Classification of flavors in cigarillos and little cigars and their variable cellular and acellular oxidative and cytotoxic responses

Flavored tobacco products are increasing in popularity but remain unregulated, with the exception of the ban on flavored conventional cigarettes. Lack of regulation of cigarillos and little cigars allows vendors to have their own version of popular flavors, each with different chemical components. A new classification system was created for flavored cigars in order to easily communicate our results with the scientific community. To understand the physicochemical characteristics of flavored little cigars and cigarillo smoke, size distribution and concentration of particulate matter in smoke were determined. Acellular reactive oxygen species generation was measured as an indirect measurement of the potential to cause oxidative stress in cells. In addition, cigarillo smoke extract treatment on bronchial epithelial (Beas-2b) cells were assessed to determine the flavor-induced cellular toxicity. Flavored cigars/cigarillos showed significant variability in the tested parameters between flavors as well as brands of the same flavor, but most of the cigars showed higher potential of generating oxidative stress, than research grade cigarettes. Flavored cigars produced maximum particle concentrations at 1.0μm and 4.0 μm compared with 3R4F reference research cigarettes. A differential cytotoxic response was observed with cigarillo smoke extract treatments, with “fruits/candy” and “drinks” being the most toxic, but were not more cytotoxic than smoke from cigarettes. These cigarillos with flavors must be well characterized for toxicity in order to prevent adverse effects caused by exposure to flavor chemicals. Our study provides insight into understanding the potential health effects of flavor-infused cigars/cigarillos and the need for the regulation of those flavoring chemicals in these products. Future research is directed to determine the flavoring toxicity of little cigars and cigarillos in vivo studies.


Introduction
In the current environment where the focus is on the dangers of cigarettes, cigars are often perceived to be safer than cigarettes due to the lack of public attention and scientific knowledge

Ethics statements
All experiments performed in this study were approved and in accordance with the University of Rochester Institutional Biosafety Committee. And, all protocol, procedure and data analysis in this study were followed the NIH guildlines and standards of reproducibility and scientific rigor by an unbiased approach. Animal or Human study protocol: None Institutional biosafety approvals: Yes

Scientific rigor statement
The approach to creating the experiment was unbiased and analysis done on the results ensured that our data are reproducible.

Cigar procurment and categorization
Little cigars and cigarillos were purchased in Rochester, NY, at various locations and vendors, to be tested in this study (see S1 Table). The types of cigars used in the experiment are classified into six basic categories based on names and descriptions on the packaging. Flavor names are generated by the manufacturer and a visual representation of the flavors can be found in Fig 1. Categories used are tobacco, menthol, fruit, candy, drinks, and spices. These categories were further sub-divided as needed which included, Black and Mild, berry, cherry, grape, mango, peach, pineapple, tropical, and alcohol. Some variation in which flavored cigar was used is present between each expirmental test due to the limitations in inventory.

Cell-free ROS assay
A fluorogenic dye was created using 0.01N NaOH, 2'7' dichlorofluorescein diacetate (H 2 DCF-DA,) (EMD Biosciences, CA) (Cat # 287810), PO 4 buffer made of sodium phosphate monobasic (JT Baker, NJ) (Cat #3828-01) and sodium phosphate dibasic (Sigma-Aldrich, MO) (Cat #2-0751), and horseradish peroxide (HRP) (Thermo Fisher, Ma) (Cat # 31491). ROS were detected based on flouresent intensity at 495/529nm. Standards that ranged from 0 to 50 μM were created using a 1mM hydrogen peroxide stock that was reacted with the fluorogenic dye at 37˚C for 15 minutes. Standards were measured on a spectrofluorometer (Turner Quantech fluorometer, Mo. FM109535) in fluorescence intensity units (FIU). Samples of cigar smoke extract were also given 15 minutes at 37˚C to react with the fluorogenic dye, then measured immediately. Sample readings were based on the hydrogen peroxide standard curve and denoted as "hydrogen peroxide H 2 O 2 equivalents". Three different methods were used for creating cigarillo smoke extract using an impinger mechanism. Multiple methods were tested to fully understand the cigar burning process. Method 1 used an SKC lab pump (model 224-PCXR8, Eighty-Four, PA, USA) with an average flow rate of 1 L/min. Little cigars or cigarillos were attached to a 50 mL conical tube containing 10 mL of freshly made fluorogenic dye. Cigarillos were lit and smoke was bubbled through the dye for 1 minute. A new cigarillo was used for each sample. Immediately following the bubbling of the dye, it was filtered through a sterile 0.45 μm polyethersulfone syringe filter to eliminate any debris. Method 2 used a standard lab vacuum to bubble the cigarillo smoke. A cigarillo was attached to a 50 mL conical tube containing 10 mL of freshly made fluorogenic dye and smoke was bubbled for 10 seconds. This process was repeated two to three times on the same cigarillo to create separate samples of extract. Immediately following the bubbling of the dye, it was filtered through a sterile 0.45 μm polyethersulfone syringe filter to eliminate any debris. Method 3 used a standard lab vacuum to bubble the cigarillo smoke. Each cigarillo was sectioned into three equal pieces by weight and attached to a 50 mL conical tube containing 10 mL of freshly made fluorogenic dye. Cigar smoke was bubbled through the dye at a constant rate as described above. Each portion of the split cigarillo was bubbled until completely burned and measured for ROS in H 2 O 2 equivalents.
Beas-2b cells were seeded in six-well plates at a density of 400,000 cells per well. Cells were grown until 85-90% confluency and then serum-deprived for 8 hours. Cells were treated with 0.5% and 1.0% concentrations of CSE. The smoke extract was created by using method #1; however, in place of fluorogenic dye being bubbled, 10 mL of 1x PBS was used. The extract was then measured on a Beckman Coulter spectrophotometer (Model DU520). An absorbance value of 1.00 ± .05 was considered to be 10% concentration, and further dilutions were done to obtain that concentration. Cells were treated for twenty-four hours. Untreated cells remained in DMEM supplemented with 1% FBS.
The viability of the Beas-2b cells was measured using acridine orange (AO) and propidium iodide (PI) staining. 20 μL of AO/PI staining and 20 μL of live cells were combined, and then 20 μL of the mixture was inserted into the Nexcelom's Cellometer (Model Auto 2000) and analyzed. The analysis included the number and concentration of live, dead, and total cells and the percent viability of the sample.

Particulate matter collection and its concentration distribution in cigar smoke
To characterize the particle size distribution in cigar smoke, TSI TM 's Dust Trak II (model 8530) was used with particle diameter cut-offs at 1 μm, 2.5 μm, 4 μm, and 10 μm at a 2 L/min sampling rate. Cigars were manually lit and puffed at 1.7 L/min for 1 minute using Scireq's Inexpose system, with 2.5s puff duration, and 16.6s inter-puff interval. Cigar smoke particle sizes were measured in an Enzyscreen chamber (Cat. CR1601) with dimensions 22 cm x 14.5 cm x 16 cm and in a DSI TM chamber with dimensions 44.8 cm x 30.1 cm x 29.6 cm [20].

Statistical analysis
Statistical analysis of significance was calculated using one-way ANOVA as well as Tukey's post-hoc test for multiple comparisons by GraphPad Prism Software version 8.1.1. The results are shown as mean ± SEM of average n = 3 to 4. Data were considered to be statistically significant for P values <0.05.

Categorization of cigars by flavor
To understand how the flavorings in cigars can be grouped, a new classification system was created (Fig 1). This design was used to allow for categories and subtypes. Presently, there are no classifications that convey the flavors in a meaningful way. This figure attempts to do that in a manner which is easy to understand and follow for toxicological studies. There are six main flavor categories, drinks, fruit, tobacco, menthol, candy, and spices with sub-categories, alcohol, black and mild, berry, tropical, pineapple, peach, grape, watermelon, mango, and cherry. Fruit category contained the most flavors as well as the largest number of subtypes.

Acellular ROS production by little cigars and cigarillos
Little cigars and cigarillos produced differential H 2 O 2 equivalents (Figs 2-6). For method #1, many of the flavored cigarillos showed a significant increase in acelluar ROS compared to air with the exeption of Black and Mild's tobacco, Swisher Sweets' blueberry, cherry from both Pt. Rillos and Jackpot, Swisher Sweets' boozy mango, tropical fusion and caramel peach and finally Black and Mild's wine. The highest aceulluar ROS produced was from Dutch Masters' Mint Fusion, Game's white grape and berry blast, White Owl grape, and Wine by Dutch Masters. Little cigars tested from Djarum, black and special, were both significantly higher in ROS than air and produced more ROS than a 3R4F. While each category has at least one low ROS producing cigar many of these cigars tested have higher ROS than a 3R4F (Fig 2).
Method # 2, similar to method #1 showed many of the cigars and cigarillos tested having signicantly higher acelluar ROS production than air (Fig 3). Tested cigars were grouped based on their categories and depicted in Fig 3. Highest significance was seen in the categories Black and Mild, Tobacco, and Spice with both Djarum cigars being the most significant (Fig 3).
Lowest ROS production was seen in categories Candy, Grape, Mango and Peach. Individual cigars and cigarillos tested for ROS productions which show a differential response (Figs 4 and 5). Also, we are interested in how different segments of cigars would affect the ROS production. There was no significant difference among all the segment (Fig 6) based on method #3 of ROS assay, which means cigar burned in a universal way, and oxidative stress generated continously and universally.
Most notably, between methods #1 and #2, it is easy to see that each method gives significantly different results even when testing the same cigarillo. Within method #1, many of the cigarillos burned down faster than others from the same brand and flavor. This resulted in a wide variations of ROS values for a single cigarillo. For example, PT Rillo's wine flavor was tested four separate times using four separate cigarillos. The range of these tests was 49.6-10,788.80 H 2 O 2 equivalents.

Fine particle emission and distribution by flavored little cigars and cigarillos
For all particle size measurements, research cigarettes (3R4F/1R6F) released lower concentration of particles, i.e., particulate matter (PM). The lowest PM from cigars/cigarillos tested was    Oxidative and cytotoxic responses of flavored cigarillos black by Djarum, and the highest was honey fusion by Dutch Masters, special by Djarum, and wine by Black and Mild ( Table 1). Most of the cigar samples tested had the highes PM concentration in categories 1 μm and 2.5 μm ( Table 1). Fig 7A shows the average PM concentration for each of the seven categories and two reference cigarettes only to be used as a visual representation of Table 1

Exposure to cigar flavorings induces variable cellular toxicity
To determine the cytotoxicity of flavored cigars, Beas-2b cells were treated 0.5% and 1% cigar/ cigarillo extract. Overall, each cigar showed a decrease in cell viability from 0.5% treatment to 1% treatment except for Djarum's black cigar (Fig 8). Surprisingly, djarum cigars, tobacco flavor and menthol flavor do not show any significant decreasing in cell viability, while all other fruit, candy, and drink flavors showed reduction of viability. Most significant cytotoxicity can be seen in PT Rillos' wine flavored cigarillo that had viability reduced to 43.9% with 0.5% extract treatment and with a 1% treatment, viability reduced further to 33.6%. At 1% extract treatments Swisher Sweet's caramel peach had a viability reduction to 36.97%, PT Rillos' Strawberry was reduced to 27.2%, and GAME's pineapple reduced to 49.2%. Cell treatment with a 1% extract of a 3R4F cigarette showed a significant reduction in cell viability as well, with an average of 27.7%. Almost all cigars tested had a reduced in cell viability compared with the control, but were not more cytotoxic than a research cigarette (Fig 8).

Discussion
This study intended to determine the toxicity of flavored cigarillos and little cigars. Selected little cigars/cigarillos from various flavor categories were combusted and acellular ROS, particle concentration/distribution, and the cell viability were assessed. Classification of flavored cigars/cigarillos should provide a convenient nomenclature/vocabulary in the scientific community as there was no scientific classification of cigar flavorings until the present. The classification system created, allows for the categorization of flavors in a way that groups like chemicals together, in an attempt to single out which flavors can be the most toxic. Otherwise, introduction of additional flavorings in cigars or any tobacco products will make comparison between categories more difficult.
The ROS produced by tobacco products are complex and not fully understood. However, it can be broken down into two phases, the gas and the tar phase [21]. Mainstream tobacco smoke constituents react with each other forming reactive oxygen and nitrogen species, including superoxide, hydrogen peroxide, hydroxyl radical, and peroxynitrites [22]. Oxidation of biomolecules, such as proteins and lipids and the formation of DNA lesions are mechanisms of toxicity of tobacco smoke exposure and disease progression [23]. It is widely known that cigarette smoke generates ROS and has significant carcinogenic properties as depicted in dosedependent studies [24,25]. In this study, we observe the ROS generated by most of the cigarillos tested in our study is significantly higher than conventional research cigarettes (University of Kentucky 3R4F or 1R6F). This suggests that cigarillos and cigarettes not only induce higher oxidative stresses, but different brands of cigarillos and little cigars showed variable toxicological effects dependent on different chemicals generated during combustion. Corroborating our data, another study showed a greater emission of total semivolative organic compounds (SVOC) and volatile organic compounds (VOC) in cigarillo smoke compared to regular cigar or cigarette smoke [26]. Most notably, between methods #1 and #2, it is easy to see that each method gives significantly different results even when testing the same cigarillo. Within method #1, many of the cigarillos burned down faster than others from the same brand and flavor, which might use multiple cigars during one run of measurement that might cause batch-batch variations. For example, wine flavors from methods one and two one can again see a wide variation, where PT Rillos is found to be significantly higher than Dutch Masters using method #2, but method #1 shows an opposite trend.Currently, regulations on cigarettes focus on very few ingredients when compared to the total 5,000 chemicals found in them, allowing for variation between brands and batches [27]. This variation is also seen in other tobacco products, such as e-cigarettes [28,29]. In agreement with our findings, Hamad et al recently reported that the flavored cigars/cigarillos (Cheyenne menthol and Swisher sweets original and cherry) have more potentially harmful constituents (HPHCs) including volatile organic hydrocarbons, increased tobacco mass, and total particulate matter (TPM) compared with conventional research reference cigarettes (3R4F) [4]. Hence, HPHCs and particulate matter (PM) including ROS released by cigars would contribute to cellular/tissue injury in the lungs. Cigar smoke produces fine particles altering iron homeostasis, which would easily deposit in the lungs and leading to damage of nearby lung tissue [18]. Our data showed various ranges of particles from flavored cigars (from 1.0 μm to 4.0 μm with higher mass concentration) vs. research-grade research and filtered cigars. In flavored cigars, the particle distribution/concentration was significantly higher in compared to other filtered cigars or research cigarettes, suggesting that composition of flavored cigars generate broader particles with higher concentrations which might be even more harmful than cigarette smoke [30][31][32]. Further studies are required to determine the grading of various commercially available cigars for their HPHCs, volatile organic compounds (VOCs), and TPM, and their products chemical analyses to ascertain the toxicity in the realm of cellular toxicity especially with flavored cigars/cigarillos and flavoring chemicals used.
In a recent study, it has been shown that increased concentration of the cigar extract led to increased cell death [1,33]. Although our study found that cigarettes had a higher percentage of cell death compared to most cigarillos or cigars, other studies have noted little cigars to be more toxic when compared to cigarettes [1,33]. This could be due to the increased amount of chemicals that are found/released in little cigars as compared to cigarettes [33], or the differential interactions of various chemicals during combustion and cellular components. When compring at other tobacco products like e-cigarettes, cell damage still occurs even with the absence of nicotine [34,35].
Other studies have shown, in particular, the flavorings cinnamaldehyde, O-vanillin, and pentanedione are significantly more cytotoxic on human cells than other flavoring chemicals [36]. Cinnamaldehyde and other aldehyde compounds are frequently used as flavorings, although the FDA and Flavors Extract Manufacturers Association (FEMA) both agree they have adverse effects on human health [37]. Grape and alcohol were common flavors for cigarillos and little cigars. The grape flavoring in food is often attributed to the compound methyl Beas 2B cells were treated with various flavored and unflavored CSE for 24 hours. Each cigarillo treatment was done in 2 concentrations, 0.5% and 1.0%, these are depicted respectively for each group. Flavor is listed with brand in parenthesis. A 3R4F reference cigarette was used for comparison and untreated cells under the same conditions were used as a control. Cell viability was measured using Nexcelom's Cellometer and AOPI dye. Viability percentages are shown as mean ± SEM, and significance was determined by one-way ANOVA. � p< 0.05, �� p< 0.01, and ��� p< 0.001 as compared with controls.
https://doi.org/10.1371/journal.pone.0226066.g008 anthranilate, and alcohol flavors come from ethyl-decadienoate. These compounds are considered safe in food; however, there is little information on the combustion and inhalation toxicology effects of these chemicals [38]. Our data demonstrated that higher cytotoxic responses with fruits/candy and drinks/beverages/alcohols flavored cigars compared to tobacco flavor, which points the harmful possibility of the flavored chemicals. More detailed studies of those flavored chemicals, such as methyl anthranilate or ethyl-decadienoate, are needed to further understand the mechanisms.
Due to the high variability of each manufacturer's/company's flavorings (including batch to batch variations), future research should be directed with cellular methods and mouse models to take a further step in studying the respiratory toxicity [35]. In addition, it would also be beneficial to determine which chemical flavoring (in flavors) is the most deleterious. This has already been conducted in flavors present in e-liquids by studying barrier dysfunction and cellular toxicity in a cellular system [39]. Once more information on the chemical make-up of flavorings and cigars are obtained, the regulatory agencies should increase regulations and/or ban these flavored cigars. Our data showing differential responses due to the different flavors, manufacturers may limit the use of the chemical ingredients in the most toxic flavors (subject to products analysis), or remove flavors all together from cigar products.
In conclusion, despite the high brand-to-brand variability between flavored cigars or cigarillos, those with fruit, candy, and drink flavors tend to show more deleterious effects in our assays. Further, it is difficult to ascertain the consistent toxicity data due to the differences in chemical composition within a specific flavor of the cigar and cigarillo. Flavored cigars produced maximum particles/particulates compared with research cigarettes. The cellular assays resulted in variable cytotoxic responses with fruits/candy and drinks/beverages being more cytotoxic. Due to the irregular composition of flavors and flavoring chemicals, it is important to regulate flavored little cigars/cigarillos. Future research may be directed to determine the cellular methods as well as in vivo models to fully determine the flavoring toxicity in flavored little cigars/cigarillos. Supporting information S1 Table. A