Mucin Secretion Induced by Titanium Dioxide Nanoparticles

Nanoparticle (NP) exposure has been closely associated with the exacerbation and pathophysiology of many respiratory diseases such as Chronic Obstructive Pulmonary Disease (COPD) and asthma. Mucus hypersecretion and accumulation in the airway are major clinical manifestations commonly found in these diseases. Among a broad spectrum of NPs, titanium dioxide (TiO2), one of the PM10 components, is widely utilized in the nanoindustry for manufacturing and processing of various commercial products. Although TiO2 NPs have been shown to induce cellular nanotoxicity and emphysema-like symptoms, whether TiO2 NPs can directly induce mucus secretion from airway cells is currently unknown. Herein, we showed that TiO2 NPs (<75 nm) can directly stimulate mucin secretion from human bronchial ChaGo-K1 epithelial cells via a Ca2+ signaling mediated pathway. The amount of mucin secreted was quantified with enzyme-linked lectin assay (ELLA). The corresponding changes in cytosolic Ca2+ concentration were monitored with Rhod-2, a fluorescent Ca2+ dye. We found that TiO2 NP-evoked mucin secretion was a function of increasing intracellular Ca2+ concentration resulting from an extracellular Ca2+ influx via membrane Ca2+ channels and cytosolic ER Ca2+ release. The calcium-induced calcium release (CICR) mechanism played a major role in further amplifying the intracellular Ca2+ signal and in sustaining a cytosolic Ca2+ increase. This study provides a potential mechanistic link between airborne NPs and the pathoetiology of pulmonary diseases involving mucus hypersecretion.


Introduction
Many published reports have demonstrated the association between NP exposure and pulmonary morbidity and mortality [1,2,3]. The adverse effects induced by NPs seem to exacerbate clinical symptoms of pre-existing respiratory illnesses such as asthma, COPD and Cystic Fibrosis (CF) [1,2,3,4,5,6]. During NP exposure, individuals with respiratory diseases showed more incidences of bronchoconstriction, medication use, bronchial hyperreactivity and lung fibrosis [2,7]. TiO 2 NPs are widely used in the nanotechnology industry due to their vast array of applications that range from household commodities, such as components of paints and carpets, to personal products that include cosmetics, textiles, sunscreens and foods [8,9]. TiO 2 is also one of the PM10 components commonly found in industries or manufacturing plants involved in processing mineral ore rutile [10]. It has been reported that .50% of TiO 2 NP exposed workers had respiratory symptoms accompanied by reduction in pulmonary function [10,11]. Other reports have also indicated that inhalation of TiO 2 NPs can induce pulmonary inflammatory responses (characterized by neutrophil recruitment), epithelial cell death and increased permeability [2,9]. Furthermore, TiO 2 NPs have been shown to play a role in inducing epithelial fibroproliferative changes, stimulating goblet cell hyperplasia and in instigating emphysema-like (such as alveolar enlargement) damages in the lungs [2,10,12]. Overall, nanotoxicity induced by TiO 2 NP exposure in both the occupational and ambient environment presents a significant and realistic health concern.
The harmful effects of NPs on the respiratory system not only encompass cellular apoptosis/necrosis, but also mucus hyperproduction which is closely associated with the pathogenesis of pulmonary diseases that include asthma, COPD and CF [2,10,13]. In these chronic pulmonary diseases, mucus hypersecretion and accumulation may lead to recurrent episodes of chronic bacterial infections, limited airflow and chronic inflammatory responses [2,14,15]. However, whether TiO 2 NPs can directly induce mucin secretion has not been resolved.
Airway mucus plays a vital role in the constant clearance of inhaled pathogens and particulates. Mucus is a large, highly glycosylated protein consisting of an array of mucin peptides (apomucin) [14]. With their oligosaccharide sidegroups, such as sialic acid, sulfate, and carboxyl (COO 2 ), mucins are usually polyanionic in nature [16]. Mucin secretion is closely regulated by cytosolic Ca 2+ concentrations ([Ca 2+ ] C ) in various epithelial cells [17]. A rise in [Ca 2+ ] C is crucial for initiating a cascade of down stream events including the mobilization of granule-bound Ca 2+ , docking of the secretory granules, fusion of the plasma-granule membrane and the formation of secretory pores, therefore leading to the exocytosis of the mucin granules [18].
Agonist-induced opening of various Ca 2+ channels expressed on the cell membrane allows the influx of extracellular Ca 2+ , which may serve as the external Ca 2+ source [19]. The initial upsurge in the [Ca 2+ ] C is usually relayed by triggering a secondary wave of Ca 2+ propagation from internal stores, such as the ER [19,20,21,22]. Ryanodine receptors (RYRs) on the ER have multiple allosteric Ca 2+ binding sites responsible for triggering Ca 2+ -induced Ca 2+ release (CICR) into the cytosol [19,20,21,22]. The resultant increase in [Ca 2+ ] C could activate other cytosolic proteins and modulate secretion of mucin, hormones or various neurotransmitters [17,23,24].
NPs have been shown to disturb cellular functions by elevating intracellular Ca 2+ levels [25,26,27,28]. For example, ultrafine carbon black NPs can elicit Ca 2+ -dependent secretion through the activation of L-type voltage-gated Ca 2+ channels [25,26,28]. However, little is known regarding the intricate calcium signaling pathway regulating the exocytotic events of secretory products. In this study, we aim to investigate the mechanism through which TiO 2 NPs induce mucin secretion via a Ca 2+ signaling mediated pathway.

Culture of ChaGo-K1 cells
The human airway bronchial epithelial cell line ChaGo-K1, obtained from American Type Culture Collection (ATCC, Manassas, VA, USA), was used because it expresses MUC proteins and secretes mucin [29]. Cells were cultured in 15 cm cell culture plates (VWR, CA, USA) in RPMI 1640 medium (Invitrogen, CA, USA) supplemented with L-glutamine, 1% penicillin/streptomycin and 10% heat inactivated fetal bovine serum (FBS). Cultures were incubated in a humidified incubator at 37uC/5% CO 2 . Cell counts were performed using trypan blue (Sigma-Aldrich, MO, USA) exclusion and a Bright-Line haemocytometer.

Nanoparticles and characterization
A mixture of anatase and rutile forms of ultrafine titanium (IV) dioxide (,75 nm) (Sigma-Aldrich, MO, USA) was used in this study because this form has been shown to result in more severe cellular injuries [30,31]. The TiO 2 NPs have a surface area of 36 m 2 /g and the dispersion conductivity is 1040 mS/cm (information from Sigma). All NP samples were sonicated before usage. The concentrations used were 1 mg/ml, 0.75 mg/ml, 0.5 mg/ml, 0.25 mg/ml, 0.1 mg/ml, and 0.05 mg/ml. The range of concentrations used was consistent with the concentrations of TiO 2 NPs found in previous reports [30]. The TiO 2 NPs were reconstituted with Hanks' solution (Invitrogen, CA, USA) before being tested. The size of NPs was independently confirmed using homodyne dynamics laser scattering (DLS) as described in previous studies [32,33].

Cell preparation
Cells were seeded at 2610 5 cells per well in a 24-well plate, and incubated for 24 hrs in RPMI 1640 supplemented with 10% FBS. Following 24 hr incubation, the RPMI medium was removed from the cells and the culture was rinsed with Hanks' solution twice before use.

Measurements of cytosolic Ca 2+ concentrations induced by TiO 2 exposure
The cells were then loaded with a Rhod-2 AM dye (1 mM) (K d = 570 nM, l Ex = 552 nm and l Em = 581) (Invitrogen, CA, USA) for 45 minutes. After the dye loading, the cells were rinsed, incubated with either normal Hanks' or Ca 2+ -free Hanks' solution, and treated with the appropriate TiO 2 concentrations. All calcium signaling experiments were carried out on a thermoregulated stage at 37uC mounted on a Nikon microscope (Nikon Eclipse TE2000-U, Tokyo, Japan). ChaGo-K1 cells were incubated with cadmium chloride (200 mM; Sigma-Aldrich, MO, USA) to block the membrane Ca 2+ channels [34], followed by TiO 2 NP stimulation. To investigate the interaction between TiO 2 and membrane Ca 2+ channels, nifedipine (10 mM; Sigma-Aldrich, MO, USA), an Ltype Ca 2+ channel blocker [27], was added to ChaGo-K1 cells prior to the exposure of TiO 2 . Antioxidant N-acetylcysteine (NAC, 250 mM; Sigma-Aldrich, MO, USA) was also added to ChaGo-K1 cells to study the involvement of reactive oxygen species (ROS) [27,35], possibly generated as a result of TiO 2 stimulation, and the activation of Ca 2+ channels. Thapsigargin (100 nM; Sigma-Aldrich, MO, USA) [18] and ryanodine (100 mM; Sigma-Aldrich, MO, USA) were added separately to deplete the ER Ca 2+ content and to inhibit the CICR mechanism [20,21], correspondingly. These two blockers were utilized to investigate the contribution from the internal ER Ca 2+ pool.

Calcein dye leakage measurements
ChaGo-K1 cells were seeded at the density of 2610 5 cells per well in a 24-well plate and cultured for 24 hrs. TiO 2 NP prepared with calcein fluorescent dye (50 mM) (Invitrogen, CA, USA) in Hanks' solution was incubated with the cells for 5 minutes at 37uC. Calcein is a biological inert green-fluorescent molecule of a molecular mass of 623 Da and an estimated molecular radius of 0.6 nm [36]. TiO 2 NP solution containing the calcein dye was then removed and cells were rinsed twice with PBS to remove possible remnants of calcein dye in the extracellular solution. The cells were subsequently stained with a fluorescent nucleus dye, hoechst (10 mM) (Sigma-Aldrich, MO, USA), for 5 mintues at 37uC and thoroughly rinsed again [33]. Fresh Hanks' solution was added into each well before taking fluorescent images of calcein and hoechst loaded cells with a Nikon fluorescence microscope. A percentage of calcein loaded cells against total number of cells, as indicated by hoechst fluorescence, was calculated for each of the TiO 2 NP concentrations used in the experiment.

Mucin secretion and ELLA Preparation
The cells were seeded at 2610 5 cells per well in a 24-well plate and cultured for 24 hrs. ChaGo-K1 cells were then rinsed with PBS and treated with BAPTA-AM (Invitrogen, CA, USA), thapsigargin (Sigma-Aldrich, MO, USA) or ryanodine (Sigma-Aldrich, MO, USA) for at least 30 minutes. Afterward the cells were stimulated for 15 minutes with the corresponding TiO 2 NP concentrations (0.75 mg/ml, 0.5 mg/ml, 0.25 mg/ml, and 0.1 mg/ml) or ionomycin (1 mM) (positive control) (Sigma-Aldrich, MO, USA), both prepared in PBS. The supernatant containing secreted mucin was collected and briefly centrifuged at 8,000 rpm to remove the residual TiO 2 NPs. The supernatant was then incubated in a 96 well (Nunc MaxiSorp, VWR, CA, USA) plate overnight at 4uC. Afterward the 96-well plate was washed with PBST (PBS + 0.05% Tween-20) and then blocked with 1% BSA. The 96 well plate was washed again with PBST and incubated with lectin (Wheat germ agglutinin, WGA) (Sigma-Aldrich, MO, USA), conjugated to horseradish peroxidase (HRP; 5 mg/ml) (Sigma-Aldrich, MO, USA), at 37uC for 1 hr. The substrate, 3,39,5,59-Tetramethylbenzidine (TMB; Sigma-Aldrich, MO, USA), was added to each well at room temperature followed by H 2 SO 4 (Sigma-Aldrich, MO, USA) in order to terminate the reaction. The optical density was measured at 450 nm [37].

Image Analysis
After staining the treated cells, image analysis was performed with an inverted Nikon Eclipse TE2000-U fluorescent microscope. Each photo was taken at a magnification of 2006 and analyzed using SimplePCI (Compix Inc., Imaging Systems, Sewickle, PA, USA). The data shown is a representative of Ca 2+ signals of more than 200 cells.

Statistical Analysis
The data was presented as means6SD. Each experiment was performed independently at least three times. Statistical significance was determined using a Student's t-test analysis with p values ,0.05 (GraphPad Prism 4.0, GraphPad Software, Inc., San Diego, CA, USA).

TiO 2 NP characterization
Dynamic laser scattering (DLS) was used to characterize the TiO 2 NPs. The particle size distribution ranged from ,9 to 80 nm due to minor aggregation or agglomeration while the predominant size is ,50 nm (Fig. 1A).

TiO 2 NPs induce cytosolic Ca 2+ concentration increase
To investigate whether TiO 2 NPs could generate an increase in [Ca 2+ ] C , ChaGo-K1 cells were loaded with Rhod-2 AM dye and exposed to 0.05-1 mg/ml of TiO 2 NPs. The change in [Ca 2+ ] C , as represented by the fluorescence intensity within ChaGo-K1 cells, was monitored for 60 seconds. Figure 1B shows that 1 mg/ml of TiO 2 NPs induced an approximate 150% increase, while lower TiO 2 concentrations (,0.1 mg/ml) caused a minor elevation (,110%) in [Ca 2+ ] C when compared with untreated cells. The effect of TiO 2 treatment on the [Ca 2+ ] C of ChaGo-K1 cells followed a concentration-dependent manner (Fig. 1B).

Extracellular source for Ca 2+ increase
To determine the main source of elevated [Ca 2+ ] C upon stimulation, ChaGo-K1 cells were exposed to TiO 2 NPs in Ca 2+free Hanks' solution. EGTA (2 mM) was added in Hanks' solution to chelate possible traces of Ca 2+ . TiO 2 (0.05 mg/ml-1 mg/ml) treatment under Ca 2+ -free conditions failed to instigate a significant increase in [Ca 2+ ] C ( Fig. 2A). Our data suggests that the extracellular Ca 2+ pool is the primary source of the observed cytosolic Ca 2+ increase. We then tested whether TiO 2 NPs can induce a Ca 2+ influx via membrane channels. Blocking the channels with CdCl 2 (200 mM) significantly inhibited an increase in [Ca 2+ ] C (Fig. 2B). Co-treatment of cells with TiO 2 NPs and nifedipine greatly blocked the NP-induced [Ca 2+ ] C increase (Fig. 2C). However, the incomplete blockage of extracellular Ca 2+ influx via channels postulates additional Ca 2+ leakage through perturbed cell membranes. To confirm whether TiO 2 can instigate membrane disruption, thereby permitting unspecific extracellular Ca 2+ entry, cytosolic leakage was assessed using the fluorescent calcein dye. It was found that the dye permeation ratio increased from approximately 4 to 13% with elevated TiO 2 concentrations ranging from 0.1 to 1 mg/ml (Fig. 2D).

Oxidative stress induced Ca 2+ influx
To demonstrate that TiO 2 -evoked [Ca 2+ ] C increase can be associated with oxidative stress, cells were pretreated with an antioxidant, N-acetylcysteine (NAC) [27]. Pre-treatment with NAC was able to partially attenuate the increase in cytosolic Ca 2+ level triggered by 1 mg/ml and 0.75 mg/ml TiO 2 exposure (Fig. 2E). These results support the idea that oxidative stress, induced by TiO 2 NPs, contributes to the observed [Ca 2+ ] C increase and promote Ca 2+ -dependent mucin secretion.

The ER as an intracellular source of Ca 2+
In order to determine the involvement of ER Ca 2+ pool, it was depleted by pre-incubating the cells with thapsigargin. Pre-treatment with thapsigargin impeded TiO 2 NPs from triggering a sustained increase in the cytosolic Ca 2+ level (Fig. 3A). We then investigated the role of the CICR mechanism by blocking RYRs (ryanodine receptors) [20]. Our results revealed that CICR was largely inhibited by ryanodine (a blocker for RYR associated with the CICR response) resulting in a significantly diminished [Ca 2+ ] C increase induced by NPs (Fig. 3B). Ca 2+ -dependency of TiO 2 -induced mucin secretion Enzyme-linked lectin assay (ELLA) was used to assess the amount of mucin secreted from ChaGo-K1 cells when stimulated with TiO 2 NPs. When compared to the control, TiO 2 NPs increased mucin secretion by 113%, 125%, 133%, 137% and 150% at 0.05, 0.1, 0.25, 0.5 and 0.75 mg/ml, respectively (Fig. 4A). Chelating the intracellular Ca 2+ with BAPTA-AM yielded a significant reduction in mucin secretion (Fig 4B). Addition of thapsigargin (Fig. 4C) or ryanodine (Fig. 4D) also resulted in diminished mucin secretion induced by TiO 2 NPs. Our data indicates that TiO 2 -induced mucin secretion is dependent on the [Ca 2+ ] C , attributed to both external and internal Ca 2+ pools (Fig. 4A-D). Ionomycin (a Ca 2+ ionophore) was used to elicit mucin secretion as a positive control (Fig. 5).

Discussion
Recently, an increasing number of reports have shown that airborne particulate pollution found in both the ambient and working environments, particularly TiO 2 NPs, can exacerbate airway diseases [1,2,3,4,5,6,10,11,38]. Aggravated clinical manifestations of COPD, CF and asthma may include intensified symptoms of mucociliary transport impairment and mucus hypersecretion [15,39]. The resultant accumulation of thick obstructive mucus usually occupies airway lumen, thereby limiting airflow and leading to morbidity [15,39]. Despite documentations of TiO 2 -induced cellular nanotoxicity effects, pulmonary inflammatory responses and emphysema-like pathology [12], whether TiO 2 NPs can directly trigger mucin secretion has not been resolved. In this study, we demonstrate that TiO 2 NPs can stimulate mucin secretion from bronchial epithelial ChaGo-K1 cells via a Ca 2+ -dependent pathway.
Our study showed that TiO 2 NPs can induce mucin secretion that increases as a function of TiO 2 NP concentration (Fig. 4A). The TiO 2 concentration range used in our study is consistent with previous reports representing the concentration found in ambience and nanotechnology industries [30,40,41,42]. While NP exposure has been long associated with increasing mucin synthesis due to goblet cell hyperplasia [13], our study indicates that TiO 2 NPs can directly trigger mucin secretion in the airway.
It has been well established that intracellular Ca 2+ plays a vital role in stimulus-secretion coupling [43]. Previous reports have documented that an elevated [Ca 2+ ] C precedes mucin granule exocytosis [17]. NP exposure has been shown to trigger an intracellular Ca 2+ increase in various cells; therefore, we examined the cellular Ca 2+ signaling pathway involved during TiO 2 stimulation [25,28,44]. At TiO 2 concentrations of 0.5, 0.75, and 1 mg/ml, there was a sustained elevation in [Ca 2+ ] C . At lower doses (0.05, 0.1 and 0.25 mg/ml), the [Ca 2+ ] C increased gradually within the 1 st minute (Fig. 1B). Our data demonstrated that TiO 2 NPs induced a concentration dependent increase in [Ca 2+ ] C , which is consistent with results from the mucin secretion measurements (Fig. 4A).
The stimulus-induced intracellular Ca 2+ signal can be evoked by the entry of Ca 2+ through voltage-gated Ca 2+ channels, or by the release of Ca 2+ from intracellular Ca 2+ stores [43,45,46]. Previous researches have suggested that extracellular Ca 2+ influx plays an important role in the elevated [Ca 2+ ] C during NP stimulation [25,27,28,47]. Data from experiments performed in Ca 2+ -free Hanks' solution confirmed that [Ca 2+ ] C failed to increase when treated with TiO 2 NPs ( Fig. 2A). To characterize the nature of the Ca 2+ influx induced by TiO 2 NPs, we first evaluated the effect of cadmium chloride (CdCl 2 ), a general Ca 2+ channel blocker [34,48]. Figure 2B shows that the [Ca 2+ ] C remained low and relatively unchanged with CdCl 2 . Secondly, nifedipine, a widely used L-type Ca 2+ channel blocker, markedly diminished the increase in [Ca 2+ ] C (Fig. 2C). The effect of nifedipine implies that TiO 2 NPs can activate L-type voltage gated Ca 2+ channels, allowing extracellular Ca 2+ influx into the cytosol. This observation is consistent with previous reports showing that ultrafine carbon black and ZnO NP-induced [Ca 2+ ] C elevation can also be attenuated by nifedipine [27,28]. In addition, several reports have suggested that oxidative stress induced by NPs can exert an impact on the intracellular Ca 2+ signaling pathway and that the activity of Ca 2+ channels may be altered by ROS [27,28,44]. Results from Figure 2E showed that NAC significantly reduced the rising [Ca 2+ ] C generated by TiO2 NPs. Huang et al, has also demonstrated that NAC can attenuate the intracellular Ca 2+ level when challenged with ZnO NPs [27]. Our results support the idea that NAC and other antioxidants may be effective in reducing NPinstigated mucin hypersecretion. NPs such as TiO 2 can damage cell membrane integrity by possible lipid peroxidation [27,31], thereby creating pores on the lipid bilayer [49] that may allow the transient influx of extracellular Ca 2+ . Our data further demonstrated that co-adminstration of TiO 2 NPs and fluorescent calcein dye lead to intracellular leakage and the permeation efficiency increased in a TiO 2 concentration dependent manner (Fig. 2D). Calcein has also been previously utilized to evaluate the efficacy of peptides in causing membrane perturbation [50]. Our result suggests that the possible membrane perturbation/transient pore formation induced by TiO 2 NPs allows an extracellular Ca 2+ influx and may account for the portion of Ca 2+ that can not be completely abolished by blocking L-type Ca 2+ channels with nifedipine.
Increasing the [Ca 2+ ] C of human goblet cells has been shown to trigger degranulation [17]. We used BAPTA (cytosolic Ca 2+ chelator) to test whether the increase in Ca 2+ induced by TiO 2 NPs could stimulate mucin exocytosis. It is evident that BAPTA significantly inhibited mucin exocytosis (Fig. 4B), indicating that TiO 2 NPs can elicit a [Ca 2+ ] C increase, thereby leading to mucin secretion.
Besides the external Ca 2+ source (Hanks' solution), the ER is one of the major internal Ca 2+ stores. Figures 3A and 4C revealed that when the ER Ca 2+ had been depleted by pretreatment with thapsigargin, the TiO 2 NP-induced [Ca 2+ ] C failed to increase significantly, and the subsequent mucin secretion was abolished. Our data indicates that the ER plays a critical role in relaying TiO 2 -induced Ca 2+ signaling. CICR is a positive feedback mechanism where the ER amplifies a small increase in [Ca 2+ ] C , (e.g. due to voltage-gated Ca 2+ influx [22]), with the activation of RYRs that will lead to the release of more Ca 2+ from the ER [19,20]. Previous studies have shown that through activation of RYRs with Ca 2+ , CICR can generate an overall increase in [Ca 2+ ] C [20,21,22]. Our data showed that ryanodine inhibited a continual rise in [Ca 2+ ] C when applying TiO 2 NPs (Fig. 3B). Therefore, it is indicative that the TiO 2 -instigated increase in [Ca 2+ ] C was also CICR dependent. The effect of ryanodine was further demonstrated by the lack of mucin secretion under TiO 2 NP stimulation (Fig. 4D). In summary, our study indicates that cellular exposure to TiO 2 NPs can activate membrane L-type Ca 2+ channels, induce ROS production and possibly disrupt the cellular membrane. Influx of extracellular Ca 2+ into the cytoplasm raises [Ca 2+ ] C , which in turn can trigger ryanodine receptors on the ER to release ER resident Ca 2+ via the CICR mechanism. A sufficient increase in the cytosolic Ca 2+ level results in subsequent mucin secretion. More importantly, our results provide a direct link between airborne particulate matters and the pathogenesis of chronic airway diseases involving mucus hypersecretion and airway obstruction. In addition, we demonstrate that once thought inert and harmless TiO 2 NPs can indeed interfere with intracellular Ca 2+ signaling, possibly leading to pathological states.