Effect of nicotine on Staphylococcus aureus biofilm formation and virulence factors

Staphylococcus aureus is a common pathogen in chronic rhinosinusitis (CRS) patients, the pathogenesis of which involves the ability to form biofilms and produce various virulence factors. Tobacco smoke, another risk factor of CRS, facilitates S. aureus biofilm formation; however, the mechanisms involved are unclear. Here, we studied the effect of nicotine on S. aureus biofilm formation and the expression of virulence-related genes. S. aureus strains isolated from CRS patients and a USA300 strain were treated with nicotine or were untreated (control). Nicotine-treated S. aureus strains showed dose-dependent increases in biofilm formation, lower virulence, enhanced initial attachment, increased extracellular DNA release, and a higher autolysis rate, involving dysregulation of the accessory gene regulator (Agr) quorum-sensing system. Consequently, the expression of autolysis-related genes lytN and atlA, and the percentage of dead cells in biofilms was increased. However, the expression of virulence-related genes, including hla, hlb, pvl, nuc, ssp, spa, sigB, coa, and crtN was downregulated and there was reduced bacterial invasion of A549 human alveolar epithelial cells. The results of this study indicate that nicotine treatment enhances S. aureus biofilm formation by promoting initial attachment and extracellular DNA release but inhibits the virulence of this bacterium.

Nicotine treatment strengthens S. aureus initial attachment but has no effect on poly-Nacetylglucosamine polysaccharide synthesis. Bacterial attachment is the initial step of biofilm formation, and in this regard, our results indicated that after treatment with 2 mg/mL nicotine, a larger number of cells of USA300 strain FPR3757 had adhered to the bottom of polyethylene wells (3.44 ± 0.36 × 10 5 , n = 3), compared with the untreated control cells (2.25 ± 0.57 × 10 5 , n = 3) (p < 0.05) ( Fig. 2A). Fibronectin-binding protein A (FnbA), which is a key surface-attached proteins and ECM-binding protein homologue (Ebh), a giant cell wall-related protein associated with virulence, are both key surface proteins, and therefore we examined the transcription level of these genes during the initial period of biofilm formation (2 h) in response to nicotine treatment. The results indicated that expression of the fnbA gene was increased whereas that of ebh was decreased after 2 h compared with the control group. (Fig. 2B) Furthermore, to examine the effects of nicotine on bacterial intercellular adhesion, we determined the synthesis of poly-N-acetylglucosamine polysaccharide (PIA) by CLSM. Observations indicated that the levels of PIA did not differ significantly between the nicotine-treated and non-nicotine-treated groups (Fig. 2C).
Nicotine promotes the autolysis of S. aureus. A Triton X-100-induced autolysis assay was performed to determine the effect of nicotine on autolysis of the S. aureus USA300 strain. As shown in Fig. 3A, the bacterial cultures treated with 2 mg/mL nicotine exhibited a significantly higher rate of autolysis than the control group culture. Consistently, the expression of autolysis-related genes, including lytN and atlA, were found to be upregulated in the nicotine-treatment group (Fig. 3B).
Next, we performed CLSM to examine the 24 h biofilms of nicotine-treated USA300 cells, using LIVE/DEAD staining, the viable cells of which were stained by green fluorescence (SYTO9) and dead cells by red fluorescence (PI). A merged image of the two staining patterns is shown in Fig. 4A. The ratio of red fluorescence intensity to total fluorescence intensity was 41.94% ± 2.00% and 26.39% ± 4.09% (n = 3, p < 0.05) in the nicotine-treatment group and the control group, respectively; while the overall green fluorescence intensity in the two groups was 980.3 ± 12.34 and 819.3 ± 33.35 (p < 0.05), respectively (Fig. 4B).
Nicotine facilitates eDNA release by S. aureus. The upregulation of autolysis is invariably associated with a higher release of eDNA. Thus, we evaluated the eDNA in 24 h S. aureus biofilms in the presence and absence of nicotine by staining with PI. The results accordingly revealed that the amount of eDNA (indicated by fluorescence intensities at a wavelength of 610 nm per OD 600 ) was higher in the nicotine treatment group (15.95 ± 1.64, n = 3) compared with the control group (10.59 ± 0.51, n = 3) (p < 0.05) (Fig. 5).

DNase I and Proteinase K inhibit nicotine-induced biofilm formation.
To evaluate whether enhanced biofilm formation is associated with an increase in eDNA release, we observed the effect of DNase I on biofilm formation in the USA300 strain with or without 2 mg/mL nicotine treatment. The results showing that OD 570 was reduced from 1.506 ± 0.04 to 0.450 ± 0.07 in the nicotine-treated group and from 0.720 ± 0.14 to 0.328 ± 0.05 in untreated group, indicating that nicotine-induced biofilm formation is sensitive to DNase I (25 U/well) (Fig. 6A).
Furthermore, we also investigated whether the enhanced biofilm formation induced by nicotine can be inhibited by Proteinase K (2 μg/mL), and accordingly found that Proteinase K has a suppressive effect on biofilm formation, with OD 570 values in the nicotine treated group decreasing from 1.300 ± 0.06 to 0.393 ± 0.02 (n = 3, p < 0.05), and those in the control group decreasing from 0.721 ± 0.14 to 0.316 ± 0.08 (n = 3, p < 0.05) (Fig. 6B) www.nature.com/scientificreports www.nature.com/scientificreports/ Nicotine treatment suppresses S. aureus virulence. As one of the major human pathogens, S. aureus can produce numerous virulence factors, including alpha hemolysin (hla) and beta hemolysin (hlb), responsible for hemolysis; nuclease (nuc), serine protease (ssp), and surface proteins A (spa), and genes responsible for pigmentation, including sigB, rsbU (global regulators), citZ, and crtN 19 . We examined the effects of nicotine on the expression levels of these virulence genes in the S. aureus USA300 strain using qRT-PCR. The results showed that in the nicotine treatment group, expression of the virulence-related genes hla, hlb, pvl, nuc, ssp, spa, and sigB was downregulated at both 12 h and 24 h, the expression of rsbU decreased at 12 h, whereas only the coagulase gene coa and the pigmentation-related gene crtN was downregulated at 24 h (Fig. 7A).
To investigate whether the altered gene expression patterns affected the virulence-related phenotypes, the hemolysis (24 h) and carotenoid pigmentation formation (24 h) of USA300 cells were investigated in the presence and absence of nicotine (2 mg/mL). As expected, the hemolysis and pigmentation phenotypes were altered upon treating with nicotine. A clear/complete hemolytic ring (β-hemolytic phenotype) was observed around USA300 . aureus strains were exposed to various concentrations of nicotine (0, 0.1, 0.2, 0.5, 1, 2 mg/mL). Crystal violet staining of biofilmassociated biomass revealed that the biofilm formation was increased in all strains in response to nicotine treatment. (B) A dose-dependent increase in biofilm formation was observed in USA300. Maximal biofilm formation was observed at 2 mg/mL. (C) Representatives of confocal images showed that denser biofilms were formed in the nicotine-treated group than in the control group. (D) The average thickness of biofilms in nicotine treatment group was 16.90 ± 0.66 μm (n = 3), while the thickness in the untreated group was 12.67 ± 0.91 μm (n = 3) (p < 0.05). (E) S. aureus USA300 strain FPR3757 growth curves under different concentrations of nicotine (0, 0.1, 0.2, 0.5, 1, 2 mg/mL) were determined by measuring the OD at a wavelength of 600 nm at 1-h intervals for 12 h. *P < 0.05 compared with untreated group. CI, clinical isolates.
colonies on blood agar plates, whereas a smaller β-hemolytic ring was formed on the blood agar plate containing nicotine (Fig. 7B). Compared with the control cells, USA300 cells cultured in TSB medium supplemented with 2 mg/mL nicotine for 24 h showed visually reduced pigment production, whose absorbance at OD 462 decreased from 0.263 ± 0.01 in the non-nicotine treated group to 0.119 ± 0.01 in the nicotine-treated group (n = 3, p < 0.05), thereby indicating that nicotine suppresses bacterial virulence (Fig. 7C).    www.nature.com/scientificreports www.nature.com/scientificreports/ Figure 6. DNase I and Proteinase K inhibit nicotine-induced biofilm formation. The biofilm formation of S. aureus USA300 strain was detected using a microtiter plate assay by measuring crystal violet stained biofilm at OD570. DNase I (25 U/well) and Proteinase K (2 ug/ml) was added to the well in both 2 mg/ml nicotine treatment group and control group. (A) Treated by DNaseI, the average thickness of biofilms varied from 1.506 ± 0.04 to 0.450 ± 0.07 in the nicotine-treated group (n = 3) and from 0.720 ± 0.14 to 0.328 ± 0.05 in untreated group. (n = 3) (B) Proteinase K(2 ug/ml) disrupted biofilm formation, in the nicotine treated group decreasing from 1.300 ± 0.06 to 0.393 ± 0.02 (n = 3), and in the control group decreasing from 0.721 ± 0.14 to 0.316 ± 0.08 (n = 3). *P < 0.05.

Figure 7.
Nicotine treatment suppresses S. aureus virulence. (A) Transcriptional levels of the virulence genes in S. aureus USA300 strain cultured in TSB and TSB supplemented with 2 mg/mL nicotine for 12 h/24 h were detected by qRT-PCR(n = 3). In the nicotine-treatment group, the expression of the virulence genes, hla, hlb, pvl, nuc, ssp, spa, sigB was downregulated at both 12 h and 24 h, while expression of the coagulase gene coa and the pigmentation-related gene crtN was downregulated at 24 h. (B) The S. aureus USA300 strain was inoculated on normal blood agar plates (left) and blood agar plates containing 2 mg/mL nicotine (right), respectively. After incubation at 37 °C for 24 h, a smaller β-hemolytic ring (β-hemolytic phenotype) was formed on the blood agar plate containing nicotine than the control. (C) Nicotine treatment suppressed carotenoid pigment formation. The S. aureus USA300 strain was cultured for 24 h in TSB (left) and TSB supplemented 2 mg/mL with nicotine (right) respectively. The pigment formed by bacterial cell pellets in nicotine group was reduced as compared to the control. Pigment production was quantified at OD 462 , decreasing from 0.263 ± 0.01 in the control group (n = 3) to 0.119 ± 0.01 in the nicotine-treated group (n = 3). These photos are representative of three independent experiment. (2019) 9:20243 | https://doi.org/10.1038/s41598-019-56627-0 www.nature.com/scientificreports www.nature.com/scientificreports/ investigated the capacity of S. aureus to invade A549 human alveolar epithelial cells. Our results showing CFUs of 3.360 ± 0.25 log 10 and 1.816 ± 0.08 log 10 (n = 3, p < 0.05) for the control and nicotine-treated groups intracellular respectively, indicated that nicotine inhibited the ability of these bacteria to invading A549 cells (Fig. 8).

Nicotine affects biofilm formation in S. aureus in an accessory gene regulator-dependent manner.
Previous studies have indicated that the activity of the accessory gene regulator (Agr) system appears to have an influence on S. aureus virulence and enhance biofilm formation by this bacterium [20][21][22] . In this study, we explored the expression of Agr system genes (agrA, agrB, agrC, and agrD) at 2 h and 24 h during the biofilm formation process in the presence or absence of nicotine. The results showed that expression of agrA, agrB, agrC, and agrD in the nicotine-treated group were lower than those in the control group, (n = 3, p < 0.05), and that the expression appeared to be both time-and dose-dependent. Figure 9A shows that gene expression gradually decreased from 0 h to 2 h and 2 h to 24 h. Furthermore, at 24 h, we found that the degree of reduction in the 2 mg/mL group was more pronounced than that in the 1 mg/mL group (Fig. 9B).
In addition, we investigated whether the strengthened initial attachment due to nicotine exposure was a result of Agr dysfunction. By observing the effect of nicotine on USA300 FPR3757 mutants harboring transposon insertions in the agrA and agrC genes, we found that nicotine-induced enhancement of the initial attachment stage was abrogated in these two mutant strains, indicating that the Agr system is involved in the nicotine-potentiated initial attachment process (Fig. 9C).

Discussion
Given that S. aureus is a frequent colonizer of the nasal mucosa and lower respiratory tract passages, it will be exposed to cigarette smoke during smoking. Previous studies have indicated that an extract of cigarette smoke induces biofilm formation in various bacterial species and that nicotine enhances biofilm formation by Streptococcus mutants, Streptococcus gordonii and Staphylococcus epidermidis 17,[23][24][25] . In the present study, we found that nicotine can enhance S. aureus biofilm formation capability in both clinical S. aureus strains and USA300 strain FPR3757. We hypothesize that the nicotine-induced enhancement of S. aureus biofilm formation may be a general phenomenon and that it might represent a protective reaction on the part of S. aureus to adverse environments, such as exposure to nicotine. Under adverse conditions, S. aureus develops a defense phenotype biofilm by altering gene expression, which results in the bacteria becoming embedded in a protective biofilm that renders them less susceptible to eradication.
In the current study, we found that nicotine treatment enhances S. aureus biofilm formation by increasing initial attachment and the release of extracellular DNA, although it also attenuates bacterial virulence, including inhibition of the expression of major virulence-related genes, reducing pigment production, and reducing bacterial invasion of epithelial cells, the latter of which is a pathological process related to activity of the Agr system.
Nicotine is one of the most toxic chemicals in tobacco. Huang et al. 17 mentioned that high levels of nicotine of up to 2.3 mg/mL are detected in the saliva of smokers. Although to date there have been no data published regarding nicotine concentrations in nasal mucus, several facts indicate that these concentrations could be similar to the levels of nicotine in saliva. Firstly, the nose is one of the most exposed organs to smoke and the nasal mucosa is one of the major routes via which nicotine is absorbed from tobacco 26 . Secondly, the nasal and oral cavities are interconnected, and are hence simultaneously exposed to nicotine during smoking. Thus, levels of nicotine in nasal mucus may reach concentrations in the range of that which we used in the present study (2 mg/mL) to investigate the underlying mechanism of nicotine-induced biofilm formation and virulence expression in vitro.
Biofilm formation proceeds in two steps, namely, an initial bacterial adhesion followed by cell proliferation and aggregation 27 . The essential step in the establishment of any Staphylococcus infection is the attachment to host www.nature.com/scientificreports www.nature.com/scientificreports/ tissues. We observed that a larger number of bacteria were attached to the polyethylene surface of culture plate wells in a nicotine-treated group than in a control group, thereby indicating that nicotine may enhance the initial step in S. aureus biofilm formation. These results are consistent with the findings of Cogo et al., who demonstrated that cotinine, the predominant nicotine metabolite, enhances the adhesion of Porphyromonas gingivalis to epithelial cell monolayers 28 . Surface proteins are the main components involved in biofilm formation, particularly during initial attachment 29,30 . FnbA is a well-known surface protein associated with the attachment stage of biofilm formation and Ebh is a giant protein found on the surface of S. aureus. In infected mice, ebh variants have been shown to be associated with the diminished virulence of S. aureus 31 . In response to nicotine treatment(2 h), we found that transcription levels of fnbA were increased, whereas the expression of ebh decreased. Furthermore, expression of the autolysis-related gene atlA, which is also associated with initial attachment, was also found to be increased.
The staphylococcal biofilm matrix has been reported to contain eDNA proteins, and polysaccharides, and we found that compared with the control group, larger amounts of eDNA were present in the matrix of S. aureus biofilms in the nicotine-treated group. Consistent with the findings of Kulkarni et al. 10 , we observed that the enhanced biofilm formation promoted by nicotine was suppressed by DNase I and proteinase K, thereby providing evidence that eDNA and proteins may play key roles in the nicotine-induced enhancement of initial attachment. Cell autolysis is a major source of eDNA in biofilms and a previous study has demonstrated that such autolysis could have a significant impact on biofilm formation 32,33 . Thus, we speculated that nicotine might increase S. aureus autolysis. However, results from previous studies to determine the effect of CS on bacterial autolysis are not always consistent: MacEachern 34 observed that exposure of S. aureus USA300 to CS resulted in an increased resistance to Triton-induced lysis, as a consequence of cell wall alteration, including changes in surface charge and hydrophobicity, Kulkarni et al. found DNase I inhibited CS-induced biofilm formation and observed 6-fold increase of autolysis associated gene, cidA, in USA300. We found that nicotine significantly increased S. aureus autolysis compared with the control group.
It has previously been shown that S. aureus biofilm formation is tightly regulated by a quorum sensing system, accessory gene regulator (Agr) system 20 , encoded by agrBDCA operon. The agrC/agrA two-component system negatively regulates S. aureus biofilm formation, partially by increasing extracellular protease production and inhibiting cell adhesion to surfaces. Our real-time PCR analysis revealed that transcription of the agr operon (agrA, agrB, agrC, and agrD) showed a marked decrease in expression in the nicotine treatment group (at 2 h and 24 h) depending on nicotine concentration and time exposure. Furthermore, we found that in USA300 mutants harboring a transposon insertion in either the agrC or agrA gene, the nicotine-induced increase in initial attachment was abolished. Taken together, our observations indicated that the augmentation of biofilm formation by nicotine may be directly or indirectly related to downregulation of the Agr system. www.nature.com/scientificreports www.nature.com/scientificreports/ As one of the major human pathogens, S. aureus can produce numerous virulence factors, whose expression is influenced by the microenvironment in the human body and is related with the process of bacterial pathogenesis. For example, it has been reported that α-toxin (α-hemolysin), a major cytolytic toxin expressed by the majority of S. aureus clinical isolates, is associated with the persistence of S. aureus cells in tissues and can also provoke cellular responses in nasal polyp cells from CRS patients, therefore, may play a role in the pathophysiology of CRS 35 . Although excess expression of virulence factors including exotoxins may enhance bacterial pathogenicity in the host, it can trigger augmented immune responses, which may lead to acute inflammations, hypersensitivity and then quick clearance of bacteria cells. Thus, the transcriptional regulation of virulence genes is vital for bacterial survival and adaption in the host. In S. aureus, the Agr system initiates RNAIII transcription and plays a pivotal role in the regulation of myriad virulence factors, especially exotoxin genes 36 . A recent extensive study conducted by Lacoma et al. 29 showed that the Agr system mediates CS-induced biofilm augmentation and CS-reduced toxin production. Based on the data in this study and in the literatures, we hypothesize that the presence of nicotine represses the expression of agrBDCA operon and thereby may also have an inhibitory effect to various virulence-related genes. In line with the expectation, the virulence genes including hla (alpha hemolysin), hlb (beta hemolysin), pvl (leucocidin), nuc (thermostable nuclease), ssp (serine protease), spa (protein A) and the genes responsible for pigmentation, including sigB, rsbU, citZ, and crtN 19 , showed decreased transcriptional levels in the nicotine treatment group at 12 h or 24 h. Consequently, nicotine treatment inhibited subsequent β-hemolysis on blood agar plates and carotenoid pigment production. Furthermore, using the A549 cell line to assess the invasion ability of S. aureus after nicotine treatment, we showed that nicotine decreased the number of intracellular bacteria. We speculate that on one hand, nicotine induces reduction in multiple bacterial virulence genes expression and may hinder the progress of acute infection. On the other hand, it enhances bacterial biofilm formation, which can protect bacteria from the host immune attack and therefore contribute to the persistence of this pathogen.
In conclusion, the results of this study provide further insights into the mechanisms whereby tobacco smoke induces S. aureus biofilm formation. We observed that nicotine can enhance biofilm formation in both clinical S. aureus strains obtained from CRS patients and the USA300 FPR 3757 strain. The altered biofilm formation was associated with enhanced initial cell attachment, elevated eDNA release, and enhanced S. aureus autolysis. Additionally, we found that nicotine repressed transcription of various virulence-related genes and inhibited β-hemolysis on blood agar plates, carotenoid pigment production and invasion in A549 cells by S. aureus. We hypothesize that the nicotine-induced reduction in bacterial virulence and enhancement of biofilm formation increases bacterial fitness and strengthens adaption to the harsh nasal environment, thereby may contribute to chronic infection.

Materials and Methods
Ethics statement. All procedures performed in studies involving human participants were in accordance with the ethical standards of the Institutional Review Board of Eye & ENT Hospital (reference number KJ2011-31) and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. In addition, informed consent was obtained from all participants.

Bacteria strains and culture media. This study was previously approved by the Ethics Committee on
Research of Eye & ENT Hospital. The clinical strains of S. aureus were collected during endoscopic surgery performed at the Shanghai Eye and ENT Hospital of Fudan University, and isolated from the middle meatus of CRS patients. The bacterial strains used in this study are listed in Table 1. S. aureus USA300 strain FPR3757 and mutants of this strain harboring transposon insertions in the icaA, ebh, agrA, and agrC genes were provided by Professor Ying Zhang at Johns Hopkins University. Tryptone soya broth (TSB; Oxoid, USA) was used for bacterial culture, and TSB supplemented with 1% glucose (TSBG) was used for bacterial biofilm formation experiments.
Detection of bacterial biofilm formation. Bacterial strains were grown in TSB with or without nicotine for 6 h at 37 °C to obtain bacteria in a mid-exponential phase. The cultures were then diluted 1:200 with TSBG supplemented with or without nicotine, and 200 μL of bacterial suspension was added to each well of a 96-well microplate and incubated at 37 °C for 24 h. To determine the effect of DNase I and proteinase K on biofilm formation, 5 μL DNase I (5 U/μL, Takara, Shanghai, China) and 2 μg/mL Proteinase K (Sango, Shanghai, China) were added to the wells. Thereafter, the wells were washed three times with phosphate-buffered saline (PBS) to remove unattached bacteria and then 200 μL of 100% methanol was added to each well to fix the attached cells at room temperature for 20 min. After removal of the methanol, the biofilms were air-dried and stained with 2% crystal violet at room temperature for 8 min. The wells were then washed with running tap water until the water was clear. Subsequently, 200 μL of 10% acetic acid was added to each well and incubated for 1 h. Finally, the www.nature.com/scientificreports www.nature.com/scientificreports/ stained biofilms were quantified by estimating the optical density (OD) at 570 nm using a microtiter-plate reader (DTX 880 Multimode Detector; Beckman Coulter, USA). The S. aureus USA300 strain and its isogenic icaA gene transposon insertion mutant were used as a biofilm-forming strain and a non-biofilm-forming control strain, respectively.
Bacterial growth curve determination. S. aureus growth curves were determined by measuring the OD at a wavelength of 600 nm using an automated growth curve detector (Biocreen C, Finland). Briefly, overnight cultures were diluted (1:200) and incubated in the presence of different nicotine concentration with shaking at 220 rpm and 37 °C. The OD 600 values of bacterial cultures were measured at 1h intervals for 12 h.

Observation of S. aureus biofilms by confocal laser-scanning microscopy (CLSM). S. aureus
biofilms were cultured in TSB with or without 2 mg/mL nicotine in glass-bottomed dishes, washed with three times with PBS, and then stained with LIVE/DEAD staining dye [1 μM of SYTO9 and 1 μM of propidium iodide (PI)] for 20 min. The biofilms were observed using a confocal laser-scanning microscope (Leica TCS SP8; Leica Microsystems, Germany) with a × 63 1.4-NA oil immersion objective. Fluorescence was quantified using Leica Application Suite 1.0 software (Leica Microsystem, Germany), and IMARIS 7.0 software (Bitplane, USA) was used to generate three-dimensional images of biofilms.
Initial bacterial attachment assays. Bacteria strains were cultured in TSB with or without 2 mg/mL nicotine to mid-exponential phase, and the bacterial cells were diluted with TSB to OD 600 = 0.1. The diluted culture was added to cell culture-treated 6-well polystyrene microtiter plates (Nunc, Denmark; 1 mL per well) and incubated at 37 °C for 2 h. Thereafter, the attached cells were washed gently with PBS (three times) and imaged. For each sample, six representative optical fields were randomly selected, and cells were counted using ImageJ software.

Detection of polysaccharide intercellular adhesin (PIA) by spectrofluorometric assay. S. aureus
biofilms were formed with or without 2 mg/mL nicotine in 96-well polystyrene microplates, washed gently with PBS, and then stained with 200 μL of 5 μg/mL wheat germ agglutinin (WGA)-Alexa Fluor 350 fluorescent conjugate (ThermoFisher, USA). After incubation at 4 °C for 2 h in the dark, the conjugate was removed, and the wells were gently washed three times with PBS. The plate was then air-dried at room temperature for 15 min, following which, 200 μL of 33% acetic acid was added to each well. The biofilms in the wells were incubated at 37 °C for 1 h, scraped thoroughly, and then 150 μL solution from each well was transferred to a solid black microplate (PerkinElmer, USA) for top reading using a Varioskan TM LUX microplate reader (ThermoFisher, USA; fluorescence at λ excitation = 346 nm and λ emission = 442 nm).
Triton X-100-induced bacterial autolysis assay. To determine the effect of nicotine on S. aureus autolysis, a Triton X-100-induced autolysis assay was performed. An overnight culture of the S. aureus USA300 strain FPR3757 was diluted 1:200 with TSB containing 1 M NaCl, and was then grown with or without 2 mg/mL nicotine to mid-log phase (OD 600 ~ 0.6). The cells were collected by centrifugation and washed twice in cold sterile PBS. After resuspension in the same volume of 0.05 M Tris-HCl (pH 7.2) containing 0.05% Triton X-100, the solution was incubated at 30 °C and OD 600 values were measured at 30-min intervals. The autolysis rate was determined by calculating the decline in OD 600 values. Data are represented as the mean ± SD of three independent experiments. Detection of extracellular DNA (eDNA). The quantity of eDNA in S. aureus cultures was determined using modifications of the methods described by Allesen-Holm et al. and Qin et al. 37,38 . Briefly, S. aureus strains were cultured overnight in TSB supplemented with or without 2 mg/mL nicotine, following which. the cultures were diluted with minimal growth medium (AB medium) supplemented with 0.5% glucose, 10% TSB, and 0.05 mM PI, to an OD 600 of 0.001. The diluted cultures were transferred to a 96-well microplate (200 μL per well) and incubated with or without 2 mg/mL nicotine at 37 °C for 24 h. The OD 600 values were measured using a microplate reader (BioRAD, USA). The fluorescence of PI-bound eDNA was measured at excitation/emission wavelengths of 535/610 nm using a Varioskan ™ LUX microplate reader. The relative amount of eDNA was expressed as the fluorescence intensity per OD 600 units.

RNA extraction.
Overnight cultures of USA300 were diluted 1:200 with TSB and incubated with or without 2 mg/mL nicotine at 37 °C to an OD 600 of 0.6 (mid-log phase). Bacterial cells were collected by centrifugation (4000 × g), washed three times with ice-cold saline, and homogenized using a Beadbeater-16 homogenizer (Biospec, USA). The bacterial RNA was purified using an RNeasy kit (Qiagen, Germany).
Quantitative real-time PCR analysis. The primers used in this study are listed in Table 2. DNase-treated RNA was reverse transcribed (Takara) to cDNA. All samples were prepared in triplicate and then quantified by qPCR using an ABI 7500 real-time PCR system (Applied Biosystems, USA) and SYBR Green I mixture (Takara). The data were normalized using gyrB as an internal control.
Analysis of hemolytic phenotype. The S. aureus USA300 strain was inoculated on blood agar plates (BioMérieux, China) and blood agar plates containing 2 mg/mL nicotine. The plates were cultured at 37 °C for 24 h, after which the hemolytic phenotype was observed and photographed.