Third hand smoke (THS) is the accumulation of second hand smoke (SHS) toxins on surfaces in homes, cars, clothing and hair of smokers. It is known that 88M US nonsmokers ≥3 years old living in homes of smokers are exposed to THS toxicants and show blood cotinine levels of ≥0.05 ng/ml, indicating that the toxins are circulating in their circulatory systems. The goal of the present study is to investigate the mechanisms by which THS causes impaired wound healing. We show that mice living under conditions that mimic THS exposure in humans display delayed wound closure, impaired collagen deposition, altered inflammatory response, decreased angiogenesis, microvessels with fibrin cuffs and a highly proteolytic wound environment. Moreover, THS-exposed mouse wounds have high levels of oxidative stress and significantly lower levels of antioxidant activity leading to molecular damage, including protein nitration, lipid peroxidation and DNA damage that contribute to tissue dysfunction. Furthermore, we show that elastase is elevated, suggesting that elastin is degraded and the plasticity of the wound tissue is decreased. Taken together, our results lead us to conclude that THS toxicants delay and impair wound healing by disrupting the sequential processes that lead to normal healing. In addition, the lack of elastin results in loss of wound plasticity, which may be responsible for reopening of wounds.
The present study investigates the effects on wound healing of tobacco smoke toxins deposited on surfaces (THS) and seeks to understand the mechanisms of THS-induced impaired healing.
We show that THS delays wound healing by, increasing oxidative stress and its associated tissue damage due to protein nitration, lipid peroxidation and DNA damage. THS exposure also causes imbalance in chemokines and cytokines that result in decrease/delay in the inflammatory response, decrease in collagen deposition, decrease in angiogenesis and increase in blood vessel leakage and clot formation. Moreover, elastase levels are elevated which reduces elastin, decreasing wound plasticity, which may be responsible for reopening of wounds.
Our findings have implications for understanding how surgical wounds in smokers heal poorly and/or re-open. They also provide insight into the mechanisms by which SHS/THS exposure delays and impairs healing. This is particularly important for children and the elderly who are bystanders exposed to THS and who may be at extensive risk of developing difficult-to-treat surgical wounds.
Wound healing is a dynamic process involving many factors and cell types including soluble mediators, blood cells, fibroblasts, endothelial cells and extracellular matrix . Injury initiates a cascade of repair processes that normally begins with blood clotting, followed by a variety of oxidative and nitro-oxidative reactions. Normal cutaneous wound healing is divided into several sequential but overlapping phases in space and time–homoeostasis, inflammation, granulation tissue formation and tissue remodelling–that must occur in an orderly manner to achieve proper healing. Impaired and chronic wounds result from an imbalance in the healing cascade and have a significant impact on human health .
Cigarette smoke is a complex mixture of more than 7000 toxic substances . Third hand smoke (THS) is created by the recurring deposition of tobacco toxins on curtains, carpet, upholstery, bedding, windows and walls of buildings; and in vehicles, on car seats and other interior surfaces. It also lingers in smokers’ hair, clothing and skin [4,5]. The chemical conversion of second hand smoke (SHS) into THS is a dynamic aging process that begins within 30 min of smoking. This causes THS to become more toxic and potentially more harmful than fresh SHS with time . Contamination of the homes of smokers by SHS residues (i.e. THS) is high, both on surfaces and in dust, including in children's bedrooms. Recently, it has also been shown that THS remains in houses and apartments even long after the smokers move out . This makes THS a serious health risk–a stealth toxin–especially for children and the elderly who are more vulnerable to these toxins than adults [8–10] and are unable to remove themselves from these environments.
Although doctors require patients to refrain from smoking before scheduled surgery and have suggested beneficial effects due to abstinence [11,12], reports have also suggested that short term cessation of smoking does not effectively reduce the risk and complications during wound healing after surgery . Because THS toxins can affect infants, children and elderly, who are defenseless to that exposure, understanding the detrimental effects of THS on wound healing is important in identifying the risks associated with the residual toxicants left from SHS and those resulting from changes occurring due to interactions with the environment that contribute to formation of THS.
We have recently developed an animal model system that mimics human exposure to THS. Our studies showed significant damage to liver, lung and cognitive behaviour . Furthermore, THS was also shown to cause delayed healing . However, the cell and molecular mechanisms associated with the development of impaired wounds has not been studied. Thus, the goal of the present study is to investigate the mechanisms by which THS causes impaired wound healing. In the present study we show that wounds of THS-exposed mice close at a slower rate than those of control mice and that they have problems in epithelial migration, collagen deposition and granulation tissue formation. We also show problems in oxidative stress, inflammation, chemokine/cytokine production and angiogenesis, all of which are major components of a healthy regenerative tissue healing process. Our findings in the present study delineate the impact of THS on wound healing processes in mice and suggest the occurrence of similar problems in humans exposed to THS.
RESEARCH DESIGN AND METHODS
All animal experiments were approved by the University of California, Riverside, Institutional Animal Care and Use Committee (IACUC). Methodology used was as previously described . Briefly, C57BL/6 mice were divided into control and THS-exposed mice with exposure starting right after weaning and proceeding until 24 weeks of age. All mice were fed a standard chow diet (percent calories: 58% carbohydrate, 28.5% protein and 13.5% fat).
Using the previously established THS exposure system , common household fabrics were placed in mouse cages and subjected to SHS cigarette smoke. Briefly, each cage contained 10 g of curtain material (cotton), 10 g of upholstery (cotton and fibre) and 2.16 in2 pieces of carpet (fibre), in order to maintain equal exposure levels across experimental groups. Two packs of 3R4F research cigarettes were smoked each day, 5 days/week and smoke was routed to a mixing compartment and distributed between two exposure chambers containing cages with the materials. The total particulate matter (TPM) was monitored twice daily and kept at 30±5 μg/m3. All cigarettes were smoked and stored in accordance with the FTC smoking regimen. Mice were then placed into cages containing materials that had been smoked for a week.
Dermal excision wound model
C57BL/6 mice exposed and not exposed to THS (Jackson Laboratories) were housed in the UCR vivarium and were treated following protocols approved by the IACUC. The wounds were performed as we previously described . Briefly, dorsal hair was removed using clippers and nair on control and THS-exposed mice. 24 h later, full thickness excision wounds were performed on the dorsum of the mice using a 7 mm biopsy punch (Acuderm). Wound tissues were collected at various time points following injury using a 10 mm diameter biopsy punch.
Tissue extracts were prepared as previously described by us .
Superoxide dismutase activity assay
Total tissue superoxide dismutase (SOD) activity was measured using a commercially available kit (Cayman Chemical) that measures all three types of SOD (Cu/Zn-, Mn- and EC-SOD). One unit of SOD is defined as the amount of enzyme needed to cause 50% dismutation of the superoxide radical. The SOD activities of the samples were calculated from the linear regression of a standard curve that was determined using the SOD activity of bovine erythrocytes at various concentrations run under the same conditions. The SOD activity was expressed as unit/ml of tissue extract.
Hydrogen peroxide activity assay
Tissue hydrogen peroxide (H2O2) levels were measured using a commercially available kit (Cell Technology) that utilizes a non-fluorescent detection reagent. The assay is based on the peroxidase-catalysed oxidation by H2O2 of the nonfluorescent substrate 10-acetyl-3,7-dihydroxyphenoxazine to a fluorescent resorufin. Fluorescent intensities were measured at 530 nm (excitation)/590 nm (emission) using a Victor 2 microplate reader. The levels of H2O2 in the supernatants were derived from a seven-point standard curve generated with known concentrations of H2O2.
Catalase activity assay
Tissue catalase activity was measured using a commercially available kit (Cayman Chemical). The enzyme assay for catalase is based on the peroxidatic function of catalase with methanol to produce formaldehyde in the presence of an optimal concentration of H2O2. The formaldehyde produced was measured spectrophotometrically with 4-amino-3-hydrazino-5-mercapto-1,2,4-triazole (purpald) as the chromogen at 540 nm in a 96-well plate. The catalase activity was expressed as nmol/min per ml of tissue extract.
Glutathione peroxidase activity assay
Tissue glutathione peroxidase (GPx) activity was measured using a commercially available kit (Cayman Chemical). The activity was measured indirectly by a coupled reaction with glutathione reductase (GR). GPx reduces H2O2 to H2O and in the process oxidized glutathione (GSSG) is produced that in turn is recycled to its reduced state by GR and NADPH. Furthermore, oxidation of NADPH to NADP+ is accompanied by a decrease in absorbance at 340 nm. Under conditions in which GPx activity is rate limiting, the rate of decrease in the absorbance measured at 340 nm, in a 96-well plate at 1-min interval for a total of 5 min using a Victor 2 microplate reader, is directly proportional to the GPx activity of the sample. GPx activity was expressed as nmol/min per ml of tissue extract.
Lipid peroxidation assay using thiobarbituric acid reactive substances
Tissue thiobarbituric acid reactive substances (TBARS) were measured by using a commercially available kit (Cell Biolabs). Lipid peroxidation results in unstable lipid peroxides that further decompose into natural byproducts such as malondialdehyde (MDA) and 4-hydroxynonenal (4-HNE). MDA forms adducts with TBARS in a 1:2 proportion. These adducts were measured fluorometrically at an excitation of 540 nm and emission at 590 nm. TBARS levels were then calculated in μM by comparison with a predetermined MDA standard curve.
Tissue nitrotyrosine levels were measured by using a commercially available kit (Cell Biolabs). The measurements are based on a competitive enzyme immunoassay. The tissue sample or nitrated BSA were bound to an anti-nitrotyrosine antibody, followed by an horseradish peroxidase conjugated secondary antibody and enzyme substrate. The absorbance was measured spectrophotometrically at 412 nm and the nitrotyrosine content in the unknown sample was then determined by comparing with a standard curve that was prepared from predetermined nitrated BSA standards.
Isolation of DNA and 8-hydroxy-2-deoxy guanosine analysis
Tissue DNA was extracted by using a commercially available kit (Qiagen). Eluted DNA was digested using nuclease P1 and the pH adjusted to 7.5–8.5 using 1 M Tris. The DNA was incubated for 30 min at 37°C with 1 unit of alkaline phosphatase per 100 μg of DNA and then boiled for 10 min. The 8-hydroxy-2-deoxy guanosine (8-OH-dG) DNA damage assay was performed by using a commercially available kit (Cayman Chemical). The measurements are based on a competitive enzyme immunoassay between 8-OH-dG and an 8-OH-dG-acetylcholinesterase (AChE) conjugate (8-OH-dG tracer) with a limited amount of 8-OH-dG monoclonal antibody. After conjugation, Ellman's reagent (used to quantify the number or concentration of thiol groups) was used as a developing agent and read spectrophotometrically at 412 nm. The intensity measured was proportional to the amount of 8-OH-dG that was expressed in pg/ml.
Large Amplicon quantitative PCR assay
For preparation of DNA, mouse skin tissues were minced and placed in a microfuge tube containing 0.5 ml lysis buffer (100 mM Tris/HCl pH 8.5, 5 mM EDTA, 0.2% SDS, 200 mM NaCl). Proteinase K was added at a final concentration of 100 mg/ml and incubated at 55°C overnight. The lysate was extracted twice with an equal volume of phenol:chloroform:isoamyl alcohol (25:24:1, v/v) (GibcoBRL). DNA in the lysate was precipitated by addition of an equal volume of propan-2-ol. The DNA precipitates were solubilized in TE buffer (10 mM Tris/HCl, pH 8.0 and 1 mM EDTA). DNA purity and concentration were measured using a NanoDrop™ spectrophotometer (Thermo Scientific). Finally, DNA was digested with Escherichia coli formamidopyrimidine DNA glycosylase (Fpg) to cleave oxidized DNA bases and induce strand breaks with its associated AP lyase activity. To examine the formation of oxidative DNA damage in two mouse genes, polymerase β (polβ) and β-globin, large Amplicon quantitative PCR (LA-qPCR) was performed as described earlier [17,18]. Briefly, a 6.5-kb long region of polβ was amplified using the LongAmp Taq DNA polymerase (New England Biolabs) with the following primers: 5′-TATCTCTCTTCCTCTTCACTTCTCCCC-3′ and 5′-CGTGATGCCGCCGTTGAGGGTCTCC-3′. Preliminary experiments were carried out to ensure the linearity of PCR amplification with respect to the number of cycles and DNA concentration. A 250-bp fragment from polβ was also amplified for normalization of the data obtained with the long fragment using the primers: 5′-TATGGACCCCCATGAGGAACA-3′ and 5′-AACCGTCGGCTAAAGACGTG-3′. Likewise, for the mouse β-globin gene, a long fragment of 8.7 kb was amplified using the primers 5′-TTGAGACTGTGATTGGCAATGCCT-3′ and 5′-CCTTTAATGCCCATCCCGGACT-3′, and normalized with a 200-bp short fragment using 5′-ACACTACTCAGAGTGAGACCCA-3′ and 5′-ATACCCAAT-GCTGGCTCCTG-3′. All of the amplified products were resolved and visualized using agarose gel electrophoresis and quantified using the VersaDoc system (Bio-Rad Laboratories). The relative amplification was normalized to the control values.
Fluorescence activated cell sorting analysis
All anti-mouse antibodies for fluorescence activated cell sorting (FACS) and OneComp eBeads were purchased from eBioscience: Blocking CD16/CD32, for monocytes (CD11c PE-eFluor®610, IgG Isoytpe control PE-eFluor®610, Ly-6C APC, IgG1 K Isoytpe control APC), for neutrophils [Ly-6G(Gr-1) PerCP-Cyanine5.5, IgG2b K isotype PerCP-Cyanine5.5, CD11b PE-Cyanine7, IgG1 K Isoytpe control PE-Cyanine7], for macrophages (F4/80 FITC, IgG1 K isoytpe control FITC, CD11b PE-Cyanine7, IgG1 K Isoytpe control PE-Cyanine7). 7 mm wound with 3 mm surrounding tissue were collected at the given time points.
Wound tissues were then cut into small pieces with scissors and combined with 100 ml of collagenase/dispase (1 mg/ml) with RPMI medium and then incubated for 45 min at 37°C. The treated tissue/cell suspension was then passed through 18 and 20 gauge needles followed by straining using a 70 μm cell strainer (BD Bioscience). Cells were then washed with RPMI. The percoll gradient method was then used to separate neutrophils and monocytes/macrophages from the cell suspension. Collected cells were washed with RPMI and incubated in 10% purified CD16/CD32 in FACS buffer (10 μl) for 5 min to block non-specific binding to antibodies. Cells were then collected and re-suspended in 100 μl FACS buffer with 1 μl of each antibody and incubated on ice for 30 min and then followed by FACS analysis.
Skin and wound tissues were homogenized as described above. Millipore customized multiplex cytokine immunoassay was performed according to the manufacturers’ protocol (Millipore) using equilibrated protein tissue homogenate. Level of the different analytes was quantified using a LuminexTM 200 instrument (Millipore) by monitoring the fluorescence associated with the bead set.
In vivo angiogenesis assay
The angiogenesis assay was done as previously described by us . Briefly, the hair on the backs of control and THS-exposed mice was removed using Nair. The next day, they were injected subcutaneously with either insulin (Humulin) (1 μg/15 μl saline) or 15 μl saline using an insulin syringe. The injections were in well demarcated areas which were sufficiently separated to avoid interference. Both insulin and PBS were injected every 24 h for 4 days. The areas surrounding the injection sites were labelled using a permanent marker to ensure that the injections were always done at the same site. Skin samples from the injected areas were collected turned upside down and observed at day 5. Blood vessels were highlighted using ImageJ software and measured as previously described  (NIH).
Frozen sections were fixed in 4% paraformaldehyde for 20 min, rinsed with PBS, incubated in a solution of 0.1 M glycine in PBS for 20 min and blocked in 10% non-immune serum of the secondary antibody species in PBS containing 0.1% Triton X-100 for 30 min. Slides were then incubated for 2 h at room temperature with the primary antibodies in 1% BSA/PBS. The primary antibodies used, and respective dilutions, were the following: 1:200 FITC-labelled mouse anti-mouse α-smooth muscle actin (αSMA, Sigma–Aldrich) and 1:100 rabbit anti-mouse fibrinogen (Abcam). Sections were then washed three times with 0.1% BSA in PBS and incubated with 1:25 goat anti-rabbit secondary antibodies (Life Technologies), for 1 h at room temperature. After washing, sections were mounted in Vectashield containing DAPI (Vector Laboratories). Immunofluorescence was visualized and imaged using a Nikon Microphot-FXA fluorescence microscope with a Nikon DS-Fi1 digital camera and Nikon NIS-Elements software (Nikon Instruments).
We used Graphpad Instat Software and Sigmaplot Software. ANOVA was used to test significance of group differences between two or more groups. In experiments with only two groups, we used a Student's t test.
Studies performed in wound healing have focused on processes that occur later during healing. Reactive oxygen species (ROS) play the most crucial roles to put the healing on a path to repair or to impair healing. We have previously shown the importance of the early ROS markers and how they can lead to the prognosis of the development of impaired and chronic wounds [20,21]. Because production of ROS, release of cytokines/chemokines, influx of inflammatory cells, angiogenesis and collagen production, all occur during the early phases of healing, the parameters in the present study were focused on the first 7 days post-wounding.
Impaired healing in THS-exposed mouse wounds
Wounds of control and THS-exposed mice were imaged periodically to record wound closure. The control wounds closed in approximately 12 days whereas the THS mouse wounds took an average 25% more time to heal (Figures 1A and 1B). In addition, after day 5 the wounds became more irregularly shaped, less contracted and the edges became thicker. To gain insight into the cellular changes leading to these differences, we performed histological examination of the wound tissue and found that the migration tongue in the wounds of the THS-exposed mice was virtually non-existent, the dermal–epidermal interactions were disrupted and the granulation tissue was poorly developed (Figure 2A). To further understand how these impaired THS wounds develop, we stained the tissue for fibrilar collagen to evaluate quality and spatial arrangement. Using the Masson Trichrome staining, we found that by day 7 the THS-exposed mouse wounds had considerably less staining in the granulation tissue, both in intensity and in area (Figure 2B). To determine whether the reduction in collagen deposition and in contraction was in part due to reduced numbers of myofibroblast (high producers of collagen), we stained day 7 wound tissues for αSMA, a protein found in myofibroblasts. We found that there were significantly reduced numbers of myofibroblasts in the wounds of the THS-exposed mice when compared with the control mice, confirming deficiency in collagen deposition (Figure 2C). However, decreased collagen deposition has also been associated with presence of excessive proteases during healing. Matrix metalloproteinases (MMPs) are enzymes critical during different phases of wound healing, including keratinocyte migration that helps in reepithelization [22,23], and remodelling of the collagen in the granulation tissue . Therefore, we examined the levels of two important MMPs in this process, MMP2 and MMP9. We found that the levels of active MMP2 in the THS-exposed mouse wounds is increased significantly at day 1 but then decreases at days 3–7 (Figure 3A, arrowhead). Levels of active MMP9, on the other hand, are higher up to day 3 in THS-exposed wounds and then decrease quickly by day 5 and 7 (Figure 3A, arrowhead).
THS-exposed wounds have delayed wound closure
Collagen deposition is impaired in THS-exposed wounds
THS-exposed mouse wounds have a high proteolytic microenvironment
Tissue inhibitors of matrix metalloproteinases (TIMPs) are known to play critical roles in maintaining the balance of active and inactive MMPs during the wound healing process . TIMP-1 has been extensively shown to be associated with impaired and chronic wounds and its expression is widely observed during re-epithelialization near the leading edge of a normally healing wound, and it is also reduced when collagen is degraded in the granulation tissue [26–28]. Therefore, we examined the levels of TIMP-1 in the wounds of THS-exposed mice and found that the levels of TIMP-1 were significantly reduced, suggesting that in these wounds the proteases are not kept in check (Figure 3B).
Because our results on the MMPs did not appear to justify the destruction we see in the granulation tissue of THS-exposed mice, we examined the levels of another enzyme, elastase, that is also known to play a critical role in wound healing [29,30]. Elastin, the major target of elastase, confers elasticity on tissues. Elastase activity in the THS-exposed mouse wounds was significantly increased when compared with that in the wounds of the control mice (Figure 3C), suggesting that the granulation tissue of THS-exposed wound has lower elastic properties than normal and therefore might therefore be susceptible to reopening.
THS wounds have redox imbalance during wound healing
One of the important parameters in wounds with impaired healing is the presence of an imbalanced redox state. To determine whether the impaired wounds of THS-exposed mice have altered oxidative stress levels, we examined several components of the oxidative stress cycle. SOD activity, an enzyme that is critical in the dismutation of superoxide anions to H2O2, was already greatly significantly elevated by 4 h post-wounding when compared with the control wounds (Figure 4A). These levels remained high during the first 7 days post-wounding. Consistent with the SOD activity, H2O2 levels were also significantly elevated in the THS-exposed mouse wounds (Figure 4B). Furthermore, we observed that unwounded skin in THS-exposed mice already has levels of H2O2 that are significantly higher than those in the skin of control mice, suggesting that the toxicants of THS affect the normal physiology of the skin and elevate stress levels even in the absence of wounding (Figure 4B). Catalase activity, one of the critical antioxidant enzymes that processes H2O2 into H2O and O2 is not changed over the course of healing (Figure 4C). However, the activity of the other critical antioxidant enzyme, GPx, was greatly decreased during the first 3 days post-wounding (Figure 4D), suggesting significant reduction in antioxidant capacity in the wound tissue of THS-exposed mice.
Oxidative stress is elevated in THS-exposed mouse wounds
The presence of oxidative stress can lead to significant molecular damage. We found that in THS-exposed mouse wounds there was a significant increase in the levels of MDA, a marker for lipid peroxidation, particularly after 5 days post-wounding (Figure 5A). This damage to lipids was particularly evident after 5 days post-wounding. Damage to proteins can occur by modification of tyrosine residues converting them to 3-nitrotyrosine (3-NT). The levels of 3-NT in the THS-exposed mouse wounds were significantly higher after 2 days post-wounding (Figure 5B). Another deleterious effect due to excessive oxidative stress occurs on DNA. 8-Hydroxylation of the guanine base (8-OH-dG) on DNA in the presence of oxidative stress, results in DNA damage. We show that the levels of 8-OH-dG in the THS-exposed mouse wounds were extremely high and much higher than the control during the entire period of healing (Figure 5C). Interestingly, we observed that the detrimental effects of oxidative stress observed in the wounds were also seen in unwounded skin of the THS-exposed mice (Figures 5A–5C).
Stress related tissue damage occurs in skin and wounds of THS-exposed mice
To further confirm the genotoxic effect of THS we measured the oxidative stress-induced DNA damage using LA-qPCR. Genomic DNA was extracted as described in Research Design and Methods followed by the digestion with the E. coli formamidopyrimidine DNA glycosylase (Fpg) enzyme that acts on 8-oxoG and other oxidized lesions, and generates strand breaks at their sites. LA-qPCR is highly sensitive to oxidative DNA damage when coupled with the repair enzyme Fpg that is specific for incision of oxidized purine lesions. We amplified both the 6.5-kb DNA polβ and 8.7-kb β-globin gene fragments with LA-qPCR. Short fragments (250–200 bp) in the same corresponding genes were also amplified under identical conditions and used for normalization of the long fragments. The intensity of the short fragments remained almost unchanged in both THS-exposed and control mouse samples, as much shorter regions of DNA would be relatively independent of oxidative DNA damage (Figures 5D and 5E). However, when compared with the controls, THS caused a marked reduction of the long amplification products for both genes, indicating the formation of oxidized DNA lesions. It can also be seen that the degree of such reduction varies in different mouse DNA samples, suggesting inter-individual differences in the formation and repair of oxidative DNA lesions. These results indicate that THS exposure causes increased levels of oxidative DNA damage in the mouse skin and in wounds.
THS exposure causes a dysfunctional immune response upon wounding
Proper wound healing results from the synchronized yet overlapping infiltration of inflammatory cells to the wound site. Disruption in the sequential entry of immune cells to sites of injury can lead to the impairment of wound closure and result in improper healing. The presence of inflammatory cells in the tissue was evaluated by flow cytometry. Singlet cell populations were defined by plotting forward scattered light height (FSC-H) compared with FSC area (Figure 6A and Supplementary Figure S1). The first cells to infiltrate the wound are the neutrophils which come into the tissue to kill bacteria. The number of neutrophils, defined by the markers Ly6G and CD11b (Figure 6A), was very similar in both control and THS-exposed mouse wounds (Figure 6B).
THS-exposure delays inflammatory response
The wave of neutrophils is followed by monocytes. Monocytes play central roles in response to injury, inflammation and host defense because they differentiate into macrophages, which are critical for wound repair. We found that the presence of monocytes at the wound site, assessed by markers CD11c and Ly6C, was significantly delayed in the THS-exposed mouse wounds. It peaked at day 3 rather than day 1 post-wounding and decreased to control levels by day 5 much like the control (Figure 6C and Supplementary Figure S2). Macrophages remove and clear neutrophils and bacteria and elicit chemokines to recruit lymphocytes, and also stimulate the migration of fibroblast, epithelial and endothelial cells for the later phases of healing and remodelling. Using the macrophage markers F4/80 and CD11b, we found that, in the THS-exposed mice, the number of these leucocytes was significantly reduced during the first 5 days post-wounding (Figures 6A and 6D, Supplementary Figure S2), indicating that most monocytes do not differentiate into macrophages. These results strongly suggest that the microenvironment of the THS-exposed mouse wounds does not promote monocyte to macrophage differentiation, a critical aspect of the inflammatory response and that the macrophage inflammatory phase is defective.
THS-exposed wounds display impaired production of chemokines
Chemotaxis of inflammatory and immune cells is heavily dependent on the expression and appropriate production of chemokines and cytokines during the course of wound healing [31,32]. To determine whether the altered number of inflammatory cells attracted to the wound site was in part due to an imbalance in the presence of chemokines/cytokine, we performed multiplex array cytokine analysis (Figure 7). Although KC, a major chemokine that chemoattracts neutrophils is lower in the THS-exposed mice (Figure 7A), another chemokine which also attracts neutrophils, LIX is at normal levels (Figure 7B), potentially explaining why we see no difference in the number of neutrophils in the THS-exposed mouse and those of the control (Figure 6B). However, neutrophil maturation is probably compromised because granulocyte colony stimulating factor (GCSF), a cytokine that is very important in neutrophil maturation and function, is significantly decreased in THS-exposed mouse wounds (Figure 7C). We also found that a major monocyte recruiting chemokine, monocyte chemoattractant protein (MCP-1/CCL2) is reduced in THS-exposed mice (Figure 7D). However, the other major chemokine that attracts monocytes, macrophage inflammatory protein 1α (MIP-1α/CCL3) is found at normal levels, therefore together they could contribute for the elevated presence of monocytes in the wound (Figure 7E) albeit with a delayed response because MIP-1α/CXCL2 peaks at day 3 much like the monocytes (Figure 6C). We also found that the number of these monocytes that differentiates into macrophages is significantly decreased in the THS-exposed mouse wounds which correlates well with the lower levels of leukaemia inhibitory factor (LIF), a cytokine that has the ability to induce the terminal differentiation of myeloid cells such as monocyte/macrophages (Figure 7F). In addition, we found that the chemokine MIG/CXCL9 levels are higher in THS-exposed mouse wounds very early in the healing process (Figure 7G). This chemokine stimulates cellular and molecular mechanisms involved in wound tissue resolution and remodelling. Therefore, presence at high levels early in the wound is detrimental to healing.
Chemokine and cytokine profile in THS-exposed mouse wounds
THS-exposed mouse wounds have leaky blood vessels
Angiogenesis is an important component in providing oxygen and nutrients to promote proper repair processes during wound healing. We first determined whether THS-exposed mice responded normally to the angiogenic factor insulin  by performing an in vivo angiogenesis assay. We found that injection of insulin subcutaneously resulted in the development of many sprouting microvessels in the skin of control mice but that insulin was unable to elicit sprouting of the blood vessels in the THS-exposed mouse skin (Figures 8A and 8B). Furthermore, we observed that the THS-exposed mouse skin blood vessels were leaky (Figure 8A). To determine whether blood vessel leakage was also observed during the wound healing process, we immunolabelled tissue sections from day 7 wounds for fibrinogen, a protein that is involved in clot formation and found that fibrinogen was present in the periphery of the blood vessels of THS-exposed mouse wounds, suggesting microvessel leakage (Figure 8C).
THS-exposure inhibits angiogenesis
First hand smoke (FHS) and second hand smoke (SHS) have previously been reported to cause impaired healing [34–37]. Doctors avoid performing surgery on patients who smoke due to known complications in clot formation, healing times and the quality of healing [12,38]. However, the effects of THS on healing are not well known. This is particularly important because THS is a stealth ‘toxin’ that permeates the homes of smokers and affects infants, children, spouses of smokers and elderly living in that environment. We have recently shown that THS has deleterious effects on multiple organs and also delays healing of skin wounds . However, the underlying cell and molecular mechanisms that lead to THS-induced impaired healing are not known. In the present study we show that THS delays wound healing by delaying wound closure, increasing oxidative stress and its associated tissue damages, causing an imbalance in chemokines and cytokines that result in decreasing/delaying the inflammatory response, decreasing collagen deposition, decreasing angiogenesis and increasing blood vessel leakage and clot formation.
After wounding, re-epithelialization was followed over time and the area of the wounds measured. We found that THS delays wound closure; improper migration of the epithelial tongue and defective dermal epidermal interaction in conjunction with poor granulation tissue formation could be the reason for improper healing of the exposed mice. The epidermal cells do not form the normal epidermal tongue that moves over the granulation tissue to cover the wound and does not form adhesions with the underlying dermis as well as proper deposition of the basement membrane to establish firm connections.
Increased collagen deposition is critical for granulation tissue formation, and remodelling plays a critical role in providing the structural integrity to the healing tissue. The decrease in the level of collagen deposition observed at day 7 post-injury in the THS-exposed mouse wounds correlates with the significantly reduced number of myofibroblasts, cells that effectively produce fibrillar collagen . Epithelial and stromal cells vital to the regenerative process in the wounded tissue have been shown to produce multiple MMPs [40,41]. Especially MMP2 and 9 have been shown to be localized to the epithelial–stromal interface behind the migrating epithelial cells, that suggest their role in remodelling of the stroma and reorganization of the basement membrane . The levels of MMP2 and 9 which are very important in reepithelization and tissue remodelling [31,32], are elevated very early on and reduced at later time points in the wound tissue of THS-expose mice suggesting its role in impaired wound closure. Furthermore, lower levels of active MMP9, especially the reduction at day 7, correlate with delayed reepithelization and disorderly collagen fibrillogenesis [14,43]. Proper function of MMPs is, therefore, critical to correct tissue restoration processes.
TIMP-1, one of the four-member family of TIMPs, was significantly reduced in the wounds of THS-exposed mice. Previous studies have reported the family of TIMPs to have inhibitory actions on multiple MMPs . However, the low levels of active MMP2 and 9 did not correlate with the lower levels of TIMP-1 we observed, suggesting that other TIMPs not included in our experiments might explain the low levels of MMP2 and 9. Furthermore, it is also possible that the increases in ROS levels we observe during the inflammatory phase of the THS-exposed mouse wounds could inhibit the pro-peptide cleavage by reacting with cysteine in the pro-peptide [44–46]. Additionally, the overall increase in elastase suggest a more proteolytic environment in the granulation tissue , which leads to elastin breakdown, a protein that is critical for elasticity of the healed tissue [47,48]. This might be one of the reasons why post-surgical wounds reopen in many smokers [49–51].
One of the primary responses during wound healing is the production of ROS; the imbalance in the levels of ROS is a critical factor in causing impaired wounds . Significantly elevated SOD activity gives rise to increases in H2O2 that if not broken down to H2O and O2 by the antioxidant enzymes, GPx and catalase, lead to increase in oxidative stress . Both superoxide anion and H2O2 can metabolize to result in peroxynitrite or hydroxyl radicals respectively which lead to oxidative damage. Molecules such as lipids, proteins and DNA are prime targets [20,54].
The significant increases in tissue damage observed in unwounded skin suggests that the THS toxins deposited on the skin lead to detrimental effects and also predispose the tissue to initiate a course of improper healing after injury. In particular, the DNA damage observed in the skin of the THS-exposed mice was very high suggesting that the cells are under significant stress from the smoke toxicants . These results indicate that THS exposure causes increased levels of oxidative DNA damage in the mouse skin tissues, which is consistent with the observation that the skin from the same exposed mice exhibit higher levels of 8-OH-dG detected by both the TBARS and LA-qPCR assays .
The occurrence of sequential but overlapping inflammatory cascade is important for proper healing. The altered levels of monocytes and macrophages in the wound tissue of the THS-exposed mice suggest a disruption of the balance of influx of inflammatory cells. Furthermore, it has been suggested that macrophages in smokers, display defective capacity to kill . Although we do see a spike in monocyte infiltration at day 3 post-wounding in the THS-exposed mouse wounds, the overall level of macrophages was significantly suppressed suggesting impaired maturation of monocytes to macrophages. These lower levels of macrophages could result in impairment of the clearance of dead neutrophils and of debris from the wound tissue and in altered production of factors that stimulate proper healing.
Because of the altered leucocyte levels in early wound healing we examined the chemokines and cytokines that lead to chemotaxis and maturation/function of these inflammatory cells. Although the levels of KC, chemoattractant for neutrophils [57,58], were reduced in the THS-exposed mouse wounds the levels of LIX/ENA-78, another chemokine that chemoattracts neutrophils was present at normal levels . This explains the presence of similar numbers of neutrophils in the THS-exposed mouse wounds. However, we observed lower levels of GCSF, which not only is a key regulator of neutrophil production  but also plays an essential role in neutrophil trafficking . This suggests that although we detect similar numbers of neutrophils infiltrating the wound tissue in both THS-exposed and control mice, these neutrophils may not be fully functional. GCSF, also directly affects monocytes and modulates their cytokine secretion . In the case of the monocytes we found that MCP-1/CCL2 was decreased at day 1 post-wounding which explains the delayed infiltration of monocytes we see . However, the fact that the levels of MIP-1α, another chemokine that attracts monocytes, are virtually the same in THS-exposed and in control mice and that the monocytes peak at day 3 post-wounding suggest that this chemokine plays a major role in the monocyte chemotaxis to the wound. Furthermore, we show that there is an overall reduced level of the interleukin 6 class protein LIF; this cytokine is critical for monocyte maturation into macrophage and for proper formation of the granulation tissue . Another chemokine that plays a very important role in the resolution and remodelling of the healing process is MIG/CXCL9. In the THS-exposed mouse wounds  the levels of this chemokine are lower after 5 days post-wounding suggesting poor remodelling of the wound tissue.
Angiogenesis is required for proper nutrition and oxygenation of the wound tissue. Our observation that THS exposure results in the development of fewer blood vessels in response to an angiogenic molecule indicates that there is a deficiency in the ability of the endothelial cells to migrate to the wound site and form vessels. The lack of angiogenesis could also be due to the inability of the endothelial cells to adhere to each other and form tubes . In support of the impaired adhesion are our findings that the vessels formed in the wounds of THS-exposed mice are leaky.
We delineate for the first time the mechanisms involved in THS-induced poor wound healing. They involve delayed rate of healing, increased oxidative stress, impaired inflammation, improper cytokine/chemokine balance and abnormal angiogenesis, driving the pathogenesis of impaired wound healing. These findings are also instrumental to understanding both FHS- and SHS-induced impaired healing. Taken together, our findings provide powerful insight into the role of tobacco smoke in wound healing and in particular targeting patients previously exposed to THS and undergoing surgery with substantial risk of developing difficult-to-treat surgical wounds.
Sandeep Dhall and Manuela Martins-Green conceived and designed the experiments. Sandeep Dhall, Raquelle Alamat, Anthony Castro, Altaf Sarker, Bo Hang, Jian-Hua Mao and Manuela Martins-Green performed the experiments. Sandeep Dhall, Alex Chan and Manuela Martins-Green analysed the data. Manuela Martins-Green and Bo Hang contributed reagents/materials/analysis tools. Sandeep Dhall and Manuela Martins-Green wrote the manuscript.
We thank Yagnika Patel for helping with wound area determination using ImageJ.
This work was supported by the Tobacco Research Disease Related Program (TRDRP) [grant numbers 22RT-0121 and 23DT-0103 (to M.M.-G.)].
fluorescence activated cell sorting
granulocyte colony stimulating factor
Institutional Animal Care and Use Committee
large Amplicon quantitative PCR
leukaemia inhibitory factor
monocyte chemoattractant protein
macrophage inflammatory protein 1α
reactive oxygen species
second hand smoke
α-smooth muscle actin
thiobarbituric acid reactive substances
third hand smoke
tissue inhibitors of matrix metalloproteinase