Inter-Laboratory Reproducibility and Interchangeability of 3R4F and 1R6F Reference Cigarettes in Mainstream Smoke Chemical Analysis and In Vitro Toxicity Assays

The Ames assay was performed according to the Organization for Economic Cooperation and Development (OECD) guideline 471 (37). Salmonella typhimurium strains TA98, TA100, TA1537, TA1535, and TA102 were provided by the National Institute of Health Sciences (Kanagawa, Japan). The TPM samples were tested using the five tester strains with and without metabolic activation by S9 mix (Ieda Trading, Tokyo, Japan) containing rat liver S9 induced by phenobarbital and 5,6-benzoflavone. Eight doses of up to 1000 μg/plate were tested for each TPM. DMSO was used as a solvent control for all conditions. The frozen stock of bacterial culture was inoculated into a flask containing Oxoid Nutrient Broth No. 2 (Thermo Fisher Scientific, Waltham, MA, USA). The bacterial culture was incubated for 9 h at 37 °C on an orbital shaker. A reaction mixture containing 100 μL of bacterial suspension (> 109 cells/mL), 100 μL of TPM, and 500 μL of the S9 mix was incubated in a test tube at 37 °C for 20 min. Following the incubation, 2 mL of molten top agar supplemented with histidine and biotin was added to the reaction mixture. Then, the mixture was poured on to a minimal-glucose agar plate. The S9 mix was replaced with phosphate buffer to determine the mutagenicity without metabolic activation. Plates were incubated at 37 °C for 72 h for TA102, and for 48 h for the other strains, TA98, TA100, TA1535, and TA1537. The number of revertant colonies was counted with an automatic colony counter (System Science, Tokyo, Japan). The mutagenic activity of each TPM obtained in the Ames assay was evaluated based on the slope values of the linear dose-response curves using linear regression analysis.

The in vitro MN assay was performed in principle according to OECD guideline 487 (2014) (38) using the Chinese hamster lung cell line (CHL/IU). CHL/IU was obtained from the National Institutes of Biomedical Innovation, Health and Nutrition (Osaka, Japan). The TPM samples were assayed using three treatment conditions: short-term exposure (“Short”), with or without S9 metabolic activation by the S9 mix and long-term exposure (“Long”) without metabolic activation. Prior to exposure to the TPM, cells were cultured at 2 × 10³ cells / 200 μL/well in minimum essential medium (MEM) (Thermo Fisher Scientific) supplemented with 10% bovine serum (BS) (Thermo Fisher Scientific) using a 96-well collagen-I coated plate (Thermo Fisher Scientific) at 37 °C in a 5% CO₂ atmosphere for 24 h. Ten doses up to 600 μg/mL were tested for each TPM and DMSO was used as a solvent control. In “Short”, with and without S9, the pre-cultured cells were treated with the medium including the TPM for 3 h. All samples were added to the medium at a final concentration of 2% DMSO. In “Short” with S9, and “Short” without S9, MEM supplemented with 10% BS and MEM supplemented with 1% BS and 5% S9 mix (Ieda Trading) were used, respectively. After removal of the supernatant including the TPM, the cells were further incubated in fresh MEM supplemented with 10% BS for 21 h. In “Long” without S9, the cells were treated with MEM supplemented with 10% BS, including each TPM for 24 h. All samples were added to the medium at a final concentration of 1% DMSO. After exposure, cells were stained with Cell mask orange solution (Thermo Fisher Scientific) for the plasma membrane, Cell stain Hoechst 33342 solution (Dojindo Laboratories, Tokyo, Japan) for the nucleus, and SYTOX Green Nucleic Stain (Thermo Fisher Scientific) for dead cells. Then, cells were fixed with 4% paraformaldehyde for more than 5 min. All plates were scanned for cell and nucleus images using a CellInsight CX5 (Thermo Fisher Scientific). The number of micronucleated cells per 2 × 10³ live cells per well was analyzed and the MN cell frequency (% MN) was calculated.

Logistic regression analysis was performed with data up to the dose at which the relative population doubling (38) was greater than 40%. The slope parameter of the logistic function was defined as the measure for genotoxic activity.

The NRU assay was performed using the Chinese hamster ovary cell line (CHO-K1) in general accordance with Health Canada Official Method T-502 (39). Ten doses up to 200 μg/mL (TPM) and 200 μg TPM equiv./mL (GVP) were tested for each sample. DMSO and PBS were used as the solvent controls for the TPM and GVP samples, respectively.

The cells were maintained with Ham's F-12 nutrient mixture medium (Thermo Fisher Scientific) supplemented with 10% fetal bovine serum (FBS) (Sigma-Aldrich, St. Louis, MO, USA) in a 5% CO2 incubator at 37 °C. The cell suspension (5 × 10⁴ cells/mL) was pre-incubated for 24 h in 96-well plates (Sumitomo Bakelite, Tokyo, Japan). The cells were exposed to each smoking fraction for 24 h. After removal of the supernatant including the TPM or GVP, the cells were further incubated in serum-free medium containing 18 μg/mL neutral red dye for 3 h. The cells were then fixed with 1% formalin solution for 2 min. After removal of the fixative, the neutral red dye taken up by the viable cells was extracted by adding 50% ethanol containing 1% acetic acid solution, and the absorbance at 540 nm was measured using a microplate reader (CORONA ELECTRIC, Ibaraki, Japan). The absorbance value at each dose level was normalized to the solvent control.

Non-linear regression analysis for the relationship between the normalized absorbance and concentration was performed based on the least squares method using a logistic function. The half maximal inhibitory concentration (IC₅₀) values were calculated by inverse estimation of the effective concentration at which the normalized absorbance was reduced by 50% as the indicator of toxicity.

Immortalized normal human bronchial epithelial BEAS-2B cells, obtained from the American Type Culture Collection (Manassas, VA, USA), were cultured in Dulbecco's MEM (Thermo Fisher Scientific) supplemented with 10% FBS (MP Biomedicals, Santa Ana, CA, USA) in a 5% CO₂ incubator at 37 °C. The cells were seeded at 5 × 10³ cells/well in a 96-well plate (Corning, NY, USA) followed by pre-incubation for 24 h. Then, the cells were exposed to the medium containing each TPM up to 400 μg/mL (1% DMSO) in a 5% CO₂ incubator at 37 °C.

Cell viability was determined using a CellTiter-Fluor cell viability assay kit (Promega, Madison, WI, USA) according to the manufacturer's protocol. Briefly, cells were exposed to each concentration of TPM in a 96-well plate (Corning) for 24 h, and the constitutive protease activity within living cells as cell viability was measured with an Infinite 200 PRO microplate reader (Tecan, Männedorf, Switzerland). Tween (3%) was used as a positive control to induce complete cell death. The value of the positive control was deducted from the original values at each dose level, then obtained values were normalized to the concurrent solvent control. The IC₅₀ values were calculated as for the NRU assay.

The GSH/GSSG ratio was determined using a GSH/GSSG-Glo assay kit (Promega) according to the manufacturer's instructions. Briefly, after exposure to each TPM for 2 h, cells were lysed with GSH and GSSG lysis buffer to measure total glutathione and GSSG, respectively. Subsequently, enzyme and substrate were added, and then luciferase activity was measured using an Infinite 200 PRO microplate reader. Then, the GSH/GSSG ratio was calculated by interpolation of glutathione concentrations from standard curves. The values at each dose level were normalized to the solvent control. The IC₅₀ values were calculated as for the NRU assay.

BEAS-2B cells were transfected with the luciferase gene as the reporter under the transcriptional regulation of ARE (18). The ARE-luciferase reporter assay was carried out using a Luciferase Assay System (Promega) following the manufacturer's instructions. Briefly, cells were lysed after exposure to each TPM for 24 h in a 96-well plate (Corning). Then, the substrate was added, and luciferase activity was measured using an Infinite 200 PRO micro-plate reader to determine the ARE gene activity. The values at each dose level were normalized to the solvent control. The ARE gene activity for each TPM was evaluated based on the slope values calculated by linear regression analysis over the dose range where cell viability was higher than 50%.

The mutagenicity of each TPM was investigated using TA98, TA100, TA1535, TA1537, and TA102, with or without S9 metabolic activation. In the current study, the 1R6F and 3R4F TPM elicited dose-dependent increases in the number of revertants in TA98 with S9, TA100 with and without S9, and TA1537 with S9, regardless of the smok ing regimen (Supplementary Table A2). Consequently, the slope values were calculated as the mutagenic activities in these conditions. In the remaining strains and conditions, both the 1R6F and 3R4F TPM showed no reproducible increases in the number of revertants.

To examine the interchangeability of 1R6F and 3R4F regarding the mutagenicity, the relative differences in the slope values between 1R6F and 3R4F in the current study were analyzed (circles in Figure 3A and B for the ISO standard and intense smoking regimens, respectively). Then, these values were compared with the CD range calculated based on historical variations in 3R4F (boxes in Figure 3A and B). The results indicated that the relative differences were within the range of the CD in TA98 with S9, TA100 with and without S9, and TA1537 with S9, regardless of the smoking regimen (“empty” circles in Figure 3A and B).

Figure 3

The inter-laboratory reproducibility was then examined by comparing the results obtained in the current study (circles in Figure 3A and B) and Jaccard study (triangles in Figure 3A and B). With S9, the relative differences in the mutagenicity were within the range of the CD in TA98, TA100, and TA1537, regardless of the smoking regimen in both the studies (“empty” circles and triangles in Figure 3A and B). Without S9, the mutagenicity of the TPM from both the reference cigarettes was found in TA98 only in the Jaccard study and in TA100 only in the current study, regardless of the smoking regimen (Figure 3A and B). Therefore, the results for the conditions with S9 were reproducible although there were some inter-laboratory differences in the mutagenic response in TA98 and TA100 without S9.

The in vitro MN assay was conducted using CHL/IU cells under three treatment conditions: “Short”, with or without S9, and “Long” without S9. The 1R6F and 3R4F TPM induced dose-related increases in the MN frequency under all the tested conditions. The slope parameter of the logistic function calculated as genotoxic activity for each condition are shown in Supplementary Table A3.

To compare the genotoxicity between 1R6F and 3R4F in the current study, the relative differences in the slope parameter of the logistic function of 1R6F compared with 3R4F were calculated (circles in Figure 4A and B for the ISO standard and intense smoking regimens, respectively), then compared with the CD range calculated based on historical variations in 3R4F (boxes in Figure 4A and B).

Figure 4

The relative differences were within the range of the CD in “Short” with S9, and “Long” without S9 under the ISO standard smoking regimen (“empty” circles in Figure 4A) and in all conditions under the ISO intense smoking regimen (“empty” circles in Figure 4B), while the relative difference was out of the range of the CD in “Short” without S9 under the ISO standard smoking regimen (“filled” circles in Figure 4A). Inter-laboratory reproducibility was then analyzed by a comparison of the results obtained in the current study (circles in Figure 4A and B) and Jaccard study (triangles in Figure 4A and B). The results from both the studies showed a good agreement. In “Short” without S9 under the ISO standard smoking regimen, the relative genotoxic activity of 1R6F was higher than the upper limit of the CD range (filled circles and triangles in Figure 4A and B). For the other conditions, the relative differences were within the range of the CD (empty circles and triangles in Figure 4A and B). Therefore, inter-laboratory reproducibility in the MN assay was confirmed.

Both the TPM and GVP collected from 3R4F and 1R6F elicited dose-dependent decreases in the absorbance, regardless of the smoking regimen. The IC₅₀ values for the reference cigarettes are shown in Supplementary Table A4. For a comparison of the cytotoxicity of 1R6F and 3R4F in the current study, the relative differences in the IC₅₀ values of 1R6F compared with 3R4F were calculated (circles in Figure 5A and B for the ISO standard and intense smoking regimens, respectively), then compared with the CD range calculated based on historical variations in 3R4F (boxes in Figure 5A and B).

Figure 5

The relative difference was out of the CD range for the TPM under the ISO intense smoking regimen (filled circles in Figure 5B), while the relative differences in the remaining conditions were within the range of the CDs (empty circles in Figure 5A and B).

We further analyzed the inter-laboratory reproducibility by comparing the results obtained in the current study (circles in Figure 5A and B) and Jaccard study (triangles in Figure 5A and B).

For the TPM under the ISO intense smoking regimen, the trend in the relative difference in the IC₅₀ values was different from the Jaccard study within ± 10% (Figure 5B).

Under the remaining conditions, the results in the current and Jaccard studies were consistent.

We carried out some additional in vitro toxicological assays to investigate oxidative stress with BEAS-2B, a cell line derived from normal human bronchial epithelium. Cell viability was determined following exposure to the TPM collected from 3R4F and 1R6F using both ISO standard and intense smoking regimens. All the TPM exposures induced dose-dependent decreases in cell viability (Figure 6). No significant difference was observed in the IC₅₀ values between 3R4F and 1R6F under either smoking regimen (Table 3).

Figure 6

The GSH/GSSG ratio and ARE gene activity were used as indicators of oxidative stress, and changes in these parameters following exposure to the TPM of 3R4F or 1R6F were evaluated under the ISO standard and intense smoking regimens. All the TPM exposures induced reduction of the GSH/GSSG ratio in a dose-dependent manner (Figure 7). A significant difference in the IC₅₀ values between 1R6F and 3R4F was not observed, regardless of the smoking regimen (Table 4).

Figure 7

Exposure to the TPM resulted in significant increases in ARE gene activity regardless of the condition (Figure 8). The decline in luciferase activity observed at a high dose level was possibly because of cytotoxicity. Although 1R6F tended to show higher activity than 3R4F in the ARE-luciferase reporter assay, no significant difference was observed between the slope activities of 3R4F and 1R6F under both smoking regimens, as calculated using Student's t-test (Table 5).

Figure 8