CC BY-NC-ND
0rr'P'WM AIIIII E IN PRESS
Mutation Research xxx (2016) xxx-xxx
ELSEVIER
Contents lists available at ScienceDirect
Mutation Research/Genetic Toxicology and Environmental Mutagenesis
journal homepage www.elsevier.com/locate/gentox Community address www.elsevier.com/locate/mutres
The mutagenic assessment of an electronic-cigarette and reference cigarette smoke using the Ames assay in strains TA98 and TA100
D. Thorne^*, I. Crooks3, M. Hollingsb, A. Seymourb, C. Meredith3, M. Gaca3
a British American Tobacco, Group R&&D, Southampton, Hampshire SO15 8TL, United Kingdom b Covance Laboratories Ltd, Otley Road, Harrogate, North Yorkshire HG3 1PY, United Kingdom
ARTICLE INFO
ABSTRACT
Article history: Received 7June 2016 Received in revised form 28 September 2016 Accepted 25 October 2016 Available online xxx
Keywords:
E-cigarettes
Cigarette smoke
Aerosol exposure
Salmonella typhimurium strains TA98 and TA100 were used to assess the mutagenic potential of the aerosol from a commercially available, rechargeable, closed system electronic-cigarette. Results obtained were compared to those for the mainstream smoke from a Kentucky reference (3R4F) cigarette. Two different test matrices were assessed. Aerosol generated from the e-cigarette was trapped on a Cambridge filter pad, eluted in DMSO and compared to cigarette smoke total particulate matter (TPM), which was generated in the same manner for mutagenicity assessment in the Salmonella assay. Fresh e-cigarette and cigarette smoke aerosols were generated on the Vitrocell® VC 10 smoking robot and compared using a modified scaled-down 35 mm air agar interface (AAI) methodology.
E-cigarette aerosol collected matter (ACM) was found to be non-mutagenic in the 85 mm plate incorporation Ames assay in strains TA98 and TA100 conducted in accordance with OECD 471, when tested up to 2400 ^g/plate. Freshly generated e-cigarette aerosol was also found to be negative in both strains after an AAI aerosol exposure, when tested up to a 1 L/min dilution for up to 3 h. Positive control responses were observed in both strains, using benzo[a]pyrene, 2-nitrofluorene, sodium azide and 2-aminoanthracene in TA98 and TA100 in the presence and absence of metabolic activation respectively. In contrast, cigarette smoke TPM and aerosol from 3R4F reference cigarettes were found to be mutagenic in both tester strains, under comparable test conditions to that of e-cigarette exposure.
Limited information exists on the mutagenic activity of captured e-cigarette particulates and whole aerosol AAI approaches. With the lower toxicant burden of e-cigarette aerosols compared to cigarette smoke, it is clear that a more comprehensive Ames package of data should be generated when assessing e-cigarettes, consisting of the standard OECD-five, TA98, TA100, TA1535, TA1537 (or TA97) and E. coli (or TA102). In addition, TA104 which is more sensitive to the carbonyl based compounds found in e-cigarette aerosols under dry-wicking conditions may also prove a useful addition in a testing battery. Regulatory standard product testing approaches as used in this study will become important when determining whether e-cigarette aerosols are in fact less biologically active than cigarette smoke, as this study suggests. Future studies should be supported by in vitro dosimetry approaches to draw more accurate comparisons between cigarette smoke, e-cigarette aerosol exposure and human use.
© 2016 Published by Elsevier B.V.
Abbreviations: AAI, air agar interface; AAN, 2-aminoanthracene; ACM, aerosol collected matter; ALI, air liquid interface; Air, air control; B[a]P, benzo[a]pyrene; CRM No 81, CORESTA recommended method No 81; DMSO, dimethyl sulfoxide; e-cigarette, electronic cigarette; HCI, Health Canada Intense; ISO, International Standards Organization; OECD, Organization for Economic Co-operation and Development; TPM, total particulate matter; SD, standard deviation; UTC, untreated control; WA, whole aerosol.
* Corresponding author. E-mail address: DavidThorne@bat.com (D. Thorne).
http://dx.doi.org/10.1016/j.mrgentox.2016.10.005 1383-5718/© 2016 Published by Elsevier B.V.
1. Introduction
The Ames or bacterial reverse mutation assay has been globally recognised by scientific communities and government agencies to determine the mutagenic potential of new chemicals and drugs, and is used as an initial screen and/or early development assay for the assessment of mutagenicity. Specific assay parameters and methodologies are covered in several international regulatory guidelines, including those by the Organization for Economic Co-operation and Development (OECD Guideline 471) [1] and the International Commission on Harmonization (ICH S2R1) [2]. These guidelines ensure the uniformity and robustness of data submit-
^lUf IIIIIIII E IN PRESS
2 D. Thorne et al. / Mutation Research xxx (2016) xxx-xxx
ted to regulatory agencies for registration or acceptance of many chemicals. The Ames assay has been used for the assessment of tobacco smoke using both TPM (total particulate matter) and whole aerosol approaches, and has been shown to be mutagenic in many strains, including TA98, TA100, TA1537, YG1024, YG1042, and TA104 (+/-S9) [3-12]. In terms of cigarette smoke assessment, the Ames assay has been extensively characterised, developed and deployed, and there is a wealth of historical data in which existing and modified tobacco products and even new categories, such as electronic-cigarettes (e-cigarettes), can be assessed against. E-cigarettes are currently being widely used and the awareness of these products are growing globally, as is the development of this new category [13].
Current research suggests that e-cigarette aerosols are less biologically active compared to traditional cigarette smoke. Some studies have demonstrated clear toxicological properties of e-cigarettes, whereas others have identified no activity at all. In contrast to cigarette smoke, which has been extensively investigated and the effects of smoking documented, e-cigarette aerosols remain poorly understood and characterised in vitro. Given the vast amount of information available on cigarette smoke and positive control chemicals known to elicit a mutagenic response, the Ames assay has been relatively under-utilised for the assessment of e-cigarette aerosols. Table 1 demonstrates the in vitro mutagenicity data available on TA98 and TA100 for cigarette smoke in both TPM and aerosol studies. In contrast, limited information exists on the comparative assessment of e-liquids, TPM and e-cigarette aerosol collected matter (ACM), which is comparable to cigarette smoke TPM and whole aerosol approaches (Table 1).
To compound this, OECD recommend a testing battery of at least five tester strains to capture the full range of chemical interactions and mutagenic events acting via different modes of action. To date, only TA98 and TA100 have been directly employed in the testing of any test e-cigarette related test matrix. TA98 is frequently deployed for cigarette smoke testing because it is sensitive to basic and neutral fractions, such as the heterocyclic amines and aromatic amines that are one of the primary sources of mutagenicity in TPM and smoke extracts [16].TA100 has also been used because of its added sensitivities compared to TA98 and its ability to distinguish between tobacco products [16]. Both strains are part of the OECD recommended protocol, which consists ofTA98, TA100, TA1535, TA1537 (or TA97) and E.coli (or TA102) [1]. To support the use of multiple strains, a study of 224 chemicals tested by the National Toxicology Program in a variety of Salmonella typhimurium strains, TA98, TA100, TA1535, TA1537 and TA97, demonstrated that TA100 was the most sensitive strain, detecting 83% of the mutagens present in the study. In contrast TA98 detected only 67%. When used in combination, TA98 and TA100 were capable of detecting approximately 93% of all mutagens present in the study. Detection sensitivity was further increased with the addition of TA1535, TA1537 and TA97 [17]. Therefore, strains TA98 and TA100 were selected for use in an initial screen and allows researchers to con-textualise the responses obtained against a wealth of historical cigarette smoke data.
Due to the lack of scientific data on the toxicological impact or mutagenic potential of e-cigarette aerosols, this study has set out to preliminarily assess two e-cigarette test matrices in tester strains TA98 and TA100. The first matrix was e-cigarette aerosol collected on a Cambridge filter pad (CFP) and eluted in a solvent. The second matrix was a freshly generated e-cigarette aerosol directly applied to the bacterial surface at the air-agar interface (AAI). TPM/ACM capture techniques focus on capturing the particulate phase of the aerosol on a CFP. The particulates captured are then eluted from the CFP surface typically using anhydrous dimethyl sulfoxide (DMSO) to create a solvent particulate based test matrix. In contrast, freshly generated aerosols are delivered directly to the agar exposure sur-
face. This efficient and direct transfer captures the interactions of the aerosol mixture and the various phases, without subsequent solvent extraction. This process captures any volatile or chemical species that are generated through the aerosolisation process. These volatiles may not be efficiently captured or even present in TPM/ACM matrices.
This study, under the conditions assessed, supports the hypothesis that e-cigarettes have a reduced mutagenic potential compared to cigarette smoke in tester strains TA98 and TA100 in both partic-ulate and aerosol test matrices.
2. Materials and methods
2.1. Chemicals and reagents
All chemicals and reagents were obtained from Sigma-Aldrich (Gillingham, UK) unless otherwise stated.
2.2. Study design
All treatments were conducted on tester strains TA98 and TA100 only. E-cigarette test matrices were compared to an equivalent 3R4F reference cigarette smoke test matrix. TPM was generated in accordance with established conditions [5]. ACM was generated in a comparable manner to TPM. TPM and ACM exposures were conducted in accordance with OECD 471. This study also investigated direct aerosol exposure methodologies in both strains under established AAI conditions [12]. All aerosol exposures were conducted for a maximum of 3 h at the AAI. TPM/ACM and cigarette smoke exposures were conducted in the presence of metabolic activation only, whereas e-cigarette aerosol exposures, were conducted in the presence and absence of metabolic activation.
2.3. Reference cigarettes and e-cigarettes
3R4F reference cigarettes were obtained from the University of Kentucky, Kentucky, USA. Prior to smoking, cigarettes were conditioned for at least 48 h at 22 ±1 °C and 60 ±3% relative humidity according to International Organisation of Standardization (ISO) 3402:1999 [18]. E-cigarettes (Vype® ePen) were obtained from Nicoventures Trading Ltd., UK (www.govype.com). Vype® ePen is a rechargeable, dual voltage, closed modular system, consisting of two segments. A rechargeable battery section and a replaceable liquid (e-liquid) containing cartridge (cartomizer), with two voltage settings, 4V and 3.6 V (4 V used in the study). Vype® ePen e-liquid cartridges (Blended Tobacco Flavour) contained 18 mg/mL nicotine and were stored at room temperature. Vype® e-liquids are formulated in the UK, using pharmaceutical/food grade ingredients (Fig. 1 and Table 2). Throughout the study, cigarettes were smoked to the HCI smoking regime as per the Health Canada Official Method T-115 [21], whereas e-cigarettes were puffed to the CORESTA Recommended Method No 81 (CRM No 81) regimen [22], as demonstrated in Table 3.
2.4. Bacterial strains and culture conditions
Bacterial strain TA98 was obtained from National Collection of Type Cultures, UK. Strain TA100 was obtained from Covance Laboratories Inc., USA. Prior to optimisation of culture conditions, each strain was checked for strain characteristics and antibiotic resistance according to Maron and Ames [19] and De Serres and Shelby [20]. Overnight cultures were prepared from frozen stocks in nutrient broth, containing appropriate antibiotics to maintain plasmids, and grown in a shaking incubator at 37 °C for 8-10 h. Metabolic activation was achieved by using Aroclor 1254-induced rat liver S9
^ ARTICLE in PRESS
D. Thorne et al. / Mutation Research xxx (2016) xxx-xxx 3
Table 1
Review of current in vitro Ames data relative to e-cigarette and cigarette smoke test matrices.
Product TPM/ACMa E-Liquida Whole Aerosolb
TA98 TA100 TA98 TA100 TA98 TA100
Reference Cigarettec e-cigarette Mutagenic [3-6] Non-mutagenic[14] Mutagenic[3-5,7,8] Non-mutagenic[14] N/A Non-mutagenic[14] Non-mutagenic [14] Mutagenic [9-12] Non-mutagenic [15] Mutagenic[9-12] NDA
TPM = total particulate matter. ACM = aerosol collected matter. N/A = not applicable. NDA = no data available. a Standard plate methodology. b Air liquid/air agar interface exposure. c Reference cigarettes 1R4F, 2R4F and 3R4F.
2.5. TPM/ACM collection
3R4F reference cigarettes were smoked on a Borgwaldt RM200A (Borgwaldt-KC, Hamburg, Germany) rotary machine using Health Canada smoking regime (55 mL puff volume, of 2 s duration, every 30 s, 100% vent blocking using a bell shaped puff profile) [21]. Vype® ePen was puffed on a Borgwaldt LM20X (Borgwaldt-KC, Hamburg, Germany) linear machine, using the CORESTA e-cigarette puffing parameters of a 55 mL puff volume, of 3 s duration, every 30 s and a square-wave puff profile. Up to 150 mg ACM were collected onto 44 mm Cambridge filter pads (Whatman, Maidstone, UK). Pads were weighed pre and post smoking/vaping, to determine deposited material. Pads were eluted with DMSO to a final stock concentration of 24mg/mL. TPM and ACM extracts were stored in single-use aliquots immediately at -80°C. Ames plates were exposed at a concentration range of 0 to 2400 ^g/plate.
Fig. 1. Schematic representation of Vype® ePen (e-cigarette) compared to a traditional combustible cigarette.
2.6. Whole aerosol exposure
Table 2
Specification of products used in the study.
Characteristics
Product
Vype® ePena
Product category Manufacturer Length (mm) Diameter (mm) Nicotine content Puff number
Cigarette e-cigarette
University of Kentucky (USA) Vype® (Nicoventures, UK) 84 153
8 20 (10 at mouth piece)
0.7-2.0 mg/cigb 18 mg/mL (1.8% v:v)c
8-10b 250-300
Test matrix assessed TPM and Aerosol
ACM and Aerosol
TPM = total particulate matter. ACM = aerosol collected matter.
a e-cigarette, closed system modulardevice, operated at 4 V with Blended Tobacco cartridges.
b Dependent on smoking regimen used (ISO vs. HCI). c As stated on the pack.
(male Sprague-Dawley rats) was obtained from MolTox®, Molecular Toxicology, Inc, USA.
The Vitrocell® VC 10 Smoking Robot (Vitrocell® systems, Waldkirch, Germany) serial number VC10/090610, was used to expose bacteria to freshly generated aerosols. The VC 10 is a rotary-head smoking machine which has a single syringe that transfers generated aerosol to an independent, continuous airflow dilution bar. Aerosol dilution is achieved via turbulent mixing with diluting air in the dilution bar and different aerosol exposure conditions are achieved by increasing or decreasing the diluting airflow. A vacuum is used to sample the aerosol (via negative pressure) from the dilution bar into the module, which docks directly under the dilution bar [25]. Diluting airflow rates within this system were maintained using mass flow controllers (Analyt-MTC GmbH, Mülheim, Germany).
Triplicate bacterial plates were exposed in Vitrocell® AMES 4 stainless steel exposure modules. The trumpet height within the module was set to 2 mm above the agar surface. Diluting airflows of 12,8,4 and 1 L/min with a fixed vacuum of 5.0 mL/min/well, were assessed. Ames plates were exposed to diluted cigarette smoke and e-cigarette aerosols between 24 and 180 min in the presence and absence of metabolic activation.
Table 3
Aerosol generation regimens.
Product PuffRegimen PuffVolume (mL) Puff Frequency (s) Puff Duration (s) Puff Profile Vent blocking Coil pre-activation(s)
Cigarette HCIa 55 30 2 Bell 100% N/A
Vype® ePenc CRM 81b 55 30 3 Square N/A 1
N/A = not applicable a HC1T-115 [21]. b CRM No 81 [22]. c Button activated.
^lUf IIIIIIII E IN PRESS
4 D. Thorne et al. / Mutation Research xxx (2016) xxx-xxx
2.7. Ames TPM/ACM assay
For all experiments, bacteria were cultured at 37 ±1 °C for 10 h in nutrient broth, containing ampicillin, to provide bacterial cultures in the range of 109 cells/mL, based on cell count data from each strain batch.
3R4F TPM and Vype® ePen ACM samples were tested for mutation and toxicity at concentrations ranging from 0 to 2400 |ig/plate. Both test articles were treated concurrently in each experiment. Vehicle and positive controls were included in quintuplicate, and triplicate respectively. Positive controls, Benzo[a]pyrene (B[a]P) and 2-aminoanthracene (AAN) were used for TA98 and TA100 in the presence of metabolic activation at concentrations of 10 |ig/plate and 5 |ig/plate, respectively.
Platings were achieved by the following sequence of additions to molten agar at 46 ± 1 °C, 0.1 mL bacterial culture, 0.1 mL test article solution or control 0.5 mL, 10% S9 mix (or buffer solution for diagnostic control treatments only) followed by rapid mixing and pouring on to Vogel-Bonner E agar plates. When set, the plates were inverted and incubated at 37 ±1 °C protected from light for 3 days. Following incubation, these plates were examined for evidence of toxicity to the background lawn, and revertant colonies were counted. The first treatment included a 60 min pre-incubation step in order to potentially maximise the detectable mutagenic spectrum of the test articles. Quantities of test article or control solution, bacteria and S9 mix, with the addition of 0.5 mL 100 mM sodium phosphate buffer, were mixed together in sterile treatment tubes and incubated for 60 min at 37 ± 1 °C with shaking, before the addition of 2.5 mL molten agar at 46 ± 1 °C. Plating of these treatments then proceeded as per plate-incorporation procedure.
2.8. Ames aerosol assay
The Ames assay used in this study is modified from the standard 85 mm methodology and is described in detail in Kilford [11] and Thorne [12]. Briefly, whole aerosol exposures were performed in a scaled-down 35 mm plate format (Grenier Bio-One). Approximately 2 x107 bacteria cells were mixed with 75 |L sodium phosphate buffer (pH 7.4) or a 10% S9 mix, prepared according to Ames [23] and complemented with 40 |g/mL histidine and 48.8 |g/mL biotin mix. The bacterial cell suspension was plated directly onto Vogel-Bonner agar using a spread plate technique and incubated at 37°C until dry (~20min) before transferring to Vitrocell® AMES 4 exposure modules for exposure at the AAI (Fig. 2).
Concurrent controls (air, untreated and positive) were included with each exposure.
Air controls were exposed to a constant flow of filtered air (0.2 L/min diluting air flow, 5.0 mL/min/well vacuum flow rate), comparable to that of the aerosol exposure. Untreated and positive controls were maintained at room temperature for the duration of the exposure and then processed as per the rest of the experiment. Positive controls Benzo[a]pyrene (B[a]P) and 2-aminoanthracene (AAN) were used for TA98 and TA100 in the presence of metabolic activation at concentrations of 0.8 | g/plate and 0.4 |g/plate, respectively. In the absence of metabolic activation, 2-nitrofluorene (2NF) and sodium azide (NaN3) were used at 0.4 and 1.0 |g/plate for strains TA98 and TA100 respectively. Concentrations of positive control chemicals were less for the aerosol technique than the TPM/ACM technique, due to the scaled down nature of the assay (35 mm compared to 85 mm plate format).
Following exposure, the plates were removed from the exposure modules, sealed, inverted and incubated at 37 °C in the dark for 3 days. Cigarette smoke exposures were conducted for up to 24 min, whereas e-cigarette aerosol exposures were conducted for up to 3 h.
2.9. Data evaluation and acceptance criteria
Plates were scored using an automated colony counter (Sorcerer Image Analyser, Perceptive Instruments, Haverhill, UK) and the background lawn inspected for signs of toxicity, defined by a thinning of the background bacterial lawn or a marked reduction in revertant numbers. Manual scoring was used where confounding factors affect the accuracy of the automated counter. Responses with positive control chemicals were compared with laboratory historical observed ranges. Observed values were comparable with historical control ranges held at Covance laboratories (Harrogate, UK) for the standard 85 mm plate assay and established ranges for the scaled-down 35 mm AAI assay. Data were evaluated using fold increase in revertant numbers, over the concurrent zero or air control plate counts, and analysed statistically using Dunnett's test. For an increase in revertant numbers to be considered as a muta-genic response, increases were required to be at least 2-fold greater than the concurrent control or statistically significant (p <0.05) using Dunnett's test, and both concentration-related and reproducible over two or more independent experiments as per OECD 471 guidelines [1]. Experiments were conducted on a minimum of 3 independent occasions, with at least 3 replicates per occasion. Results are presented as mean revertants/plate ± standard deviation (SD).
3. Results
Tables 4, 5 and 6 summarises all the data obtained in the presence and absence of S9 metabolic activation for both TPM/ACM and aerosol treatments respectively.
3.1. TPM/ACM
All TPM/ACM experiments were conducted using final concentrations of 0, 50, 100, 150, 200, 250, 300, 500, 1000 and 2400 |g/plate, plus vehicle and positive controls. In all experiments, the vehicle and positive controls were within the historical control ranges for the laboratory. All samples were completely soluble in the aqueous assay system at all concentrations tested and no precipitation was observed.
For cigarette smoke TPM exposure, both tester strains demonstrated marked increases in revertant numbers which were reproducible over all experimental occasions and statistically significant when the data were analysed at the 1% level using Dunnett's test. In TA98 and TA100 increases were concentration related, in most cases only up to 1000 |g/plate, above which the mutagenic responses tailed off, which was considered to be an indication of toxicity, through thinning of the bacterial lawn.
No consistent or concentration-related effects were observed with e-cigarette ACM in either TA98 or TA100 in the presence of S9. Observed fluctuations in the revertant numbers were attributable to normal biological variability and low vehicle controls counts rather than any compound-related effects. Data demonstrated that there were no increases above the two-fold threshold or that were statistically significant. There were no concentration-related or reproducible increases and no evidence of toxicity or thinning of the bacterial lawn were observed at any concentration tested in either strain.
Positive controls run concurrently with each exposure, demonstrated a clear positive response under sub toxic conditions and the untreated control demonstrated background spontaneous rever-tant numbers for each strain. For example, benzo[a]pyrene (B[a]P) at 10 |g/plate produced a clear positive response in TA98 at approximately a 6-fold increase in revertant numbers whereas, 2-aminoanthracene (AAN) at 5 | g/plate produced a clear positive
G Model
MUTGEN-402782; No.ofPages10
ARTICLE IN PRESS
D. Thorne et al. / Mutation Research xxx (2016) xxx-xxx
Fig. 2. A schematic representation of the Vitrocell® VC 10 aerosol exposure system. [A] Software and air-flow controller. [B] Smoking Robot and ventilation hood for aerosol generation. [C] Piston/syringe which draws the puff and delivers the aerosol to the dilution system. [D] Dilution, transit and delivery of aerosol occurs in the dilution bar. [E] Smoke exposure module which holds the scaled-down 35 mm Ames plates [24].
Table 4
Results—3R4F TPM compared to Vype® ePen ACM using 85 mm plate incorporation (+S9).
TA98 Concentration|xg/plate 3R4F MeanRevertants/Plate (±SD) MeanFold Increase Statistical analysis Vype® ePen MeanRevertants/Plate (±SD) MeanFold Increase Statistical analysis
0 39.2 (2.4) 39.2 (2.4)
50 63.8 (5.9) 1.6 NS 43.7 (7.2) 1.1 NS
100 105.3(11.2) 2.7 * 43.4 (4.6) 1.1 NS
150 164.2(11.8) 4.1 * 36.9 (2.7) 0.9 NS
200 212.4(16.8) 5.4 * 35.9 (2.2) 0.9 NS
250 253.6(10.5) 6.5 * 36.2 (4.2) 0.9 NS
300 339.0 (31.6) 8.6 * 33.9(1.7) 0.9 NS
500 453.9 (31.3) 11.6 * 35.7 (4.5) 0.9 NS
1000 550.8 (42.9) 14.0 * 37.2 (4.3) 0.9 NS
2400 498.7 (44.3) 12.7 * 36.7 (4.9) 0.9 NS
0 113.1 (12.2) ~ 113.1 (12.2) ~ ~
50 142.4(7.6) 1.3 NS 125.9 (5.5) 1.1 NS
100 139.2 (9.0) 1.2 NS 122.3(4.6) 1.1 NS
150 156.0(156.1) 1.4 NS 120.0(7.3) 1.1 NS
200 176.0 (6.9) 1.6 NS 118.4 (4.7) 1.0 NS
250 195.2(11.3) 1.7 NS 122.7(8.0) 1.1 NS
300 226.0(10.1) 2.0 * 135.8(13.2) 1.2 NS
500 310.0(16.7) 2.7 * 131.5 (5.6) 1.2 NS
1000 393.0(13.3) 3.5 * 120.7 (6.3) 1.1 NS
2400 366.7 (27.8) 3.2 * 119.7(8.0) 1.1 NS
TPM = total particulate matter. ACM = aerosol collected matter. NS = not significant.
* = significantly different compared to Zero concentration (p = <0.01). ~ = not analysed.
response in TA100 at approximately a 5-fold increase in revertant numbers.
The results show that e-cigarette ACM did not induce mutations in TA98 or TA100 above the background untreated/vehicle control, whereas cigarette smoke TPM in both strains elicited a positive response which was statistically significant at the majority of concentrations, reproducible between experiments and concentration dependant (Fig. 3).
3.2. Aerosol exposures
All aerosol experiments were conducted in strains TA98 and TA100 in the presence (and absence for e-cigarette exposure) of metabolic activation (S9). Exposures were based on a scaled-down 35 mm plate format using dilutions of 12, 8, 4 and 1 L/min with a 5 mL/min vacuum as previously described [11,12].
3.3. Cigarette smoke aerosol
This study investigated cigarette smoke generated under a Health Canada Intense (HCI) smoking regimen, which delivers a higher puff volume over a shorter timeframe compared to an ISO equivalent. The results from this study demonstrate that a 24min HCI exposure with cigarette smoke is enough to elicit a mutagenic response in both TA98 and TA100 at the lowest smoke dilution of 1 L/min in the presence of metabolic activation (Fig. 4). At the 1 L/min dilution, both strains TA98 and TA100 showed signs of thinning of the background lawn, indicating toxicity had been reached. Both strains showed a positive response to cigarette smoke in the presence of metabolic activation, indicated by a greater than two-fold increase over air control and statistically positive dose response when assessed using Dunnett's test. Untreated control and air controls were not significantly different from each other and remained negative in the absence and presence of metabolic activation, indicating that exposure conditions had no effect on
^lUf HIIIIII.E IN PRESS
6 D. Thorne et al. / Mutation Research xxx (2016) xxx-xxx
Table 5
Results—Cigarette smoke compared to e-cigarette aerosol using a scaled down 35 mm AAI methodology (+S9).
Treatment 3R4Fa Vype® ePenb
MeanRevertants/Plate (±SD) MeanFold Increase Statistical analysis MeanRevertants/Plate (±SD) MeanFold Increase Statistical analysis
UTC 5.9(1.2) 4.9 (0.1)
B[a]P 38.0 (9.5) 10.0 * 56.4(10.1) 8.6 *
Air Controlc 3.8 (0.7) 6.6 (0.1)
12L/min 7.o(l.8) 1.8 NS 4.2(1.2) 0.6 NS
8 L/min 21.7 (5.2) 5.7 * 5.7 (0.9) 0.9 NS
4L/min 43.7 (8.0) 11.5 * 4.8 (0.7) 0.7 NS
1 L/min 16.7(l0.8) 4.4 * 4.2 (0.2) 0.6 NS
UTC 20.7 (0.5) 20.2 (4.5)
AAN 159.2 (49.3) 8.1 * 323.0 (55.2) 16.0 *
Air Controlc 19.7(2.8) 20.3 (3.9)
12 L/min 22.5 (3.5) 1.1 NS 18.2(3.1) 0.9 NS
8 L/min 32.2 (0.2) 1.6 NS 19.5 (7.3) 1.0 NS
4 L/min 41.7 (0.5) 2.1 * 17.5 (4.5) 0.9 NS
1 L/min 30.3 (19.8) 1.5 NS 20.2 (8.2) 1.2 NS
AAI = air agar interface. UTC = untreated control. NS = not significant.
* = significantly different compared to Air Control (p = < 0.05). ~ = not analysed.
B[a]P (benzo [a] pyrene) = 0.8 |g/plate. AAN (2-aminoanthracene) = 0.4 |g/plate. a 3R4F aerosol generated over 24 min. b ePen aerosol generated over 180 min.
c Data compared to Air Control obtained concurrently during exposure.
Table 6
Results—e-cigarette aerosol using a scaled down 35 mm AAI methodology (-S9).
Treatment Vype® ePena
MeanRevertants/Plate (±SD) MeanFold Increase Statistical analysis
UTC 4.9(1.1)
2NF 143.9(13.1) 23.1 *
Air Controlb 6.3 (0.4)
12L/min 3.3 (2.4) 0.5 NS
8 L/min 5.5(1.2) 0.9 NS
4 L/min 2.8(1.6) 0.5 NS
1 L/min 5.2 (0.7) 0.8 NS
UTC 22.1 (4.1)
NaNa 335.6 (60.9) 17.3 *
Air Controlb 20.5 (8.7)
12 L/min 17.2 (4.5) 0.8 NS
8 L/min 18.3(10.8) 0.9 NS
4 L/min 24.0(1.9) 1.2 NS
1 L/min 18.2 (6.4) 0.9 NS
AAI = air agar interface. UTC = untreated control.
NS = not significant when compared to Air Control. 2NF (2-nitrofluorene) = 0.4 |xg/plate. NaN3 (sodium azide)= 1 |xg/plate. * = significantly different compared to UTC (p = <0.05). ~ = not analysed. a ePen aerosol generated over 180 min.
b Data compared to Air Control obtained concurrently during exposure.
revertant numbers. Positive controls B[a]P and AAN provided significant increases in strains TA98 and TA100 respectively in the presence and absence of metabolic activation (Fig. 4).
3.4. E-cigarette aerosol
Preliminary investigations assessed a 64 min e-cigarette aerosol exposure in the presence of S9. No response was observed in either TA98 or TA100 tester strains. Therefore, in the presence and absence of metabolic activation, e-cigarette aerosol exposures were
extended to 3 h using established conditions. A statistically positive response over a two-fold threshold did not occur in either TA98 or TA100 in any of the experiments tested, with or without metabolic activation. All e-cigarette aerosol treatment data were comparable to the air control and demonstrated no statistically significant differences, despite clear mutagenic responses from positive controls, B[a]P and AAN. No thinning of the background lawn other or toxic effects were observed (Fig. 5).
^ ARTICLE IN PRESS
Fig. 3. Response to TPM treatment in the presence of S9 metabolic activation. [A] TA98 control responses. [B] TA98 responses to 3R4F cigarette and Vype® ePen e-cigarette particulates [C] TA100 control responses. [D] TA100 responses to 3R4F cigarette and Vype® ePen e-cigarette particulates. UTC = untreated control/vehicle control, B[a]P = Benzo[a]pyrene [10 |xg/plate], AAN = 2-aminoanthracene [5 |xg/plate].
Table 7
Positive controls in the presence and absence of metabolic activation.
Exposure method TA98 TA100
-S9 +S9 -S9 +S9
85 mm plate incorporation technique3 35 mm AAI spread plate method' n/a 2NF0.4 |xg/plate B[a]P 10 |xg/plate B[a]P 0.8 |xg/plate n/a NaN3 1.0 |xg/plate AAN 5 |xg/plate AAN 0.4 |xg/plate
n/a= not assessed. B[a]P = Benzo[a]pyrene. AAN = 2-aminoanthracene. 2NF = 2-nitrofluorene. NaN3 = Sodium azide. AAI = air agar interface. a Standard 85 mm plate incorporation assay used forTPM/ACM exposure matrices. b Scaled down 35 mm air agar interface spread plate methodology developed for aerosol exposures.
3.5. Controls
Irrespective of exposure conditions, positive controls were run concurrently with each exposure in the presence and absence of metabolic activation, to confirm strains were positively responsive. Untreated (UTC) and/or air controls (Air) were also run concurrently with every experiment to establish spontaneous revertant numbers. Positive control information can be found in Table 7.
4. Discussion
This study has investigated the mutagenic potential of a commercially available e-cigarette (Vype® ePen, Nicoventures, UK), compared to reference cigarette smoke (3R4F), in tester strains TA98 and TA100 using two different exposure scenarios. Firstly, particulate based TPM trapping techniques were used, which have been extensively described for cigarette smoke assessment [6]. For
e-cigarette ACM capture, a comparable technique to TPM was used, whereby an e-cigarette aerosol was captured on a Cambridge filter pad and eluted for testing. Secondly, freshly generated smoke and e-cigarette aerosols were directly assessed, using an aerosol generation system and an AAI methodology [24-26].
Using TPM trapping techniques, with metabolic activation and exposure up to 2400 ^g/plate, 3R4F TPM produced a clear mutagenic response in strains TA98 and TA100. This response is consistent with previous reported literature based on reference cigarette smoke (1R4F, 2R4F and 3R4F) [3-8]. However, e-cigarette ACM exposed up to 2400 ^g/plate, did not demonstrate any muta-genic activity under the same test conditions. Particulate trapping techniques focused on the particulate fraction of the aerosol and require the capture and elution of a solvent-based test matrix. Whole aerosol testing techniques capture the aerosol dynamics and represents an exposure matrix more comparable to human use [27]. For example, there are no solvent dilutions, extractions, or inter-
ШШШВШШ^Ш ARTICLE IN PRESS
D. Thorne et al. / Mutation Research xxx (2016) xxx-xxx
Fig. 4. Response to 24 min 3R4F cigarette smoke in the presence of S9 metabolic activation. [A] TA98 control responses. [B] TA98 3R4F aerosol treatments compared to air control. [C] TA100 control responses. [D] TA100 3R4F aerosol treatments compared to air control. UTC = untreated control, Air=air control, B[a]P = Benzo[a]pyrene [0.8 |xg/plate], AAN = 2-aminoanthracene [0.4|xg/plate].
mediary steps that may transform the test matrix. Aerosols were generated, diluted and delivered to the cell system in a relatively short space of time (~8s) [27]. In strains TA98 andTA100, cigarette smoke has been widely demonstrated to produce a positive mutagenic response [9-12]. In this study, a positive response in both strains was observed within a 24 min cigarette smoke exposure. Within this timeframe, at the highest smoke concentration only, clear thinning of the background lawn was observed, indicating toxicity. E-cigarette exposures were extended to 180 min in the presence and absence of metabolic activation and no mutagenic activity was observed in either strain under any of the test conditions. Assessing the e-cigarette aerosol is of particular importance as it captures the aerosolisation process of the e-liquid formulation, without trapping or requiring solvent extraction techniques. Estimates on coil heating performance suggests coil temperature can be as high as 280° C and in the absence of e-liquid significantly higher [28-30]. Aerosolising e-liquid introduces the potential for thermal breakdown of e-liquid compounds leading to formation of a wider range of chemical products in the aerosol [31].
This study has investigated varying exposure matrices in tester strains TA98 and TA100 and irrespective of exposure, no mutagenic activity has been shown in response to an e-cigarette test matrix. Whereas, in contrast, cigarette smoke TPM and whole smoke are clearly positive in both strains TA98 and TA100. When comparing e-cigarette aerosol test matrices to tobacco smoke, strains TA98 and TA100 are an obvious place to start, which is why these strains were initially selected. However, the choice of strains for e-cigarette testing needs careful consideration. Moreover, it is clear that with the lower toxicant burden of e-cigarette aerosols compared to cigarette smoke, that a more comprehensive Ames package of data should be generated, for example, the use of the standard Ames OECD-
five, TA98, TA100, TA1535, TA1537 (or TA97) and E.coli (or TA102) should be employed. In addition, TA104 which is proposed to be more sensitive for the carbonyl based compounds [32] that have been reported to be found in e-cigarette aerosol under dry-wicking conditions (the condition where the wick dries and overheats in the absence of e-liquid) [33], could also prove a useful addition to the Ames testing battery. Finally, these tests could be supplemented with strains YG1024 and YG1042, which are reported to be more sensitive derivatives ofTA98 andTA100. In a comprehensive testing strategy, the additional sensitivity of these two strains over their parent strains may provide valuable insight into the mutagenic potential and mode of action of e-cigarette test matrices, where responses are low, marginal or non-existent.
This study demonstrates that at comparable exposure times and doses tested in strains TA98 and TA100, that Vype® ePen aerosol was non-mutagenic. This observation is supported by chemical analysis [34], where studies have demonstrated lower levels of chemicals and contaminants in Vype ePen compared to cigarette smoke, the same product and e-liquid formulation as used in this study. For example, a recent study by Margham et al. [34] reported the levels of e-cigarette emissions from a Vype® ePen, the same e-cigarette used in this study, were 92-99% lower than those from a 3R4F reference cigarette. Of the 150 chemicals measured, only 43 were detectable in ePen, 17 were present at levels too low to be quantified. Of the 26 aerosol constituents that could be quantified, 13 were identified as analytical contaminants and 13 were generated in whole or in part by the e-cigarette. Compounds measured included the major e-liquid constituents, nicotine, propylene glycol (PG) and glycerol (VG), recognised impurities in pharmacopoeia quality nicotine and eight species identified as thermal decomposition products of PG or VG [34].
ашИЙНЯВИВ^И ARTICLE IN PRESS
D. Thorne et al. / Mutation Research xxx (2016) xxx-xxx
Fig. 5. Response to 3 hr e-cigarette aerosol in the presence of S9 metabolic activation. [A] TA98 control responses. [B] TA98 responses to Vype® ePen e-cigarette aerosol. [C] TA100 control responses. [D] TA100 responses to Vype® ePen e-cigarette aerosol. UTC = untreated control, Air=air control, B[a]P = Benzo[a]pyrene [0.8 |xg/plate], AAN = 2-aminoanthracene [0.4 |xg/plate].
The results reported in this study have extended e-cigarette aerosol exposures to 3 h in whole aerosol, whilst conducting a comparable OECD test protocol for particulate material, with consistent negative findings. The authors propose that in order to assess e-cigarettes to their full potential and understand the nature of the product in its extremes, that longer and more extensive exposure options are investigated. However, any such study should be con-textualised against human consumption data, to ensure that these products are being tested within the constraints of their manufacturers' specifications and intended use. Furthermore, in vitro dosimetry approaches should be considered to draw more accurate comparisons between cigarette smoke, e-cigarette aerosol exposures and human use.
5. Conclusions
In the presence and absence of metabolic activation, e-cigarette ACM and aerosol were deemed non-mutagenic in tester stains TA98 and TA100, under the test conditions described previously, despite clear positive control responses. Conversely, 3R4F cigarette smoke TPM and freshly generated whole smoke were clearly positive. In the case of freshly generated cigarette smoke, a positive response in both strains was observed within 24min, whereas e-cigarette aerosols remained negative up to 3 h. This study adds data to support the growing evidence base that e-cigarettes are significantly less harmful compared to cigarettes. However, these findings are based on only two tester strains and only one e-cigarette variant. Further investigations are required on extended exposure conditions, additional tester strains and other products before it becomes clear on the mutagenic potential of e-cigarettes. This study demonstrates, compared to cigarette smoke, Vype® ePen
e-cigarette particulates and aerosols are non-mutagenic under the test conditions assessed.
Declaration of interest
The authors are employees of British American Tobacco or Cov-ance Laboratories Ltd. Covance Laboratories Ltd., Harrogate, UK, conducted all experimental work and were funded by British American Tobacco. Nicoventures Ltd., UK, is a wholly-owned subsidiary of British American Tobacco.
Author contributions
David Thorne, Ian Crooks, Clive Meredith and Marianna Gaca designed the study. Michael Hollings and Adam Seymour managed and conducted all aerosol testing. David Thorne drafted the manuscript. All authors approved the final manuscript.
Acknowledgements
The authors acknowledge Annette Dalrymple, Julie Clements and Mark Ballantyne for their scientific review and technical input.
References
[1] OECD, Bacterial Reverse Mutation Test, in: OECD guideline for the testing of chemicals, Test Guideline 471 (1997).
[2] ICH-S2R1, Guidance on Genotoxicity Testing and Data Interpretation for pharmaceuticals intended for human use (2011).
[3] D. Doolittle, C. Lee, J. Ivett, J. Mirsalis, E. Riccio, C. Rudd, G. Burger, A. Hayes, Comparative studies on the genotoxic activity of mainstream smoke condensate from cigarettes which burn or only heat tobacco, Environ. Mol. Mutagen 15 (1990) 93-105.
ШШШВШШ^Ш ARTICLE IN PRESS
D. Thorne et al. / Mutation Research xxx (2016) xxx-xxx
[4] R. Baker, E. Massey, G. Smith, An overview of the effects of tobacco ingredients on smoke chemistry and toxicity, Food Chem. Toxicol. 42S (2004) 53-83.
[5] K. Scott, J. Saul, I. Crooks, O. Camacho, D. Dillon, C. Meredith, The resolving power of in vitro genotoxicity assays for cigarette smoke particulate matter, Toxicol. In Vitro 27 (2013) 1312-1319.
[6] D. DeMarini, R. Gudi, A. Szkudlinska, M. Rao, L. Reico, M. Kehl, P. Kirby, G. Polzin, P. Richter, Genotoxicity of 10 cigarette smoke condensates in four test systems: comparisons between assays and condensates, Mutat. Res. 650 (2008) 15-29.
[7] R. Steele, V. Payne, C. Fulp, D. Rees, C. Lee, D. Doolittle, A comparison of the mutagenicity of mainstream cigarette smoke condensates from a representative sample of the U.S. cigarette market with a Kentucky reference cigarette (K1R4F), Mutat. Res. 342 (1995) 179-190.
[8] E. Roemer, F. Tewes, J. Meisgen, D. Veltel, E. Carmines, Evaluation of the potential effects of ingredients added to cigarettes, part 3: in vitro genotoxicity and cytotoxicity, Food Chem. Toxicol. 40 (2002) 105-111.
[9] M. Aufderheide, H. Gressmann, A modified Ames assay reveals the mutagenicity of native cigarette mainstream smoke and its gas vapour phase, Exp. Toxicol. Pathol. 58 (2007) 383-392.
[10] M. Aufderheide, H. Gressmann, Mutagenicity of native cigarette mainstream smoke and its gas/vapour phase by use of different tester strains and cigarettes in a modified Ames assay, Mutat. Res. 656 (2008) 82-87.
[11] J. Kilford, D. Thorne, R. Payne, A. Dalrymple, J. Clements, C. Meredith, D. Dillon, A method for assessment of the genotoxicity of mainstream cigarette-smoke by use of the bacterial reverse-mutation assay and an aerosol-based exposure system, Mutat. Res. 769 (2014) 20-28.
[12] D. Thorne, J. Kilford, M. Hollings, A. Dalrymple, M. Ballantyne, C. Meredith, D. Dillon, The mutagenic assessment of mainstream cigarette smoke using the Ames assay: a multi-strain approach, Mutat. Res. 782 (2015) 9-17.
[13] K. Farsalinos, R. Polosa, Safety evaluation and risk assessment of electronic cigarettes as tobacco cigarette substitutes: a systematic review, Ther. Adv. Drug. Saf. 5 (2014) 67-86.
[14] M. Misra, R. Leverette, B. Cooper, M. Bennett, S. Brown, Comparative in vitro toxicity profile of electronic and tobacco cigarettes, smokeless tobacco and nicotine replacement therapy products: e-liquids, extracts and collected aerosols, Environ. Res. (11 2014) 11325-11347.
[15] J. Murphy, F. Lowe, I. Fearon, O. Camacho, E. Minet, A framework forthe assessment of reduced-risk tobacco and nicotine products, Conference Proceedings, Tobacco Science Research Conference (2015).
[16] J. Wan, M.Johnson,J. Schilz, M. Djordjevic,J. Rice, P. Shields, Evaluation of in vitro assays for assessing the toxicity of cigarette smoke and smokeless tobacco, Cancer Epidemiol. Biomark. Prev. 18 (2009) 3263-3304.
[17] E. Zeiger, Carcinogenicity of mutagens: predictive capability of the salmonella mutagenesis assay for rodent carcinogenicity, Cancer Res. 47 (1987) 1287-1296.
[18] International Organization for Standardization, 1999. Tobacco and tobacco products—Atmosphere for conditioning and testing. ISO 3402:1999. Geneva : International Organization for Standardization.
[19] D. Maron, B. Ames, Revised methods for the Salmonella mutagenicity test, Mutat. Res. 113 (1983) 173-215.
[20] F. De Serres, M. Shelby, Recommendations on data production and analysis using the Salmonella/microsome mutagenicity assay, Mutat. Res. 64 (1979) 159-165.
[21] Health Canada Official Method T-115. Determination of Tar, Nicotine and Carbon Monoxide in Mainstream Tobacco Smoke. Prepared by the Department of Health, December31,1999.
[22] CORESTA Recommended Method No 81, 2015. Routine analytical machine for e-cigarette aerosol generation and collection—Definitions and standard conditions. CRM No 81, 2015, http://www.coresta.org/Recommended_ Methods/CRM_81.pdf, (Accessed 4th November 2015).
[23] B. Ames, J. McCann, E. Yamasaki, Methods for detecting carcinogens and mutagens with the Salmonella/mammalian-microsome mutagenicity test, Mutat. Res. 31 (1975) 347-364.
[24] D. Thorne, J. Kilford, R. Payne, J. Adamson, K. Scott, A. Dalrymple, C. Meredith, D. Dillon, Characterisation of a Vitrocell® VC 10 in vitro smoke exposure system using dose tools and biological analysis, Chem. Cent. J. 7 (2013) 146.
[25] J. Adamson, D. Thorne, G. Errington, W. Fields, X. Li, R. Payne, T. Krebs, A. Dalrymple, K. Fowler, D. Dillon, F. Xie, C. Meredith, An inter-machine comparison of tobacco smoke particle deposition in vitro from six independent smoke exposure systems, Toxicol. In Vitro 28 (2014) 1320-1328.
[26] S. Majeed, S. Frentzel, S. Wagner, D. Kuehn, P. Leroy, P. Guy, A. Knorr, J. Hoeng, M. Peitsch, Characterization of the Vitrocell® 24/48 in vitro aerosol exposure system using mainstream cigarette smoke, Chem. Cent. J. 8 (2014) 62.
[27] D. Thorne, J. Adamson, A review of in vitro exposure systems, Exp. Tox. Path 65(2013)1183-1193.
[28] J. Bertholon, M. Becquemin, M. Roy, F. Roy, D. Ledur, M. Annesi, B. Dautzenberg, Comparison of the aerosol produced by electronic cigarettes with conventional cigarettes and the shisha, Rev. Mal. Respir. 30 (2013) 752-757.
[29] S. Talih, Z. Balhas, T. Eissenberg, R. Salman N. Karaoghlanian, R. Baalbaki, N. Saliba, A. Shihadeh, Effects of user puff topography device voltage, and liquid nicotine concentration on electronic cigarette nicotine yield: measurements and model predictions, Nicotine Tob. Res. 17 (2015) 150-157.
[30] T. Schripp, D. Markewitz, E. Uhde, T. Salthammer, Does e-cigarette consumption cause passive vaping? Indoor Air 23 (2013) 25-31.
[31] S. Costigan, C. Meredith, An approach to ingredient screening and toxicological risk assessment of flavours in e-liquids, Regul. Toxicol. Pharmacol. 72 (2015) 361-369.
[32] D. Dillon, R. Combes, E. Zeiger, The effectiveness of Salmonella strains TA100: TA102 and TA104 for detecting mutagenicity of some aldehydes and peroxides, Mutagenesis 13 (1998) 19-26.
[33] K. Farsalinos, V. Voudris, K. Poulas, E-cigarettes generate high levels of aldehydes only in 'dry puff conditions, Addiction 8 (2015) 1352-1356.
[34] J. Margham, K. McAdam, M. Forster, C. Liu, C. Wright, D. Mariner, C. Proctor, Chemical composition of an e-cigarette aerosol—a quantitative comparison with cigarette smoke, Chem. Res. Toxicol. (2016), http://dx.doi.org/10.1021/ acs.chemrestox.6b00188.
