Biomarkers of disease can be detected in mice as early as 4 weeks after initiation of exposure to third-hand smoke levels equivalent to those found in homes of smokers

Cigarette toxins affect biological processes as well as organ systems and are detrimental to human health [1,2]. However, most of the emphasis on the dangers of cigarette-smoke exposure has been centered on first-hand smoke (FHS) and second-hand smoke (SHS). FHS is the smoke inhaled directly by the smoker whereas SHS consists of the smoke and toxins emitted when the smoker exhales the smoke of the cigarette into the air, as well as the smoke that is released from the burning end of a cigarette. SHS exposure occurs when people are in the vicinity of someone smoking. A new form of exposure to cigarette toxins is known as the third-hand smoke (THS). THS consists of deposition of SHS toxins on surfaces such as, household upholstery, bedding, carpeting, and furniture where smoking has occurred as well as the clothing, skin, and hair of people who have smoked. These toxins accumulate and age by reacting with the ambient air and changing into carcinogenic chemicals such as tobacco-specific nitrosamines [3]. These toxins cannot be seen but can be smelled and remain on surfaces for many years, resistant to even strong cleaning agents [4]. Therefore, THS people are usually unaware that they are being exposed. Bystanders inhale not only the contaminated air but also are exposed to THS by absorption through the skin [4]. Children are at the highest risk because they are in close contact with household surfaces, they frequently put their hands in their mouths, and have a high surface-to-volume ratio leading to the absorption of these toxins through the skin very effectively. In fact, children living in homes where smoking has occurred contain tobacco metabolites in their urine [5].

We have developed a model system in which the exposure of animals to THS mimics that of human exposure [6]. We found that THS causes alterations in multiple organ systems such as liver, lung, skeletal muscle, and healing of wounds which show various biomarkers of disease [6–9]. In the liver, we found that THS leads to increased lipid levels, indicating non-alcoholic fatty liver disease (NAFLD). This results in increased lipid levels in circulation that potentially contribute to cardiovascular disease. Furthermore, dyslipidemia can result in increased inflammation that could result in development of cirrhosis [10–12]. In the lung, collagen production is increased in addition to the increase in the levels of inflammatory cytokines, suggesting the potential development of fibrosis with implications for inflammation-induced diseases such as chronic obstructive pulmonary disease and asthma [13–14]. In the skin, healing of wounds mimics poor healing of surgical incisions observed in human smokers with an increased propensity of the wound to re-open [15]. In the brain, we find alterations in behavior, including increased hyperactivity [6]. In addition, mice develop hyperinsulinemia, yet they cannot effectively clear glucose from the blood, leading to hyperglycemia as a result of molecular changes in the insulin–insulin receptor signaling pathway [7]. When exposed to THS and fed a high-fat and carbohydrate diet (‘western type diet’), these molecular changes and symptoms of insulin resistance (IR) become worse. THS exposure also increases the levels of proinflammatory cytokines in the serum, including the interleukins 1, 4, 6, tumor necrosis factor α (TNF-α), and granulocyte macrophage colony-stimulating factor (GM-CSF), which suggests that the body is in a proinflammatory state. Finally, the most striking attribute that we observe in mice exposed to THS is a generalized increase in oxidative stress and a decrease in the antioxidant enzymes, catalase, and glutathione peroxidase (GPx) in most of the tissues [7]. As a result, we observe oxidative stress mediated damage to DNA, proteins, and lipids [7,9].

It is now important to determine what is the minimum time of exposure to THS required to identify key diagnostic biomarkers of damage and disease associated with THS exposure. These biomarkers, once validated in humans, can be used as critical indicators of exposure to THS, and how long this exposure has occurred. Ideally, in a clinical setting, urine or blood samples can be taken from children that are suspected to be sick because of THS exposure to determine, when, how long, and how much damage THS has caused. For these studies, we exposed different cohorts of mice to THS for 1, 2, 4, and 6 months and then tested these mice for hormonal alterations, IR, metabolic syndrome, and oxidative stress mediated damage to the liver (because it is the detoxifying organ) and to the brain because of the previously observed behavioral changes. Our findings are significant because they show critical times needed for THS exposure to cause health problems and have the potential to be used as diagnostic tools to detect THS exposure in humans, especially children.

Male C57BL/6 mice were exposed to THS for 1, 2, 4, 6, months starting at weaning. Age-matched control mice were kept in clean air. Both control and THS-exposed mice were maintained under controlled environmental conditions – 12-h light/dark cycle in conventional cages with ad libitum access to standard chow (percent calories: 58% carbohydrate, 28.5% protein, and 13.5% fat) and water.

All animal experimental protocols were approved by the University of California, Riverside, Institutional Animal Care and Use Committee (UCR-IACUC). Mice were killed by carbon dioxide (CO₂) inhalation, which is the most common method of killing used by NIH. The amount and length of CO₂ exposure were approved by UCR-IACUC.

Common household fabrics such as curtain material, upholstery, and carpet were placed in empty mouse cages and exposed to SHS generated by the Teague smoking machine as previously described [16]. Each cage contained 10 g of curtain material (cotton), 10 g of upholstery (cotton and fiber), and two 16 in² pieces of carpet (fiber) to maintain equal exposure levels across all experimental groups. Two packs of 3R4F research cigarettes were smoked each day, intermittently to mimic the behavior of human smokers, 5 days per week. The smoke was routed to a mixing compartment to mix with air and the mixture distributed between two exposure chambers, each containing four cages with the materials. The gravimetric method was used to determine the total particulate matter (TPM) concentrations in each smoke chamber as previously done. The TPM values were adjusted to 30 ± 5 µg/m³ which is within the values found by the EPA in the homes of smokers. All cigarettes were smoked and stored in accordance with the Federal Trade Commission (FTC) smoking regimen. At the end of each week, cages were removed from the exposure chamber, bagged, and transported to the vivarium where mice were placed into the cages with the THS-exposed material. Exposure was performed as previously described [6].

Brain or liver tissues were collected after the appropriate exposure to THS and immediately frozen and stored at –80°C. Tissues were homogenized in radioimmunoprecipitation assay buffer (RIPA), centrifuged, and the supernatant collected for the studies unless otherwise specified by the assay procedures. All procedures described below have been performed previously by us [7] and are therefore described briefly.

Blood was collected directly from the heart and allowed to coagulate on ice for 20–30 min. The samples were then centrifuged at 10000 rpm for phase separation, and the serum was used immediately or frozen and stored at –80°C for future assays.

Fasting blood glucose levels were measured using a commercially available kit (TRUE result) and gold sensor laser accuracy strips. Mice were fasted for 12 h in order to ensure that both glucose and insulin were at the base levels. Both variables were affected by food and water and that effect varied from animal to animal. Therefore, in order to make sure that we could compare the levels of glucose and of insulin amongst animals, we always fasted mice under the same conditions. To obtain fasting blood glucose levels, a nick was made on the tail vein and the glucose levels were measured by bringing the strip into contact with the drop of blood, which then the TRUE result glucose reader provided a readout of the levels of glucose present in the blood.

We used the ALPCO Mouse Insulin ELISA Kit (Cat# 80-INSMS-E01) and followed the protocol provided by the manufacturer. The plates were read at λ = 450 nm. Concentration of the samples was obtained from the standard curve.

Homeostatic model assessment equation (HOMA) is a method for assessing IR derived by (fasting blood glucose × fasting insulin)/22.5.

We used Cayman Superoxide Dismutase Assay Kit (Cat# 706002) to measure the superoxide dismutase (SOD) levels in liver and brain tissues. SOD activity was quantitated by measuring the dismutation of superoxide radicals generated by xanthine oxidase and hypoxanthine. A standard curve was generated and used to quantitate SOD activity.

We used Cayman Hydrogen Peroxide Assay Kit (Cat# 600050) to measure H₂O₂ levels in liver and brain tissues. H₂O₂ was detected using the probe 10-acetyl-3,7-dihydroxyphenoxazine (ADHP) in the presence of horseradish peroxidase, producing a highly fluorescent resorufin which was then quantitated using a spectrophotometer. A standard curve was generated to quantitate hydrogen peroxide.

We used Cayman Catalase Assay Kit (Cat# 707002) to measure catalase activity in the liver and brain tissues. Catalase activity was quantitated by measuring the reaction of catalase with methanol in the presence of an optimal concentration of H₂O_2. The chromogen 4-amino-3-hydrazino-5-mercapto-1,2,4-triazole (purpald) was used to measure the reaction rate, using spectrophotometer. A standard curve was generated and used to quantitate the activity of catalase.

We used Cayman GPx activity assay kit (Cat# 703102) to measure GPx (Cat# 703102) activity in the liver and brain tissues. GPx activity was quantitated by the coupled reaction of GPx with glutathione reductase (GR). GSSG, produced upon reduction of an organic hydroperoxide by GPx, was recycled to its reduced state GR and NADPH. The oxidation of NADPH to NADP⁺ was accompanied by a decrease in absorbance at 340 nm. The rate of decrease in the A340 was directly proportional to the GPx activity in the liver and brain samples.

DNA damage was measured using ELISA Kit from Cell BioLabs, Inc. (Cat# STA-320). DNA damage was quantitated by measuring levels of 8-hydroxy-2-deoxyguanosine (8-OH-dG) in brain and liver tissues. DNA was extracted from tissues as previously described [7]. A standard curve was generated and used to quantitate 8-OH-dG in the extracted brain and liver DNA.

Protein damage was quantitated by measuring the nitrosylation of tyrosine residues in proteins using ELISA Kit from Cell BioLabs (Cat# STA-305). Proteins were extracted from liver and brain samples as previously described. Samples were then loaded on to a 96-well ELISA plate precoated with anti-nitrotyrosine antibody followed by an HRP–conjugated secondary antibody. Nitrotyrosine levels were then interpolated from a standard curve.

Lipid peroxidation in the liver and brain tissues was determined by the levels of the by-product of lipid peroxidation known as malondialdehyde (MDA) using a kit from Cell BioLabs (Cat# STA-330). The unknown MDA containing samples or MDA standards were first reacted with thiobarbituric acid (TBA) to form what are known as TBA reactive substances which can then be quantitated by comparison with the predetermined TBARS standard curve.

ALT levels were measured using ELISA Kit from Cayman Chemicals (Cat# MAK055), and quantitated by incubation of liver and serum samples in an anti-ALT antibody coated plate, followed by HRP–conjugated secondary antibody. A standard curve was generated and used to quantitate ALT in the tissue and serum.

The ATP levels in liver and brain tissues were quantitated using a kit from BioVision (Cat# K354-100). ATP levels were quantitated by measuring the phosphorylation of glycerol. The product of this phosphorylation reaction was read at λ = 570 nm and the absorbance readings of the standards were used to generate a standard curve. The unknown levels of ATP in the samples were interpolated from a standard curve.

The levels of cytokines in the serum were quantitated using ELISA array kit from R&D systems (Cat# ARY006). This ELISA assay kit allowed for the analysis of multiple cytokines and chemokines simultaneously, interleukin-1α (IL-1α), MCP-1, IL-1β, IL-2, interleukin-4 (IL-4), interleukin-6 (IL-6), interleukin-10 (IL-10), IL-12, IL-17, IFN-γ, TNF-α, G-CSF, and GM-CSF. Serum was incubated in corresponding ELISA wells of each cytokine tested, incubated with the corresponding antibody, followed by HRP–conjugated secondary antibody. A standard curve was generated and used to quantitate the levels of each cytokine tested.

The adrenocorticotropic hormone (ACTH) levels in the serum were quantitated using ELISA Kit from MyBioSource (Cat# MBS52341). ACTH levels were quantitated by incubation of serum in an anti-ACTH antibody-coated plate, followed by HRP–conjugated secondary antibody. A standard curve was generated and used to quantitate ACTH levels.

The corticotropin-releasing hormone (CRH) levels in the serum were quantitated using an ELISA Kit from MyBioSource (Cat# MBS16352). CRH levels were quantitated by incubation of serum in an anti-CRH antibody-coated plate, followed by HRP–conjugated secondary antibody. A standard curve was generated and used to quantitate CRH levels.

The dopamine levels in the serum were quantitated using an ELISA Kit from MyBioSource (Cat# MBS32332). Dopamine levels were quantitated by incubation of serum in an anti-dopamine antibody-coated plate, followed by HRP–conjugated secondary antibody. A standard curve was generated and used to quantitate dopamine levels.

The epinephrine levels in the serum were quantitated using ELISA Kit from MyBioSource (Cat# MBS12232). Epinephrine levels were quantitated by incubation of serum in an anti-epinephrine antibody-coated plate, followed by HRP–conjugated secondary antibody. A standard curve was generated and used to quantitate epinephrine levels.

The nuclear respiratory factor 1 (NRF1) levels in the liver were quantitated using ELISA Kit from MyBioSource (Cat# MBS1235327). NRF1 levels were quantitated by incubation of liver in an anti-NRF antibody-coated plate, followed by HRP–conjugated secondary antibody. A standard curve was generated and used to quantitate NRF1 levels.

We used the procedure previously published by us [7]; for details, see the S1 Methods in [7].

The proopiomelanocortin (POMC) levels in the serum were quantitated using ELISA Kit from MyBioSource (Cat# MBS1315234). POMC levels were quantitated by incubation of serum in an anti-POMC antibody-coated plate, followed by HRP–conjugated secondary antibody. A standard curve was generated and used to quantitate POMC levels.

For the statistical analysis of experiments, we used GraphPad InStat Software (GraphPad, La Jolla, CA, U.S.A.). Statistical comparisons between two groups was performed using the unpaired Student’s ttest. We also used a 2 × 3 ANOVA comparing the effects of treatment (THS or control) with time (2, 4, and 6 months). All data are mean ± S.D. represented by the error bars. Means were considered significantly different when P<0.05.

For all the studies shown here, we investigated levels of various molecules chosen to detect liver and brain damage as well as metabolic syndrome markers. We exposed different cohorts of mice to THS for 2, 4, and 6 months and tested these mice against 24 biomarkers (Table 1). Fifteen biomarkers were tested in serum and nine biomarkers were tested in liver and brain tissue of the mice exposed to THS in a time-dependent manner. For those variables that showed increases at 2 months of exposure, we then determined if they were already altered at 1 month of exposure.

The data are presented in four sections: effects on insulin metabolism, on oxidative stress, on hormonal changes, and on inflammation. Fasting blood glucose and insulin were used to test for metabolic syndrome. H₂O₂, SOD, catalase, GPx, nitrotyrosine, lipid peroxidation, and DNA damage were used to test for liver and brain oxidative stress. CRH, ACTH, epinephrine, and cortisol were used to test for hormonal alterations. IL-1α, IL-4, IL-6, IL-10, GM-CSF, and TNF-α were used to test for markers of inflammation.

In order to determine whether there is a time-dependent increase in markers of IR with increased THS exposure and at what length of exposure the body becomes insulin resistant, we tested fasting blood glucose and insulin levels in the exposed mice. We first fasted the mice for 12 h and then obtained blood from a nick in the tail to determine the blood glucose levels. The mice were then killed and blood obtained from the heart to test for insulin levels. We found that fasting blood glucose and serum insulin levels increased after 3–4 months of THS exposure (Figure 1A,B). To quantitate IR, we used the homeostatic IR index (HOMA-IR) and found that mice exposed to THS for 4 and 6 months had significantly higher indices than controls (Figure 1C). At 2 months, we begin to see the general trend that fasting blood glucose and insulin levels are increasing. However, it takes the body approximately 3–4 months for the differences to become significant.

Cigarette smoke toxins, such as those found in THS, are potent sources of reactive oxygen species (ROS) [17]. To determine whether mice exposed to THS have oxidative stress in the liver, the organ that detoxifies the body, we measured the levels of H₂O₂ and found that by 2 months of exposure, H₂O₂ was already significantly increased in the liver tissue of THS-exposed mice (Figure 2A). Because H₂O₂ is produced by SOD we measured the activity of this enzyme to determine whether it is also elevated. Indeed, we found that SOD was significantly increased by 2 months (Figure 2B), much like H₂O₂. H₂O₂ is a reactive molecule and therefore needs to be quickly broken down to H₂O and O₂ by catalase and GPx, two antioxidant enzymes, before it damages the cells. However, catalase activity was the same as in control at 2 months, but by 4–6 months the activity of this enzyme was significantly decreased (Figure 2C). GPx activity was also not significantly changed earlier but it decreased at 6 months (Figure 2D). Taken together, these results show that THS exposure reduces the antioxidant potential of the liver and leads to increased oxidative stress. Furthermore, the increased levels of H₂O₂ can react with Fe²⁺ ions in the tissue and undergo the Fenton reaction leading to production of hydroxyl radicals [18–21]. These, in turn, can lead to protein modification, formation of DNA adducts, and lipid peroxidation. We found that THS exposure causes nitrotyrosine modification in proteins by 2 months of exposure (Figure 2E) and DNA adduct formation by 3–4 months (Figure 2F). Lipid peroxidation was not seen until 6 months of exposure (Figure 2G). Because of these molecular changes, we examined the possibility that these alterations could impair mitochondrial function by measuring the levels of ATP produced, given that ATP is the energy currency of the cell. We found that by 2 months, the levels of ATP were already significantly lower than the control (Figure 3A) and that there was a concomitant increase in lactate levels (Figure 3B), suggesting a switch in energy production from oxidative phosphorylation to glycolysis. We also measured two other biomarkers that indicate liver malfunction when elevated, aspartate aminotransferase (AST) and urea. Both were elevated by 2 months of exposure (Figure 3C,D). Given that there is an increased oxidative stress, molecular damage, as well as mitochondrial dysfunction in the liver, we examined the levels of NRF1, a nuclear transcription factor that becomes up-regulated, in the liver, in response to oxidative stress or inflammation. We found that in the liver of 2 months old THS-exposed mice, NRF1 levels were much higher than in controls (Figure 3E), suggesting that as early as 2 months of exposure to THS, the body is under significant oxidative stress, however NRF1 is failing to combat this oxidative stress because that oxidative stress continues to see it worsen as time of exposure increases.

Because we have previously shown that THS exposure affects behavior, we measured the levels of H₂O₂ as well as the levels of SOD and catalase and GPx activity in brain tissue. H₂O₂ levels increased significantly by 4 months of THS exposure (Figure 4A). Interestingly, we did not observe a significant increase in SOD activity in the brain (Figure 4B) as we did in the liver, and catalase and GPx activities were only significantly decreased by 6 months after THS exposure (Figure 4C,D). We also observed oxidative stress mediated molecular damage in the brain of THS-exposed mice with significantly elevated levels of nitrotyrosine that occurred in 2 months of THS exposure (Figure 4E). However, DNA damage (Figure 4F) and lipid peroxidation (Figure 4G) were not seen until 6 months of THS exposure. The differences in the findings between brain and liver suggest that the brain might have less oxidative stress due to THS exposure because the blood–brain barrier might play a protective role regarding the THS toxins.

Given that we observed molecular damage in the brain, we examined the levels of ATP production and found that by 2 months of THS exposure, mice had decreased ATP production (Figure 5A). This decrease was accompanied by an increase in lactate (Figure 5B) much like we observed in the liver, indicating a switch to glycolysis as an alternative way for energy production.

During times of physiological stress (e.g. THS toxins), the body activates the hypothalamus–pituitary axis (HPA axis), a complex set of neuroendocrine-mediated changes that regulates many body processes, including insulin sensitivity, the immune system, mood, energy storage/expenditure, and physically prepares the body to combat stressors [22–24]. Upon detection of stress by the brain, the hypothalamus secretes a hormone known as CRH which then stimulates the anterior pituitary to release ACTH which, in turn, travels in the blood to the adrenal glands to stimulate the secretion of hormones such as cortisol and epinephrine which are stress responders. We observed that CRH was only significantly more elevated at 6 months of exposure (Figure 6A) whereas ACTH was elevated as early as 2 months (Figure 6B). Similarly, epinephrine was increased by 2 months of exposure (Figure 6C) but cortisol was only seen to be elevated at 4 months of exposure (Figure 6D). Given the fact that ACTH was increased independently from CRH levels, we determined whether this was due to the increased production of the POMC hormone. This hormone is involved in an alternative pathway of ACTH production. During times of oxidative stress in the brain, the hypothalamus gland will secrete POMC hormone, which will activate the pituitary gland to produce ACTH [25]. We found that, both in the serum and brain of mice exposed to THS for 2 months, there were increased levels of the POMC hormone (Figure 6E,F).

In order to determine whether exposure to THS toxins results in an increase in inflammation in a time-dependent manner, we examined the levels of cytokines in the serum. We found several proinflammatory cytokines were elevated (for a list of cytokines in the array, see the Materials and methods section). Mice exposed to THS for as early as 2 months had the proinflammatory cytokines IL-1α, IL-6, GM-CSF, and TNF-α levels significantly higher than those of the controls (Figure 7A–D). Collectively these cytokines indicate a proinflammatory condition in response to THS exposure. THS-exposed mice also have increased anti-inflammatory cytokines IL-4 and IL-10 (Figure 7E,F). These data taken together show that as early as 2 months, THS-exposed mice suffer from significant inflammation. Because it is known that pathways involved in inflammatory responses that are driven by cytokines commonly involve the mitogen-activated protein kinases, P38 and ERK [26–28], we determined whether these kinases are involved in the THS-induced inflammatory response. We found that P38 levels were increased in the liver of mice exposed to THS for 2 months and so were the levels of p-ERK, which is the form of functional ERK (Figure 7G,H).

To determine whether the biomarkers that were altered at 2 months after THS exposure were already altered at 1 month of exposure, the mice were exposed to THS for only 1 month, and the biomarkers that we tested and showed significance at 2 months were examined at 1 month. These biomarkers included serum ACTH, epinephrine, AST, and urea, nitrosylation of proteins in the liver and brain, and the proinflammatory cytokines, IL-1α, IL-4, IL-10, GM-CSF, and TNF-α. We found that 1 month of THS exposure resulted in significant increases in, epinephrine, TNF-α, GM-CSF, and serum AST but not the other markers. (Figure 8A–H).

The goal of the present study was to determine the minimum amount of time required to cause physiological changes in mice when they are exposed to THS, using an exposure system that mimics human exposure. In order to verify that THS exposure over time had a significant difference in the changes to the variables we measured, we used a mixed model repeated measure two-ANOVA to determine whether the effects of THS are dependent on exposure time and we found a positive time-dependent significant correlation with increase time of THS exposure and the effects it had on all the variables we measured. Figure 9 shows that of the biomarkers tested, the ones altered in the serum after only 1 month of THS exposure are GM-CSF, TNF-α, epinephrine, and AST and that in the liver and brain tissues we have increased nitrotyrosine levels. In addition to continued alteration in the biomarkers we found to be changed at 1 month, after 2 months of exposure, we found increased fasting blood glucose and IL-6 in the serum. In the liver, we find increased hydrogen peroxide levels and increased SOD activity but decreased catalase activity, as well as increased DNA damage. At this time of exposure in the brain, we find increased hydrogen peroxide levels. After 4 and 6 months of exposure, we also find increased serum insulin and CRH levels. We also show molecular alterations to pathways of oxidative stress response through increased presence of NRF1, stress hormone production pathways through POMC, and inflammatory response pathways through P38, and ERK. Our studies can indicate whether or not, and for how long, a mouse has been exposed to THS, by measuring these variables in the blood, making these biomarkers both easy to test and highly correlative with THS exposure.

In our previous research, we observe that mice exposed to THS for 6 months, develop symptoms of IR and metabolic syndrome, with increases in fasting blood glucose, fasting serum insulin, and HOMA-IR index [7]. In the studies presented here, we show that there is an increase in biomarkers of IR with increased THS exposure, as well as identifying critical times of exposure at which the body becomes insulin resistant. Fasting blood glucose, serum insulin, and the HOMA-IR index were increased in a time-dependent manner with THS exposure, showing that as early as 4 months of THS exposure, the animals already have a propensity for type 2 diabetes. In humans, small increases in blood glucose, over time, are highly damaging to the blood vessels and other tissues [29]. Formation of glycation end products on proteins present in cell membranes have been associated with increased blood glucose levels and have been one of the leading causes of cardiovascular diseases [30].

We have reported in previous studies the development of oxidative stress in multiple organs, including the lung, liver, skeletal muscle, brain, adrenal, and skin, after 6 months of THS exposure [6–9]. The work presented here focusses on the liver and brain because the liver plays a major role in metabolism and detoxification, and the brain also plays significant roles in behavior.

In the liver, the data we obtained on the oxidative stress levels after THS exposure show increases in H₂O₂ levels and SOD activity as early as 2 months. SOD is the primary enzymatic defense in the liver against the damaging effects of O₂^•–, by converting O₂^•– into H₂O₂ which is a substrate for catalase and GPx [31]. Because catalase and GPx activity levels are unchanged in the THS-exposed mice, H₂O₂ is not broken to H₂O and O₂^•. Under these conditions, H₂O₂ participates in the Fenton reaction leading to OH radicals, which are damaging to proteins, lipids, and DNA. The implications of this damage in an organ such as the liver is significant because it can hinder the capability of the organ to detoxify the body, leading to more exacerbated damage by THS toxins. Nitrotyrosine is widely used as a marker of post-translational modification by the nitric oxide (^•NO, nitrogen monoxide) derived oxidant peroxynitrite (ONOO^•–), formed when ROS are not effectively cleared by the natural antioxidant defense systems and as a result can react with proteins in the cell, nitrating them which potentially changes functionality, and as a result, compromises tissue function [32]. We observe high levels of nitrated proteins as early as 2 months of THS exposure, which can show how sensitive the liver is to these toxins and how ineffective it becomes to clear these ROS when exposure to THS occurs. Hydroxyl radicals react with DNA and, as a result, lead to the formation of adducts on the DNA bases, most commonly in the form of modified 8-OH-dG, which has been shown to cause misreading of DNA during replication and affects the activity of DNA polymerase [33]. Our data show that THS exposure results in DNA damage as early as 2 months exposure, and this risk becomes significant compared with controls after 4 and 6 months of exposure. In addition, cell membranes can be modified by lipid peroxidation products, such as trans-4-hydroxy-2-nonenal (HNE), 4-hydroperoxy-2-nonenal (HPNE), and MDA [34]. Lipid peroxidation products can also modulate signaling molecules and alter functions of enzymes and proteins in the cell. Our data show that significant lipid peroxidation occurs by 6 months of exposure, indicating that there is a longer length of time required for damage to lipid components of cells. Taken together, these data reveal that THS exposure results in oxidative stress mediated damage which can result in the development of chronic liver diseases such as NAFLD. Protein nitrosylation and lipid peroxidation have been associated with smoke induced development of NAFLD in rodents and humans and our results are showing a time-dependent correlation with THS exposure and on set of fatty liver disease.

In the brain, we observe that it takes much longer for THS exposure to result in the accumulation of H₂O₂ and alteration in the activity of the antioxidant enzymes. We propose that this delay is due to the presence of the blood–brain barrier preventing the THS toxins from entering the tissue freely. However, we see that by 4 months, H₂O₂ levels are significantly high, and that the activity of SOD, catalase, and GPx, are reduced. In the brain we again observe the increase in nitrosylation of proteins as early as 2 months as well as increased lipid peroxidation and DNA damage by 6 months of THS exposure. Thus, compared with the brain, the liver is more susceptible to oxidative stress mediated damage.

We previously showed that mice exposed to THS for 6 months showed increased hyperactivity, stress, and anxiety, compared with their control counterparts [6]. We hypothesized that these behavioral changes caused by THS exposure might be detected early by testing for increases in stress hormones in the serum. We found that the major hormones of the hypothalamus, pituitary, adrenal stress axis (HPA), and their precursors were significantly altered in THS-exposed mice. This finding was significant not only because increases in these hormones are associated with the behavioral changes we detected, but also because these hormones have a secondary effect on insulin signaling and metabolism [35]. During times of stress, the body, through the release of epinephrine and cortisol, converts the stress signals received by the brain to physiological responses by binding to receptors in tissues such as the liver and skeletal muscle [36]. We attributed the potential role of these hormones as one of the mechanisms by which THS-exposed mice develop IR. In these studies, we determined whether these hormones are altered in a time-dependent manner and at what time point of THS exposure do these alterations occur.

In the present work, the HPA-axis signaling cascade hormones CRH, ACTH, cortisol, and epinephrine increased in a time-dependent manner, with epinephrine and ACTH being high as early as 2 months of THS exposure. Cortisol showed significant increases at 4 months of exposure. We noticed, however, that CRH only increased at 6 months of exposure. At 2 months of THS exposure in the brain and serum there were increased levels of POMC, which probably explains why ACTH is so high at 2 months of exposure, independent of CRH levels. Further evidence comes from the fact that at 6 months of exposure when CRH levels are high, the ACTH levels are also increased, most likely to be because ACTH is also stimulated by CRH.

We found that as early as 2 months the proinflammatory cytokines, IL-1α, IL-4, GM-CSF, and TNF-α were increased compared with controls. IL-1α is a cytokine that is increased during states of inflammation and has roles in furthering the inflammatory response by increasing production of other cytokines such as TNF-α. We see that in mice exposed to THS for 2, 4, and 6 months, the levels of IL-1α increases which indicates that these mice are in a proinflammatory state which then is further increased by the production of other cytokines. Increases in levels of IL-1α have also been shown to have negative roles in lipid and glucose metabolism pathways, both of which occur in our THS-exposed mice [37].

GM-CSF is a cytokine that is increased during states of inflammation and has roles in monocyte recruitment and stimulation, which then go into circulation, invade tissues, and become macrophages, which in turn secrete inflammatory cytokines [37]. Furthermore, GM-CSF has been shown to be a signal transducer and transcription activator in macrophages, which leads to their production of ROS as a conserved mechanism for fighting infection. GM-CSF levels are increased in THS-exposed mice as early as 2 months adding to the evidence that these mice are in a proinflammatory condition and further explaining the high amounts of ROS and oxidative stress in the tissues.

TNF-α is a cell cytokine involved in systemic inflammation, and is mostly produced by macrophages which are activated to do so by IL-1α. TNF-α has many downstream pathways most of which induce cell death, mitochondrial fission, and/or breakdown, as well as activating the p38 and ERK inflammatory pathways [38]. We found that as early as 1 month of THS exposure TNF-α levels were increased, not only indicating inflammation, cell damage, and death, but also potentially explaining mitochondrial dysfunction that we observe with THS exposure. In addition to the increase in the levels of these inflammatory cytokines, we also see anti-inflammatory cytokines such as IL-4 and IL-10 [39] increased in THS-exposed mice as early as 2 months of exposure, as the body attempts unsuccessfully to resist the inflammation caused by THS exposure.

In summary, we show that THS toxins have major effects on the liver and brain, and on circulating hormones and cytokines, which show the general trend of increased susceptibility of the body to metabolic diseases and inflammation as the time of THS exposure increases. These results could potentially be used to estimate the length of exposure to THS. Furthermore, these data may be important to inform the public, in particular the parents of children living in the homes of smokers, about the danger of exposure to THS. We summarize in Table 2, a key set of easily testable serum biomarkers altered as a result of THS exposure at different times. We can see that at 1 month of exposure, which we label as ‘mild exposure’, we detect alterations in GM-CSF, TNF-α, epinephrine, AST. ‘Moderate exposure’ takes ~2 months and consists of increased ACTH, urea, POMC, IL-1α, IL-10, IL-4 in the serum. After 4 months of exposure, ‘high exposure’, we detect fasting blood glucose, insulin, IL-6, CRH, and cortisol in the serum. These biomarkers can potentially be used to classify how long one has been exposed to THS toxins and the extent of cellular and molecular damage that could lead to disease. Maturation of humans is much slower than in mice, hence we expect that the exposure times shown here would be longer for each stage.

• The present study investigates the time-dependent effects of THS exposure on health, with the goal of correlating exposure time with a specific set of biomarkers.

• We showed THS exposure as early as 1 month, can lead to changes in markers of health in the serum, liver, and brain tissues. These markers increase in a time-dependent manner showing that the body has symptoms of metabolic syndrome, oxidative stress mediated cellular damage, and systemic inflammation, which progressively worsen with increased time of exposure to THS.

• Our findings will help identify the approximate length of exposure to THS as well as the extent of damage. They also provide insight into the mechanisms by which THS exposure leads to physiological changes that lead to the onset of disease. This is particularly important for children and the elderly who are sensitive bystanders exposed to THS.

This work was funded by Tobacco Research Disease Related Program (TRDRP) grants under projects 22RT-0121and 23DT-0103. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

The authors declare that there are no competing interests associated with the manuscript.

N.A., Y.C. and M.M.G. conceived and designed the experiments. N.A. and Y.C. performed the experiments. N.A., Y.C. and M.M.G. analyzed the data. M.M.G. contributed reagents/materials/analysis tools. N.A., Y.C. and M.M.G. wrote the paper.

ACTH

adrenocorticotropic hormone

ALT

alanine aminotransferase

AST

aspartate

CRH

corticotropin-releasing hormone

EPA

Environmental Protection Agency

G-CSF

granulocyte colony-stimulating factor

GPx

glutathione peroxidase

GR

glutathione reductase

HOMA

homeostatic model assessment equation

HPA

hypothalamus–pituitary axis

IFN-γ

interferon-γ

IL

interleukin

IR

insulin resistance

MCP-1

monocyte chemoattractant protein-1

MDA

malondialdehyde

NAFLD

non-alcoholic fatty liver disease

NRF1

nuclear respiratory factor 1

POMC

proopiomelanocortin

ROS

reactive oxygen species

SHS

second-hand smoke

SOD

superoxide dismutase

TBA

thiobarbituric acid

TBARS

thiobarbituric acid reactive substances

THS

third-hand smoke

TNF-α

tumor necrosis factor α

TPM

total particulate matter

UCR-IACUC

University of California, Riverside, Institutional Animal Care and Use Committee