Dual Exposure to E-Cigarette Vapour and Cigarette Smoke Results in Poorer Airway Cell, Monocyte, and Macrophage Function Than Single Exposure

E-cigarette users predominantly also continue to smoke cigarettes. These Dual Users either consume e-cigarettes in locations where smoking is not allowed, but vaping is, or to reduce their consumption of cigarettes, believing it will lead to harm reduction. Whilst it is known that e-cigarette vapour is chemically less complex than cigarette smoke, it has a distinct chemical profile, and very little is known about the health impacts of exposure to both chemical profiles vs. either alone. We simultaneously exposed cells in vitro to non-toxic levels of e-cigarette vapour extract (EVE) and cigarette smoke extract (CSE) to determine their effects on 16HBE14o- airway epithelial cell metabolism and inflammatory response, as well as immune cell (THP-1 cells and monocyte-derived macrophages (MDM) from healthy volunteers) migration, phagocytosis, and inflammatory response. We observed increased toxicity, reduced metabolism (a marker of proliferation) in airway epithelial cells, and reduced monocyte migration, macrophage phagocytosis, and altered chemokine production after exposure to either CSE or EVE. These cellular responses were greater after dual exposure to CSE and EVE. The airway epithelial cells from smokers showed reduced metabolism after EVE (the Switcher model) and dual CSE and EVE exposure. When EVE and CSE were allowed to interact, the chemicals were found to be altered, and new chemicals were also found compared to the CSE and EVE profiles. Dual exposure to e-cigarette vapour and cigarette smoke led to worse functional outcomes in cells compared to either single exposure alone, adding to limited data that dual use may be more dangerous than smoking only.


Introduction
E-cigarettes have been shown to cause a widening range of effects, beyond toxicity, in cells, animal models, and humans (reviewed in Bozier et al. [1]).The effects observed in the lung include enlarged human lung fibroblasts with induced spindle formation and vacuolization [2] and the loss of lung endothelial barrier function [3].Bozier et al., the National Academies of Sciences, Engineering, and Medicine review, and Banks et al. have provided excellent systematic reviews on the known health effects of e-cigarettes using in vitro, ex vivo, and in vivo studies [1,4,5].Data from previous studies, including our own, are forming a clear picture that even nicotine-free e-liquids may not be harmless, likely because of the flavourants present [1,4,[6][7][8].
One of the largest groups of e-cigarette users are those who smoke tobacco cigarettes and vape e-cigarettes, called "Dual Users" [9,10].Some of the reasons for dual use are reduced cost, the ability to use e-cigarettes in places where smoking tobacco cigarettes is

Dual Exposure of Bronchial Epithelial Cells Results in Increased Toxicity and Impaired Proliferative Capacity
We have previously reported that 10% CSE, 100% banana EVE, and 100% chocolate EVE were sufficient to cause increased 16HBE cell death when full monolayers were exposed, and that the LDH assay proved to be very high sensitivity toxicity assay [8].We first performed experiments to identify non-toxic single and dual exposure concentration in the 16HBE cells so that cell metabolism, as measured by MTT conversion as a marker of proliferative capacity, would not be confounded by the dying cells.We found that reductions to 5% CSE, 25% banana EVE, and 25% chocolate EVE were required to prevent significant toxicity, whether alone or as a dual exposure, as measured by LDH release.Increased toxicity was observed when the cells were dually exposed to 100% and 50% banana or chocolate EVE with 5% CSE vs. 5% CSE alone.Additionally, increased toxicity was observed upon dual exposure to 100% banana EVE with 5% CSE or 100% and 50% chocolate EVE with 5% CSE vs. the respective EVE alone (Figure 1A).Exposure to 100% nicotine, mango, and PGVG EVE did not induce an increased release of LDH alone or in dual combination with 5% CSE.
Using the cellular metabolism of MTT to formazan as a proxy for proliferative capacity, a key attribute for wound healing, the 16HBE cells displayed reduced proliferative capacity after exposure to tobacco EVE, dual tobacco, dual banana, chocolate, and dual chocolate, relative to control.Interestingly, individual exposure to CSE and banana was not enough to significantly reduce the proliferative capacity of the 16HBE cells compared to control; however, when combined, a significant reduction was observed compared to both the control and CSE treatments.Dual banana and dual chocolate exposures also displayed proliferative capacity significantly below to that of CSE exposure alone.

Models of Dual Use and Switching Using Smoker NHBE Show Reduction in Proliferative Capacity
As NHBE cells from smokers (SHNBE) were commercially available and able to be passaged to expand to the numbers needed for this work, they were used to model the impact of both "Dual User" and "Switcher" usage habits, whereby cigarette smokers either transition to dual use with e-cigarettes or switch to exclusive e-cigarette use.

Models of Dual Use and Switching Using Smoker NHBE Show Reduction in Proliferative Capacity
As NHBE cells from smokers (SHNBE) were commercially available and able to be passaged to expand to the numbers needed for this work, they were used to model the impact of both "Dual User" and "Switcher" usage habits, whereby cigarette smokers either transition to dual use with e-cigarettes or switch to exclusive e-cigarette use.
In the Dual-User model, exposure to 5% CSE showed no significant decrease in proliferative capacity compared to control, likely due to the exposure of the cells to cigarette smoke prior to collection.Exposure to dual tobacco, dual mango, dual banana, and dual PG:VG significantly decreased the proliferative capacity of SNHBE cells below that of both CSE-only and their matching EVE-only exposures.Dual chocolate exposure had a significantly decreased proliferative capacity compared to chocolate EVE-only but was not significantly less than the CSE-only exposure (Figure 2).In the Dual-User model, exposure to 5% CSE showed no significant decrease in proliferative capacity compared to control, likely due to the exposure of the cells to cigarette smoke prior to collection.Exposure to dual tobacco, dual mango, dual banana, and dual PG:VG significantly decreased the proliferative capacity of SNHBE cells below that of both CSE-only and their matching EVE-only exposures.Dual chocolate exposure had a significantly decreased proliferative capacity compared to chocolate EVE-only but was not significantly less than the CSE-only exposure (Figure 2).

Models of Dual Use and Switching Using Smoker NHBE Show Reduction in Proliferative Capacity
As NHBE cells from smokers (SHNBE) were commercially available and able to be passaged to expand to the numbers needed for this work, they were used to model the impact of both "Dual User" and "Switcher" usage habits, whereby cigarette smokers either transition to dual use with e-cigarettes or switch to exclusive e-cigarette use.
In the Dual-User model, exposure to 5% CSE showed no significant decrease in proliferative capacity compared to control, likely due to the exposure of the cells to cigarette smoke prior to collection.Exposure to dual tobacco, dual mango, dual banana, and dual PG:VG significantly decreased the proliferative capacity of SNHBE cells below that of both CSE-only and their matching EVE-only exposures.Dual chocolate exposure had a significantly decreased proliferative capacity compared to chocolate EVE-only but was not significantly less than the CSE-only exposure (Figure 2).In the Switcher model, exposure to tobacco EVE or mango EVE significantly reduced proliferative capacity below control.However, this is likely due to the fact that PG:VG alone caused a significant reduction in proliferative capacity compared to control, which persisted even at a 25% PG:VG EVE concentration.Interestingly, exposure to nicotine EVE, banana EVE, and chocolate EVE had no significant reduction compared to control, despite the presence of PG:VG.Thus, comparisons were further assessed against the matching PG:VG EVE exposure or dual PG:VE + CSE exposure (Table 1).In the Dual-User model, the proliferative potential was significantly decreased upon exposure to dual nicotine, dual tobacco or dual mango compared to dual PGVG, whilst in the Switcher model, the proliferative potential was significantly higher in nicotine, banana, and chocolate EVE-exposed cells when compared against exposure to the e-liquid base: PGVG EVE.

Dual Exposure Induces Decreased Migration and Spontaneous Differentiation to Macrophages in THP-1 Monocytes
Experiments to identify non-toxic single and dual exposure conditions for THP-1 monocytes were conducted prior to the migration assay to prevent any confounding by the presence of dead/dying cells (Figure 3A).There was no difference in toxicity noted between single and dual exposures.
We also assessed whether the exposures induced spontaneous differentiation that resulted in adherence, so that the migration assays would not be confounded by differentiating cells.Exposure to 20% CSE, 100% banana EVE, dual 100% banana, 100% chocolate EVE, and dual 100% chocolate increased THP-1 monocyte toxicity, as measured by LDH release, and also induced differentiation/adherence. Exposure to 10% CSE, 50% banana, dual 50% banana, 50% chocolate, and dual 50% chocolate was not toxic but did induce differentiation compared to control (Figure 3B).Thus, 5% CSE, 25% banana, and 25% chocolate were used for the migration assays.
Exposure to 5% CSE, nicotine EVE, and all EVE flavours alone or dual with CSE reduced THP-1 cell migration compared to control (Figure 3C).Dual nicotine and all dual flavours also reduced THP-1 cell migration below that of 5% CSE exposure.Significant reductions in migration were noted between every dual exposure and its single exposure counterpart.Interestingly, 100% PGVG also reduced migration, but 25% PGVG did not.Thus, further statistical analysis was performed against the matching PG:VG controls (Table 2).When compared against the corresponding PGVG exposure, THP-1 migration was significantly decreased by exposure to nicotine, tobacco, mango, banana, dual banana, chocolate, and dual chocolate migration.reductions in migration were noted between every dual exposure and its single exposure counterpart.Interestingly, 100% PGVG also reduced migration, but 25% PGVG did not.Thus, further statistical analysis was performed against the matching PG:VG controls (Table 2).When compared against the corresponding PGVG exposure, THP-1 migration was significantly decreased by exposure to nicotine, tobacco, mango, banana, dual banana, chocolate, and dual chocolate migration.

Dual Exposure Decreases Normal Healthy Monocyte Migration and Further Exacerbates Decreased Phagocytic Capacity
Exposure to CSE, nicotine, and all individual EVE flavours alone or dually with CSE reduced donor monocyte migration compared to control.Dual nicotine and all dual exposure treatments also reduced migration compared with CSE alone.When comparing single vs. dual exposure matching pairs, Nicotine EVE and all flavours EVE showed a significant decrease with dual exposure vs. single EVE exposure (Figure 4A).As observed in THP-1 monocytes, 100% PGVG also reduced migration, while 25% PGVG did not.Therefore, further statistical analysis against the corresponding PG:VG controls was performed (Table 3).When compared with their matching PGVG control; dual nicotine, dual tobacco, dual mango, banana, dual banana, chocolate, and dual chocolate showed further decreased migration.This was also visualised via microscopy (some representative images are shown in Figure 4C).Cytochalasin D was used as a negative control for migration.We have previously reported that CSE nicotine and flavoured EVE could reduce the alveolar macrophage phagocytosis of bacteria [7].Healthy donor monocyte-derived alveolar-like macrophages were used as a model for alveolar macrophages due to the numbers required for dual exposure analysis.CSE, nicotine, dual nicotine, and all single and dual flavours reduced phagocytosis compared to control, as expected.PGVG did not affect phagocytosis.Dual nicotine and all dual-flavoured EVEs reduced phagocytosis compared to CSE alone.When comparing single vs. dual exposures, nicotine and all flavoured EVEs showed a significant reduction in phagocytosis in the dual exposure vs. the single exposure (Figure 4B).

Dual Exposure Resulted in Greater Disruption of Macrophage Cytokine Production Than Single Exposure
Exposure to CSE, nicotine EVE, and all flavoured EVEs reduced the secretion of MCP-1 and RANTES, while increasing secretion of IL-8 from MDMs compared to the control exposure.All flavours in dual EVE exposure further enhanced this change compared to CSE alone, whilst dual nicotine-only enhanced this change compared to CSE for RANTEs and IL-8, but not for MCP-1.PGVG had no effect on MPC-1, RANTES, or IL-8 secretion.Dual nicotine and all EVE flavours in dual EVE exposure showed a further significant deviation from their corresponding single EVE exposure for all three cytokines (Figure 5).

Interaction between Cigarette Smoke and E-Cigarette Vapour in the Soluble Phase Results in Changes in VOCs
One possible reason for the increased effects of dual exposures vs. single exposure may be due to the interaction of cigarette smoke with e-cigarette vapour.We modified our existing extract production method to recapitulate, in a soluble phase, what would happen in the lung lining fluid of a Dual User.The preparation of 100% CSE was first made, and then, immediately after, either 100% chocolate or 100% PGVG e-cigarette vapour was bubbled through the CSE.The reverse procedure was also performed.Samples were either frozen immediately or incubated at 37 • C with 5% CO 2 for 24 h to allow for longer

Interaction between Cigarette Smoke and E-Cigarette Vapour in the Soluble Phase Results in Changes in VOCs
One possible reason for the increased effects of dual exposures vs. single exposure may be due to the interaction of cigarette smoke with e-cigarette vapour.We modified our existing extract production method to recapitulate, in a soluble phase, what would happen in the lung lining fluid of a Dual User.The preparation of 100% CSE was first made, and then, immediately after, either 100% chocolate or 100% PGVG e-cigarette vapour was bubbled through the CSE.The reverse procedure was also performed.Samples were either frozen immediately or incubated at 37 °C with 5% CO2 for 24 h to allow for longer interaction (aged).Dual extracts were then VOC profiled and compared against 100% CSE, 100% chocolate EVE, and 100% PGVG EVE.
Firstly, we observed that the PG:VG sample, which was made from PG and VG stocks purchased from a vape store, had chemicals present that were not present in the chocolate sample, therefore it was not suitable for comparing with the chocolate single-extract samples (Table S1).
In the extracts frozen immediately after collection, we observed aniline and 2,2′-Bipyrazine in the CSE made into the chocolate EVE, and 1-ethoxy-2-methylbenzene, 3-pyridinamine, phosphoryl fluoride, and N-methyl-benzeneacetamide in the chocolate EVE made into CSE, which were not present in CS or chocolate single extracts.We further observed 2-Pentanone, acetic acid ethenyl ester, methylpyrazine, 2-ethenyl-pyridine, 3-methyl-2-cyclopenten-1-one, 1-methoxy-4-methyl-benzene, 2-oxo-3-cyclopentene-1-acetaldehyde, and 3-methylpyridine in CS or chocolate single extracts that were not detected in Firstly, we observed that the PG:VG sample, which was made from PG and VG stocks purchased from a vape store, had chemicals present that were not present in the chocolate sample, therefore it was not suitable for comparing with the chocolate single-extract samples (Table S1).
In the extracts frozen immediately after collection, we observed aniline and 2,2 ′ -Bipyrazine in the CSE made into the chocolate EVE, and 1-ethoxy-2-methylbenzene, 3-pyridinamine, phosphoryl fluoride, and N-methyl-benzeneacetamide in the chocolate EVE made into CSE, which were not present in CS or chocolate single extracts.We further observed 2-Pentanone, acetic acid ethenyl ester, methylpyrazine, 2-ethenyl-pyridine, 3-methyl-2-cyclopenten-1-one, 1-methoxy-4-methyl-benzene, 2-oxo-3-cyclopentene-1-acetaldehyde, and 3-methylpyridine in CS or chocolate single extracts that were not detected in the CSE made into the chocolate EVE.Additionally, 2-Pentanone, acetic acid ethenyl ester, methylpyrazine, 2-ethenylpyridine, 3-methyl-2-cyclopenten-1-one, 1-methoxy-4-methyl-benzene, cyclopentanone, 2,2dimethyl-propanal, allyl pentyl ester oxalic acid, 4-ethylphenol, and 2,5,6,7-tetrahydro-3H-Cyclopenta[c]pyridazin-3-one/3-methylene-2-oxo-cyclohexanecarboxylic acid, methyl ester were found in CS or chocolate single extracts that were not detected in the chocolate EVE made into the CSE (Table 4).This demonstrates the rapid generation of new VOCs and the likely alteration/degradation of VOCs due to the interaction of the chemicals in CSE with chocolate EVE.In the extracts allowed to further interact at 37 • C with 5% CO 2 for 24 h after collection, we observed pyrrole, 3-methyl-2-cyclopenten-1-one, and cyclopentanone in the CSE made into the chocolate EVE and pyrrole and 3-methyl-2-cyclopenten-1-one in the chocolate EVE made into CSE, which were not present in CS or chocolate single extracts.We further observed methylpyrazine, methyl alcohol, methylglyoxal, and 3-pyridinol, in CS or chocolate single extracts that were not detected in the CSE made into the chocolate EVE.In addition, methylpyrazine, methyl alcohol, methylglyoxal, and 3,4-dimethyl-2-cyclopenten-1-one/2,3-dimethyl-2-cyclopenten-1-one were found in CS or chocolate single extracts that were not detected in chocolate EVE made into the CSE (Table 5).This demonstrated that the interaction of the chemicals in CSE and chocolate EVE over time at physiological temperatures results in the generation of new VOCs and the alteration of VOCs, which was mostly different from the changes immediately after the interaction of the two extracts.The changes between PGVG EVE and interacting CSE and PGVG EVE are documented in Supplemental Tables S1 and S2.
When we compared freshly collected extracts vs. aged extracts, we also observed the detection of new VOCs, as well the disappearance of many others, likely due to alterations such as thermal decomposition or degradation (Tables S3-S9).

Discussion
Whilst e-cigarettes have only been mainstream for about 10-15 years, there has been a recent surge in promoting them as a form of harm reduction, resulting in a large number of smokers who "Switch" to e-cigarette use, 30-40% of whom continue to use e-cigarettes long term ≥ 1 yrs [21,22], and those who "Dual use", both smoking and vaping [9,10].
Whilst there is no dispute about the harms of smoking cigarettes, and data continue to mount against vaping being harm-free [23], including a growing body of evidence suggesting that the harms of vaping extend beyond the lungs [24][25][26][27][28], we still understand very little about dual use.Dual use exposes users to two different sets of chemicals, which potentially activate different molecular pathways, leading to increased harm, a very different scenario to the expected harm reduction many users believe they are achieving via dual use.
We and others have previously shown that flavouring chemicals, in particular, contribute to e-cigarettes' negative health impact [6][7][8]29].Thus, in this study, we investigated the impacts of dual exposure to four different flavoured e-liquids, with and without added cigarette smoke, including: two with previously observed high cellular impact and two observed to have lower cellular impacts [7,8].
We have previously shown that the detection of LDH release was a more sensitive indicator of bronchial epithelial cell toxicity than annexin V or nucleic acid staining [6,7], so this parameter was utilised to determine the non-toxic doses of EVE and CSE.Toxicity to bronchial epithelial cells was assessed by the loss of 16HBE membrane integrity and subsequent LDH release, which showed that dual exposure to two out of four flavoured EVEs and CSE was worse than exposure to CSE or to flavoured EVE alone.Whilst toxicity due to CSE or EVE was expected, including flavour-linked toxicity [30][31][32], no studies have yet investigated their combined impact on airway cell survival.This suggests that dual use may lead to higher toxicity in the lungs than smoking.
Damage to the airways usually triggers increased cellular metabolism, leading to wound healing through cell proliferation, needed to replace damaged cells, a process which is hindered by cigarette smoke [33,34].We investigated the cellular metabolism of MTT in single vs. dual exposures in 16HBE cells.We observed a decreased MTT conversion after exposure to two out of four flavours.However, one of the non-significant flavours was quite diluted to reduce it to a non-toxic level, so its impact at higher doses could not be assessed in this assay.Only one study has studied the effects of e-cigarettes on wound healing, and this study observed a negative impact on the gingival cells [35].However, in this study, for two out of four flavours, dual exposure had a greater impact compared to CSE exposure.
Commercially available smoker NHBE cells allowed us to validate our results using both a Dual-User model-smokers who transition to both smoking and vaping, as well as a Switcher model-smokers who switch exclusively to e-cigarettes.Smoker bronchial epithelial cells showed reduced proliferative capacity in single exposure to three out of four flavours of EVE, whereas the fourth flavour was diluted to a non-toxic level, so its impact at higher concentrations could not be assessed in this assay.We also observed that dual exposure to CSE and EVE was greater than exposure to CSE for three out of four flavours.Interestingly, we also observed that the PGVG EVE alone also reduced the cellular proliferative potential, albeit to a lower extent than flavoured EVE, providing evidence that all e-liquids are likely to cause a negative impact, which can be exacerbated by the addition of flavours to varying extents.This study therefore provides the first early evidence that Dual Users may expose their airways to a higher toxicity risk.
In order to perform their primary function, monocytes need to be able to migrate to the site of infection and then differentiate into macrophages, so we assessed the impacts of CSE, EVE, and dual exposure on monocyte migration.First, we needed to confirm whether dual exposure caused increased cellular toxicity, so that we could be sure that cell death was not contributing to any declines.We did not observe any increased toxicity from dual exposure.However, the chocolate and banana flavours needed dilution to prevent toxicity.We then assessed whether EVE, CSE, or dual exposure affected monocyte differentiation and/or adhesion to ensure this was not reducing cell numbers.We found that CSE and two flavours needed diluting to avoid inducing differentiation, and thus adhesion to the plates, and falsely impacting the results.All four flavours and nicotine alone reduced monocyte migration.Dual exposure further reduced migration, which was also significantly reduced compared to CSE alone.Interestingly, PGVG-only EVE impacted THP-1 migration, possibly due to the coating of the cells with a viscous substance.Whilst there is little information on the effects of e-cigarettes on monocyte migration, a recent paper showed that THP-1 monocytes exposed to DHA had reduced migration [36].We then confirmed these data in healthy non-smoker donor monocytes, with the same results.The existing literature shows conflicting information on the effects of CSE on monocyte migration, with some studies showing increased migration and others decreased migration, which may be due to methods of extraction, cell type, dose, or even different migration assay designs [37][38][39][40][41][42][43][44][45].
There are not any other studies yet investigating the impacts of e-cigarettes on monocyte migration, let alone dual exposure.
Once at the site of an infection, monocytes can differentiate into macrophages and phagocytose bacteria.Studies by us and others have shown that exposing immune cells to e-cigarette vapour causes reduced bacterial phagocytosis [6][7][8]23,[46][47][48][49][50][51], with support from some mouse infection studies [24,51,52], but none have looked at dual exposure.This effect was most likely due to the decreased levels of phagocytic receptors TLR-2, TLR-4, CD44, CD36, and SR-A1 we observed on the surface of the cells [6][7][8]23].This is the first study that demonstrates that exposure to both cigarette smoke and e-cigarette vapour causes a further reduction in the phagocytic capacity of macrophages.
Tissue-resident macrophages, including alveolar macrophages, act as first responders to infections and then release chemokines to recruit monocytes to the site of infection [53,54].Thus, we assessed two key chemokines secreted by MDMs with single and dual exposure.We have previously shown that the secretion of the monocyte chemokine MCP-1 by THP-1 macrophages was decreased after exposure to CSE, nicotine, and some flavoured EVEs in single exposure [8], which was also found by Sinha et al. [55].In mice, Wang et al. found increased MCP-1 and RANTES in bronchoalveolar lavage fluid (BALF) after exposure to vapour from unflavoured nicotine e-liquid [56].The subsequent paper found that the inflammatory profile changes were often gender specific [57].Moshensky et al. found no changes in the RANTES gene expression in nicotine-vaped mouse lung tissue, but only two flavours were tested [26].MCP-1 increased in the BALF of mice exposed to unflavoured nicotine vapour [58].In humans, one study showed that vapers had increased RANTES in airway cells [59], but another showed reduced RANTES in the sputum of vapers [60].Hickman et al. found no change in MCP-1 in vapers' sputum [61].Cigarette smoke has been associated with reduced MCP-1 and RANTES in several studies [62][63][64][65][66][67][68], although there are conflicting studies as well [69][70][71][72].No studies have yet looked at the effect of dual exposure on monocyte recruitment via chemokine production from macrophages.Our data show that dual exposure exacerbated the reduction in chemokine secretion.
We then assessed whether the neutrophil chemoattractant, IL-8 [73], was also reduced by single and dual exposures, but we found that this was actually increased with CSE, single EVE, and dual exposures, with greater increases for dual exposures.We have previously demonstrated increased IL-8 secretion by donor airway cells that was specific to certain flavours, but only a trend of increased IL-8 levels in saliva of vapers [7] (also seen by Perez et al. [74]), which we believe to be related to mixed flavour use in the vapers.This was supported by mixed results in the literature in studies which used a range of flavours [2,[75][76][77][78].IL-8 secretion by monocytes and macrophages, specifically, was also found to be related to certain flavours [79][80][81].The literature on the role of cigarette smoke on macrophage IL-8 secretion consistently reported a positive association, similar to our findings [64,66,[82][83][84], but, again, there are not any studies yet looking at dual exposure.
VOCs are widely recognised as toxic, with some being recognised as carcinogenic, and concentrations in workplaces are often regulated [85,86].VOCs, due to their volatile oxidative nature, lead to oxidative stress and the formation of reactive oxygen species.When inhaled, they can cause pulmonary harm [87,88].Whilst there are minimal data available in the literature on the direct impacts of VOC exposure without particulate matter, these few studies have shown it leads to the disruption of cytokine airway levels [89,90], including increased airway IL-8 [91][92][93].
VOCs in e-cigarettes have been shown to have a distinctly different VOC profile compared to cigarette smoke [19,94,95], not just a decrease in the VOCs present in cigarette smoke.This has been linked to the presence of flavourants in e-cigarettes vs. cigarettes [96].Some previous studies have even shown that e-liquid is reactive, and that new products are formed between the flavouring chemicals and the glycol bases, even at room temperature [97][98][99].Many e-liquids contain more than five different flavouring chemicals and as high as 50 [7,100].We have previously assessed 10 different flavoured e-liquids and observed a correlation between the benzene ring-containing flavourants and impacts on macrophage phagocytosis and airway epithelial cell toxicity [7].
However, there has been little investigation into what may occur when the chemicals of both cigarette smoke and e-cigarette vapour are allowed to interact, i.e., in the lung lining fluid of a Dual User.We therefore allowed the soluble chemicals from cigarette smoke and e-cigarette vapour to interact and demonstrated that the interaction between the two can indeed lead to the alteration of the VOCs present in either e-cigarette vapour or cigarette smoke, as well as the formation of new ones, providing a plausible reason why dual exposure has been linked with poorer health outcomes.These chemicals include aniline, which has been known to be toxic for almost 100 years and can be absorbed through skin or inhaled and can cause methemoglobinemia, or reduced ability of haemoglobin to carry oxygen, resulting in breathlessness [101].Phosphoryl fluoride was also detected, a toxic fluoride gas produced in lithium battery fires [102], as well as pyrrole, which is associated with hepatotoxicity [103].3,5-dimethylphenol/3.5-DMP is known to be toxic to the waterways [104], and furfural has been shown to cause lung injury [105].Whilst there is very little in the literature in this area, the study by Cirillo et al. showed that the level of carbonyls produced by e-liquids without nicotine was higher than that in those with nicotine.This was linked to the total concentration of chemicals in the e-liquid being lower and allowing more room for carbonyl creation [106].
We also observed a number of VOCs in single exposure extracts that were not present in the dual exposure, even in the samples frozen immediately after production.This suggests that the VOCs, which are by nature difficult to destroy without powerful catalytic compounds, were altered, perhaps by metabolism, oxidation, or thermal degradation.VOC by-products or intermediary VOCs formed from these processes are also often considered toxic.This may represent another possible pathway for the increased harm being observed in Dual Users.
Dual use if often considered harm reduction due to reduction in exposure to the chemicals and VOCs in cigarette smoke [107,108].However, these data provide proof-ofconcept evidence that using both tobacco cigarettes and e-cigarettes will expose users not only to the different sets of chemicals in cigarette smoke and e-cigarette vapour, but also to a likely new set of chemicals formed due to the interaction between the chemicals in cigarette smoke and e-cigarette vapour.This may be one of the reasons why increased harm is observed in dual use vs. smoking or vaping only.Further extensive investigation is warranted into the interaction of e-cigarette vapour with and without flavour or nicotine with cigarette smoke to further understand the additional harms dual use may present.
There are limitations to our study, in that we utilized in vitro or ex vivo systems, and that the airways exist as a complicated system involving many interacting cell types.We only tested four e-liquids, and there are currently thousands sold worldwide.As we used PG and VG sold by the vaping company as our PG:VG, the purity of these reagents could not be guaranteed.We only assessed the interaction of soluble chemicals.Assessment of VOCs and larger, dedicated mass spec studies are needed to further expand on our data.Due to the large number of cells needed for migration and phagocytosis assays, we were unable to assess these in Switcher and Dual-User models.We were also unable to assess impacts based on gender.

EVE and CSE Preparation
E-cigarette vapour extract (EVE) was produced using an EVOD-2 3.7 V, which uses a 1.5 Ω. dual coil, as previously described [8].EVE from four flavoured e-liquids (banana, chocolate, mango, and tobacco), in a 50% Propylene Glycol:50% Vegetable Glycerine base (PG:VG), nicotine (at 18 mg/mL in PG:VG), and PG:VG were only used in this study.These flavours were initially tested as part of a panel of 10 flavours in previous studies, due to their popularity amongst non-smokers (sweet/fruit), smokers, and ex-smokers (tobacco).These four flavours were selected from the panel of 10 for future studies, as they represented flavours with low epithelial toxicity (tobacco and mango) and high epithelial toxicity (banana and chocolate) [7].The control medium (C) was obtained by using the same apparatus to pass air through the culture medium for the same duration as e-cigarette use.Cigarette smoke extract (CSE) was prepared as previously described [8].For migration assays, 5 cigarettes (250% CSE) or 60 × 3s puffs (120% EVE) per 10 mL of serum-free media were required to make combination treatments that could be added to the cells without diluting the EVE below 100%.

Cell Maintenance
The THP-1 monocytic cell line (ATCC, Manassas, VA, USA) was maintained in RPMI with 10% FCS and differentiated into macrophages, as previously described [23].Experiments were carried out within 10 passages of each other.Cells were routinely screened for mycoplasma.
Normal human bronchial epithelial cells from healthy patients with a history of cigarette smoking (SNHBE; CC-2540, Lonza, Durham, NC, USA) were cultured as per manufacturer's instructions in basal epithelial growth medium (BEGM) and used within 3 passages.

Isolation of Monocytes and Monocyte-Derived Macrophages
Monocytes were collected from non-smokers with no history of chronic respiratory disease.Written, informed consent was obtained from all participants, and ethics approval was granted by the Royal Adelaide Hospital HREC for this research.Peripheral blood mononuclear cells (PBMCs) were isolated as previously described [109] and seeded for 45 min to allow monocyte adherence.Non-adherent cells were rinsed off.Monocytes were either differentiated into monocyte-derived macrophages (MDMs), as previously described [109], or lifted by pipetting after 10 min in warm 10× TrypLE Select (Life Technologies, Carlsbad, CA, USA) for seeding into assay plates.

MTT Assay
Both 16HBE and smoker NHBE cells were seeded, in triplicate, in 200 µL of growth medium at 1 × 10 4 cells/well in 96-well plates and allowed to adhere overnight.The growth medium was replaced with 100 µL of treatment for 24 h with 10 µL of 5 mg/mL MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide; Sigma Aldrich), in PBS, added for the final 4 h of incubation.Treatment media were carefully removed and the formazan crystals were solubilised with 50 µL of dimethyl sulfoxide (Sigma Aldrich) before absorbance at 570 nm was measured.

Phagocytosis Assay
Non-typeable Haemophilus influenzae (NTHi) were prepared, and the phagocytosis assay was performed with a 50:1 target-to-macrophage ratio, as previously described [8].Treated macrophages that were not exposed to targets were used as a gating control, as previously demonstrated [8], and phagocytosis was assessed by collecting events on a FACSCanto II with FACS DIVA 7.0 (BD Biosciences, Milpitas, CA, USA).

Migration Assay
THP-1 monocytes, at a concentration of 1 × 10 6 cells/mL, were labelled with 0.4 µM Calcein AM for 30 min at room temperature in serum-free media.Excess Calcein was removed by centrifugation 150× g for 5 min.The cell pellet was resuspended twice in serum-free media and centrifuged.A mixture of 120% EVE and 250% CSE was used to allow 100% EVE + 5% CSE treatment.Cells were treated under serum-free conditions for 22.5 h before 50,000 cells were added to the triplicate wells of a ChemoTX™ chemotaxis plate (5 µm pore, #111-5, Neuroprobe, Gaithersburg, MD, USA), with 10% FCS + 100 ng/mL IL-2 in the bottom chambers as a chemoattractant.The top filter was wiped with a cotton bud and then gently rinsed with PBS.The plate, with the filter still attached, was centrifuged at 150× g for 5 min and then read at 490/520 nm.Pre-treatment with 10 µM Cytochalasin D was used as a negative control, and the cells added directly to the bottom well served as a positive control, to validate the assay method.

Differentiation Assay
To assess whether EVE and CSE treatments induced spontaneous THP-1 monocyte differentiation, as measured by induction of adhesion, 50,000 THP-1 monocytes were seeded and treated as per the migration assay for 24 h.Differentiation was induced by 45 µM phorbol 12-myristate 13-acetate (PMA) as a positive control.Non-adherent cells were gently removed and adherent cells gently rinsed once with sterile PBS.Adherent cells were stained with 1% Crystal Violet in methanol for 30 min and washed with water 3 times.Plates were air dried for 5 min before adding 95% ethanol and placed on a shaker for 15 min to dissolve the crystal violet.Absorbance at 570 nm was measured.

LDH Assay
For epithelial cell toxicity analysis, the supernatant was collected from MTT assay plates, and for THP-1 monocytes, the supernatant was collected from differentiation assay plates.Supernatants were analysed for lactate dehydrogenase (LDH) activity as previously described [8].The assay was performed as per manufacturer's instructions (Roche, Basel, Switzerland).In a separate well, Tween-20 was added at 2% v/v, mixed and incubated for 5 min to lyse cells as a 100% control of maximal LDH release.

Cytokine Analysis
The supernatant was collected from treated MDMs being used for phagocytosis assays.The analysis of MDM-secreted cytokines via cytometric bead array (CBA, BD Biosciences) was performed as per the manufacturer guidelines for cell culture supernatants.CBA was analysed on a FACSCantoII with FCAP array software v3.0.Human soluble protein flex sets were used.The targeted cytokines were RANTES, IL-8, MCP-1, IP-10, and MIG.The supernatant was tested at a 1:4 dilution.(Perth, Australia) for volatile organic chemical (VOC) markers using gas chromatography mass spectrometry (Agilent Technologies, Mulgrave, Australia).The extract samples were diluted with ultra-high-purity water and spiked with relevant internal standards prior to analysis using a Restek Rxi-624Sil MS GC column (Restek Corporation, Bellefonte, PA, USA), with specifications of 20 m × 0.18 mm, 1.0 µm film thickness, fitted with a 1 m × 0.32 mm deactivated guard.The inlet temperature was set at 250 • C and samples (0.5 µL) were injected with a 5:1 split.Helium was used as the carrier gas, and the GC temperature program was as follows: 60 • C for 8 min;

Statistical Analysis
The data are presented as mean ± SEM for cell line work or median and individual data points for ex vivo work.The Kruskal-Wallis non-parametric ANOVA with the Mann-Whitney U test was employed for statistical analysis for cell line work.Friedman's nonparametric test with the Wilcoxon signed ranks pairwise test was performed for analysis of the primary cell work.Differences from control of p < 0.05 were considered significant.All statistical tests were performed using Graphpad Prism 9.

Conclusions
These data, combined with previous work by others, highlight the fact that dual use of tobacco cigarettes and e-cigarettes may place Dual Users at risk of increased harm instead of the marketed concept of harm reduction.Our data suggest Dual Users may be at greater risk from infections than smokers by way of reduced phagocytosis of bacteria, reduced monocyte recruitment signals and, consequently, reduced monocyte migration towards the infection site.We provide one possible reason for this: the effects of altered chemicals created by the interaction of reactive chemicals in cigarette smoke with the chemicals in e-cigarette vapour, and vice versa, which could occur in the lungs of Dual Users.When one considers that dual use exposes the user to an increased number of chemicals compared to smoking alone, along with exposure to chemicals formed by the interaction between cigarette smoke and e-cigarette vapour, dual use should not be considered as a form of harm reduction.
Author Contributions: M.P.W. performed the majority of the experiments with the aid of R.H., N.A.B., A.W. and L.T., and also conducted data analysis and manuscript preparation.All authors have read and agreed to the published version of the manuscript.Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Figure 2 .
Figure 2. Proliferative capacity of Smoker bronchial epithelial cells in a Switcher and Dual-User model.Conversion of MTT to formazan was measured in SNHSBE cells exposed to EVE (Switcher), EVE + CSE (Dual User), or controls for 24 h.Unless indicated otherwise, 100% EVE and 5% CSE

Figure 2 .Figure 2 .
Figure 2. Proliferative capacity of Smoker bronchial epithelial cells in a Switcher and Dual-User model.Conversion of MTT to formazan was measured in SNHSBE cells exposed to EVE (Switcher), EVE + CSE (Dual User), or controls for 24 h.Unless indicated otherwise, 100% EVE and 5% CSE Figure 2. Proliferative capacity of Smoker bronchial epithelial cells in a Switcher and Dual-User model.Conversion of MTT to formazan was measured in SNHSBE cells exposed to EVE (Switcher), EVE + CSE (Dual User), or controls for 24 h.Unless indicated otherwise, 100% EVE and 5% CSE were used.Data are plotted as the median plus individual data points.a: p < 0.05 compared to control.b: p < 0.05 compared to 5% CSE, * p < 0.05 between pairs ** p < 0.01 between pairs.

Figure 3 .
Figure 3. Dual exposure of THP-1 monocytes.THP-1 monocytes were seeded in CSE, EVE, or EVE + CSE and (A) LDH release, (B) differentiation, and (C) migration was measured.Unless indicated otherwise, 100% EVE and 5% CSE were used.N = 6.Mean ± SEM. a: p < 0.05 compared to control.b: p < 0.05 compared to 5% CSE, ** p < 0.01 between pairs.PMA was used as a positive control for differentiation, and Cytochalasin D was used as a negative control for migration.

Figure 3 .
Figure 3. Dual exposure of THP-1 monocytes.THP-1 monocytes were seeded in CSE, EVE, or EVE + CSE and (A) LDH release, (B) differentiation, and (C) migration was measured.Unless indicated otherwise, 100% EVE and 5% CSE were used.N = 6.Mean ± SEM. a: p < 0.05 compared to control.b: p < 0.05 compared to 5% CSE, ** p < 0.01 between pairs.PMA was used as a positive control for differentiation, and Cytochalasin D was used as a negative control for migration.

Figure 4 .
Figure 4. Effects of dual exposure on monocyte migration and MDM phagocytosis.(A) Healthy nonsmoker donor monocytes were exposed to CSE, EVE, or EVE + CSE, and monocyte migration was assessed.(B) Monocyte-differentiated macrophages from healthy non-smokers were exposed to CSE, EVE, or EVE + CSE, and their ability to phagocytose NTHi bacteria was assessed.(C) Examples of healthy monocytes migrated to lower well via microscopy.Unless indicated otherwise, 100% EVE and 5% CSE were used.The data show the median value plus individual data points.a p < 0.05 compared to control.b p < 0.05 compared to 5% CSE, ** p < 0.01 between pairs.

Figure 4 .
Figure 4. Effects of dual exposure on monocyte migration and MDM phagocytosis.(A) Healthy non-smoker donor monocytes were exposed to CSE, EVE, or EVE + CSE, and monocyte migration was assessed.(B) Monocyte-differentiated macrophages from healthy non-smokers were exposed to CSE, EVE, or EVE + CSE, and their ability to phagocytose NTHi bacteria was assessed.(C) Examples of healthy monocytes migrated to lower well via microscopy.Unless indicated otherwise, 100% EVE and 5% CSE were used.The data show the median value plus individual data points.a p < 0.05 compared to control.b p < 0.05 compared to 5% CSE, ** p < 0.01 between pairs.

4. 10 .
Mass Spectrometry of E-Cigarette Vapour, Cigarette Smoker, and Dual Extracts EVE, CSE, and dual extracts were externally analysed by Envirolab Services Pty Ltd.
2 • C/min to 90 • C; 5 • C/min to 150 • C; 20 • C/min to 230 • C; then hold for 2 min.The MS detector was operated at 300 • C with a solvent delay of 3 min across a m/z scan of 20-550.

Funding:
M.P.W. has received funding from the Royal Adelaide Hospital and The Faculty of Medical Sciences, University of Adelaide, for this work.Institutional Review Board Statement: The study was conducted in accordance with the Declaration of Helsinki, and approved by the Royal Adelaide Hospital Ethics Committee HREC/17/RAH/285 (R20020811, 20 August 2022).

Table 1 .
Significance values for comparisons between single and dual exposure treatments on the proliferative potential of SNHBE cells.

Table 3 .
Significant differences between comparisons of single and dual exposures on healthy non-smoker monocyte migration.
Nicotine, mango, and tobacco were used at 100%, whilst banana and chocolate were used at 25%. p < 0.05 for all the values presented.NS = not significant.-, comparison not applicable.

Table 4 .
Mass spectrometry analysis of VOCs in single-and dual-extract samples frozen immediately after collection.
STILL DETECTED = detected in single extracts and also in dual extracts.NO LONGER DETECTED = detected in single extracts and no longer detected in dual extracts.REDUCED DETECTION = full detection in single extracts, only trace detection in dual extracts.NEW = only detected in dual extracts and not detected in single extracts.

Table 5 .
Mass spectrometry analysis of VOCs in single-and dual-extract samples incubated at 37 • C with 5% CO 2 for 24 h immediately after collection.
STILL DETECTED = detected in single extracts and also in dual extracts.NO LONGER DETECTED = detected in single extracts and no longer detected in dual extracts.NEW = only detected in dual extracts and not detected in single extracts.