of TFN nicotine. Synthetic nicotine can be identical to tobacco derived nicotine, i.e., > 99% (S)-nicotine (CAS number 54-11-5) or can be a racemic mix of (R)- and (S)-nicotine (50/50 with the CAS number 22083-74-5) or vary in the ratios of (R)- and (S)-nicotine; all of which can have different CAS numbers.

The FDA has only recently placed regulations on TFN. Furthermore, recent synthetic advances have clearly demonstrated the capability to synthesize pure nicotine enantiomers. Herein could thus lie the analytical challenge: How to quantitatively document the source of nicotine in a TFN versus a TDN environment? It would seem from the references provided in this review that recent advances in relatively rapid chiral chromatographic approaches, particularly, chiral HPLC and chiral SFC offer excellent approaches to address such issues.