The role of pulmonary ORCC and CLC-2 channels in the response to oxidative stress

Exposure of human lung epithelial cells to the oxidant pollutant ozone (O3) alters cell Cl− currents inducing an outward rectifier effect. Among the various Cl− channels, ClC-2 and ORCC seemed to be involved in this response. To identify the channel related to O3 induced current changes. Down regulating the expression of ORCC and ClC-2 genes and analyzing the membrane current show that the enhancement of the current disappeared when ORCC was silenced. The contribution of ORCC and ClC-2 channels in control and O3 treated cells was obtained by a mathematical approach. We suggest that O3 activates ORCC channels and slightly inhibited ClC-2 channels in the negative voltage range. These findings open the possibility of identifying the biomolecular changes induced by O3 allowing a possible pharmacological intervention towards chloride current due to oxidative stress.


Introduction
In Europe, air pollution is considered the second biggest threat to human health after climate change, causing about 500,000 premature deaths affecting mainly children and older people (European Commission 2017; European Environment Agency 2019). Among the different pollutants, Ozone (O 3 ) derived from the photochemical smog, has been defined as one of the most toxic outdoor stressors to human health (Mudway and Kelly 2000).
Because of its anatomical structure, location, and function, the lung represents one of the main targets of air pollution (Bargagli 2009) and making this tissue susceptible to damage and to develop pulmonary pathologies (Bowatte 2017;Malhotra et al. 2016;Tomczak 2016;Seaton et al. 1995;Guarnieri and Balmes 2014). Several studies in rodents and humans documented that exposure to a high concentration of O 3 can damage cells in the respiratory epithelium leading to bronchial and alveolar conditions (Harkema et al. 1987) with a decrease in lung functions, airway inflammation and hyperactivity, impaired defense mechanisms and a consequent increase in the incidence of respiratory infections (Mudway and Kelly 2000;Ciencewicki et al. 2008;Cross et al. 2002). O 3 interaction with the ELF (epithelial lining fluid) leads to the formation of Rita Canella and Giuseppe Valacchi contributed equally to the work. bioactive molecules including Reactive Oxygen Species (ROS), and peroxidation products (Laisk et al. 1989;Sarti et al. 2002;Del Corso et al. 1998;Hamilton et al. 1998;Dianzani 1998;Kreuzer et al. 1998;Selley 1998;Bosch-Morell et al. 1999;Parola et al. 1999;Bhalla 1999).
Cultured type II pneumocytes (A549 cell line (Foster et al. 1998)) are a good model to study the effect of O 3 on ion balance between cells and the surrounding environment (Johnson et al. 2006). On the membranes of these cells, there are several types of ion channels and transporters, involved in the homeostatic ions flow between the apical alveolar fluid and the basolateral interstitium (Shabbir et al. 2013). Chloride channels have acquired strong interest from researchers, both because of their activity in epithelial fluid secretion, regulation of cellular volume, proliferation and migration, myelination in the fetus and newborn, ischemia in white brain matter and myocardial cells (He 2017;Zhao et al. 2015) and their involvement in human channelopathies, among which the most known is cystic fibrosis (Verkman and Galietta 2009;Bi et al. 2014).
Since the damaging mechanisms of O 3 on ion channels have not yet been fully described, our study focused to evaluate the identification of the lung Cl-channels responsive to O 3 (Nilius and Droogmans 2003).
Among all the Cl-channels present in this cell type (Shabbir et al. 2013;Xiang et al. 2013;Brochiero et al. 2004;Chen 2005) our functional studies by means of selective blockers (Verkman and Galietta 2009;Canella et al. 2017Canella et al. , 2018Sato-Numata et al. 2016), identified the Outward Rectifier Chloride Channels (ORCC) activated by a strong depolarizing event or in conditions of oxidative stress, physiologically analogous to VSORs (outwardly rectifying Cl-volume-sensing channels) and often identified with them (Shimizu et al. 2004) and ClC-2 channels as the channels most likely responsive to O 3 , although the contribution of each channel needed more investigation.
In this study, we have performed siRNA technique to knock down Anoctamine 6 protein (ANO6), the functional component of ORCC channels (Martins et al. 2011), a member of a family of calcium-activated chloride channels (Ano1-Ano10) (Kunzelmann et al. 2011) indispensable for its operation and of about 65% for ClC-2 protein. We observed the almost complete block of ORCC and ClC-2 functioning. We then used the trace subtraction technique to get the contribution of each specific type of channel. This technique has been used extensively in electrophysiology from the basic papers of Hodgkin and Huxley (1952) to the present day and it is able to provide cleaner traces than those obtainable with inhibitors, protein knockdown or gene silencing, which cannot always be applied simultaneously on the channels (Martini et al. 2013;Yamanishi et al. 2018). So, with a mathematical approach, we were able to identify with good approximation, the contribution of each channel type to the total current induced by O 3 exposure.

Materials and methods
Cell culture and ozone exposure A549 cells (ATCC, Manassas, VA) were cultured in Dulbecco's modified Eagle's medium high glucose (DMEM, Lonza ® , Milan, Italy) until the ideal confluence was reached as previously described (Valacchi et al. 2011). Cells were exposed to 0.1 ppm of O 3 for 30 min as in the previous studies (Valacchi et al. 2015).

Cytotoxicity determination
Cytotoxicity study was performed after siRNA transfections by lactate dehydrogenase (LDH) (EuroClone, Milan, Italy) as previously described (REF PMID: 32504424). Positive control with 100% toxicity was treated with 2% (V/V) Triton X-100 in culture media for 30 min at 37 °C.

Western blot
Total cell lysates were extracted in RIPA buffer containing protease and phosphatase inhibitor cocktails (Sigma-Aldrich Corp.) and protein concentration was determined by Bradford assay (Biorad Protein assay, Milan, Italy). After protein quantification, 60 μg boiled proteins were loaded into 10% sodium dodecyl sulphate-polycrylamide electrophoresis gels and blotted onto nitrocellulose membranes. The membrane was then blocked and incubated overnight as previously described (Benedusi et al. 2018).

Patch-clamp technique
The procedure was performed as previously reported (Canella et al. 2019). The chemicals were purchased from Sigma-Aldrich. Patch-clamp protocol has a duration of 1400 ms ranging from + 70 to − 110 mV with 20 mV increments from a starting holding potential of − 30 mV.
The O 3 treated cell samples were examined with patchclamp technique immediately after 30 min of exposure.

Data analysis and statistical procedures
Data were analyzed with GraphPad Prism v.6 and reported as mean ± standard error. Statistical tests performed for control-treated comparisons were Student's t test for means and one-way and two-way ANOVA. P < 0.05 was the significance level. P values are indicated both in text and graphics.

Evaluation of ORCC and ClC-2 channels involvement in O 3 effect on cell current
To better elucidate which chloride channel is involved in the action of O 3 on chloride ions flow, the proteins ANO6 or ClC-2 were respectively silenced in A549 cells and their protein expression levels were evaluated after 24 h. To overcome the problem of interaction between ORCC and ClC-2 silencing with the functionality of other Cl − channel types, we compared the current-voltage (I-V) curves obtained treating the cells with specific inhibitors of the two channels (DCPIB for ORCC and CdCl 2 for ClC-2) obtaining results not significantly different from those obtained by siRNA technique (Insert in Fig. 1; two-way ANOVA test). Since the two current blocking mechanisms are very different, but the results are very similar, we believe that we are reasonably comfortable that siRNA technique does not alter the behavior of other channels.
As shown in Fig. 1, after silencing ANO6 its levels decreased significantly, compared to the controls (P < 0.0001; one-way ANOVA; post-hoc Bonferroni test). No changes were noticed in the siRNA negative control or Lipofectamine alone. Concerning ClC-2, as shown in Fig. 1, after silencing its protein levels decreased significantly (− 60% VS control cells, Lipofectamine, negative control; P < 0.0001; one-way ANOVA; post-hoc Bonferroni test); also in this case the negative control siRNA or Lipofectamine alone did not show any significant differences. These data indicated that our experiments of siRNA-mediated protein knockdown were efficient and that the knocking down effect achieved by ANO6 or ClC-2-siRNA was specific and effective.
By patch-clamp technique, we analyzed the movement of chloride ions before and after O 3 exposure, in control, and when ANO6 or ClC-2 protein were knocked down. Figure 2 shows the example current traces obtained from a control cell (panel A), an O 3 -treated cell (panel B), an ANO6 silenced cell (panel C), an ANO6 silenced cell treated with O 3 (panel D), a ClC-2 silenced cell (panel E), a ClC-2 silenced cell treated with O 3 (panel F). The recordings were obtained under voltage-clamp conditions as described in the Materials and Methods section. In panel 2A, the typical time-dependence of the current due to ClC-2 is hidden by the activity of the other channels, especially ORCC, whose contribution is quite important also to negative potentials. The silencing of ANO6 (panels 2C and 2D) makes this phenomenon evident at the more negative potentials.
Patch-clamp protocol was applied to sample of control cells (n = 21), cells knocked down for ANO6 (n = 9) and cells knocked down for ClC-2 (n = 9), obtaining the I-V curves shown in Fig. 3a. Two-way ANOVA test indicated that the control curve as a whole and control knocked down curves were not significantly different (P(I tot vs I siANO6 ) = 0.12; P (I tot vs I siClC-2 ) = 0.21); because of the little amplitude of these currents near the resting membrane potential, the contribution of I leak becomes a more relevant part of the total current, making the comparison of the whole curves not significant. Bonferroni test showed a significant decrease (P < 0.05) for the inward part of the siClC-2 curve at − 110 mV and a significant decrease of the siANO6 current at the most negative (P < 0.01 for V = − 110 and − 90 mV) and most positive potential values (P < 0.01 for V = + 70 and + 50 mV).
The cellular cur rent reversal potentials of both silenced genes (E rev(siANO6) = 35.7 ± 4.1 mV; E rev(siClC-2) = 31.7 ± 5.7 mV) are more positive than control cells (E rev(tot) = 17.4 ± 2.6 mV), indicating the lower contribution of the chloride ions to the total current in these conditions (E Cl = − 2.5 mV). These results were not surprising for ClC-2, since its contribution to the membrane potential is well known, but it was intriguing for ANO6, which role in control conditions is not yet well clarified.

Identification of total current components in control conditions
To dissect the total control current (I tot ) in its components, we assumed that in our experimental conditions (internal K + substituted with Cs + ), I tot is due to the contribution of ClC-2 channels (I ClC-2 ), ORCC channels (I ORCC ) and leak channels (I leak ) that include the other non-voltage dependent channels present in the cells, as described in the equation (1): By knocking down ANO6, we consistently reduce I ORCC , so, by the functional response, we can assume that: (1) I tot = I ORCC + I ClC -2 + I leak To estimate the contribution of ORCC channel to the total control current, we can subtract (2) to (1): Since: we can apply the same formula to calculate I ClC-2 : Knowing I ORCC and I ClC-2 it is possible to obtain I leak : (2)  the cell is important since the current is inward at all the explored voltage.

Identification of total current components after O 3 exposure
As depicted in Fig. 4a, O 3 exposure increases the total current (I tot(O3) , n = 13) confirming our previous work (Canella et al. 2017). However, when the ANO6 protein is silenced, the effect of O 3 decreases and the I-V relationship of the control cells (I tot ) and the silenced ones (I siANO6(O3) , n = 4) are not significantly different (two-way ANOVA; P = 0.38). Moreover, the reversal potential of the current, shifts towards more positive values (E rev(siANO6(O3)) = 33.8 ± 6.7 mV) with respect to the reversal potential of I tot(O3) (E rev(tot(O3)) = 7.6 ± 1.7 mV). Treating the ClC-2 silenced cells with O 3 resulted in the gray triangles curve (I siClC-2(O3) , n = 4). In these cells, the reaction to O 3 exposure is mainly due to ORCC channels and the curve is greater than the control, and was not significantly different from I tot(O3) (two-way ANOVA; P = 0.30). Reversal potential of I siClC-2(O3) (E rev(siClC-2) = 7.5 ± 0.5 mV) is identical to that of I tot(O3) . From these data, we can extrapolate that ORCC channels significantly respond to O 3 exposure. To graphically clarify the contribution of each channel type we operated mathematically.
We can describe the current obtained after O 3 exposure as: (4) I tot(O3) = I ORCC + I ClC -2 + I leak + I ORCC(O3) + I ClC -2(O3) where I ORCC(O3) is the additional part of current induced by O 3 on ORCC channels and I ClC-2(O3) is the additional part of current induced by O 3 on ClC-2 channels.
Panel B shows the part of total current due to O 3 action (black diamonds) obtained subtracting (1) to (4): Subtracting (2) to (5) we can obtain an estimate of the contribution to I O3 of ClC-2 channels. The current is always positive indicating that the effect of O 3 on ClC-2 channels is absent or even inhibiting at negative potentials and becomes slightly activating at positive potentials (Fig. 5).
Subtracting (3) to (6) we can obtain an estimate of the contribution to I O3 of ORCC channels in the same condition.

Effect of O 3 on ORCC and ClC-2 channels
To visualize the effect of O 3 on the current passing through the ORCC and ClC-2 channels, we compared the total ORCC current after O 3 treatment (I ORCC + I ORCC(O3) ) with

Fig. 4
Current-voltage relationship before and after cell silencing, after O 3 treatment: a Current-voltage curves of control cells (I tot , n = 21), exposed to O 3 (I tot(O3) , n = 13), after silencing of ANO6 and O 3 exposure (I siANO6(O3) , n = 4) and after silencing of ClC-2 and O 3 exposure (I siClC-2(O3) , n = 4) cells; the comparisons between the control group and the siANO6 group and the comparison between the O 3 group and the siClC-2 group were not significant (two-way ANOVA: P (I tot vs I siANO6(O3) ) = 0.38; P (I tot(O3) vs I siClC-2(O3) ) = 0.30). b Current-voltage curves mathematically obtained as indicated in Results: current due to O 3 action (I O3 ), part of the I O3 due to ORCC channels activation (I ORCC ) and part of the I O3 due to ClC-2 channels activation (I ClC-2 ) the ORCC current in the control condition. The same comparison has been done for ClC-2 current. The results are shown in Fig. 5, panel A. O 3 induces a significant increase both of inward (two-way ANOVA; P < 0.001) and outward current (two-way ANOVA; P < 0.001) due to ORCC channel, while ClC-2 channels activity is slightly enhanced only for the outward part of the curve (two-way ANOVA; not significant; P = 0.11).
Adding the components of the total current calculated under control conditions (Eqs. 2a, 3a and 3b), with the contribution given to the current by the ORCC and ClC-2 channels when the cells are treated with O 3 (Eqs. 7a and 7b), the I tot(O3) has been reconstructed.
The comparison between the experimental curve and the reconstructed one is depicted in Fig. 5b.
Two-way ANOVA test between observed data and calculated ones is not significant (P = 0.47), supporting our reconstruction procedure.

Discussion
Our previous work demonstrated that, in A549 cells, exposure to O 3 causes a considerable increase of the total membrane current, mainly in its outward component. The functional experiments implemented to identify the responsible channels for this alteration by means of ORCC and ClC-2 channels inhibitors did not unambiguously identify the O 3 target. In this study, we discern the contribution of ClC-2 and ORCC channels to the current induced upon O 3 exposure, by siRNA technique and a mathematical approach applied to patch-clamp traces.
The gene encoding for ANO6, a functional part of ORCC (Martins et al. 2011), was silenced and the currents were analyzed before and after O 3 exposure ( Fig. 3a  and 4a).
Surprisingly, patch-clamp recordings on knocked down cells in the control condition, showed that an important part of the chloride current is due to ORCC channels. This finding indicated that the ORCC channels are not only activated by depolarizing potentials or in presence of oxidative stress, as reported in the literature (Toczyłowska-Mamińska and Dołowy 2012; Wanitchakool et al. 2016), but they contribute to the physiological membrane potential. This observation is of particular interest since the role of this channel in control conditions has not been yet analyzed. The overall I-V relationship of silenced cells is non-significantly different from the control, but the current values are lower than those of the control at all considered potentials. In particular, those evoked at the most positive (+ 70 and + 50 mV; P < 0.001; P < 0.05) and negative (− 110 and − 90 mV; P < 0.001; P < 0.01) potentials are significantly different from the control (two-way ANOVA, Bonferroni test).
When we silenced ClC-2 gene, the comparison between the obtained I-V curve and the control one, shows a significant decrease of the current at very negative potentials (− 110 mV; P < 0.05; two-way ANOVA, Bonferroni test), indicating the important contribution of this channel at the membrane resting potential and its inward rectifier behavior already reported in literature (Cuppoletti et al. 2000).

Fig. 5
Contribution of ORCC and ClC-2 channels to the membrane current after O 3 treatment: a current-voltage curves mathematically obtained as indicated in Results: contribution to I tot(O3) due to ORCC channels (I ORCC + I ORCC(O3) ) and to ClC-2 channels (I ClC-2 + I ClC-2(O3) ) compared to the corresponding basal current (I ORCC and I ClC-2 ). See Results for statistical analysis. b Current-voltage curves of cells exposed to O 3 (I tot(O3) , n = 13) and mathematical reconstruction of I tot(O3) . The comparison between the two curves was not significant (two-way ANOVA; P = 0.47)

3
The reversal potential of both currents after silencing, shifts towards very positive potentials, far from the equilibrium potential for chloride ions in our model (E Cl = − 2.5 mV). Since in our experiments, only chloride ions have an equilibrium potential round 0 mV, this result confirms the closing of chloride conductance in both cases.
With the mathematical approach, we attempt to identify the contribution to the total control current of ORCC and ClC-2 channels. Our experiments show that from the functional point of view, the silencing approach almost completely eliminates the current carried by the type of targeted channels. In any case, the subtraction of traces makes the presence of any residual current irrelevant.
The reversal potential of both calculated currents is very close to E Cl indicating the ability of the mathematical approach to isolate the control current components. On the other side, leak current is a compound inward current that does not revert in the analyzed voltage range, indicating the cations prevailing role for leakage current.
When the silenced cells are exposed to O 3 , the current of siANO6 cells does not show any enhancement and its reversal potential becomes very positive (E rev(siANO6(O3)) = 33.8 ± 6.7 mV), while the current of siClC-2 cells results not statistically different from the current of O 3 treated cells and the reversal potential is similar to E rev(tot(O3)) (see Results). This finding indicates that the current induced by O 3 exposure is mostly due to ORCC channels activation.
Dissecting the contribution of each channel type to the O 3 effect, we can also show that O 3 slightly inhibits ClC-2 inward current: the calculated values of ClC-2 current at the negative voltage values, were positive as if an outward current was added to the inward original one. In fact, subtracting an inward current by closing the corresponding channels, is mathematically equivalent to adding an outward current. Unfortunately, the effect of O 3 on the negative current due to ClC-2 did not reach a statistically significant value. Figure 5a shows the ORCC current after O 3 treatment as the summation of the basal activity of ORCC (I ORCC ) and the activity due to O 3 action (I ORCC(O3) ) and compares it with the basal ORCC current (I ORCC ) alone. The comparison indicates an increase of both the outward and the inward components of the current after O 3 exposure. This finding is intriguing because ORCC channels are considered outward rectifier (Mall and Galietta 2015), nevertheless, our results indicated that the current is almost linear and that after O 3 exposure, the outward effect in our cells is mainly due to the interplay of the three main current present in the cells: I ORCC , I ClC-2 and I leak . These data will induce us to investigate, in our future studies, whether O 3 influences, in addition to the number of open channels, also their voltage-dependence.
We have made the same calculation for ClC-2 current. In this case, the comparison between the ClC-2 current after O 3 exposure (I ClC-2 + I ClC-2(O3) ) and the basal I ClC-2 , clarifies what observed (Fig. 4b): the inward part of the current is very marginally influenced by the action of O 3 , while the outward part has a slight increase. This effect is not sufficient to compensate for the loss of efficacy of ORCC channels in O 3 -treated and ANO6 silenced cells and to make the total current significantly different from control.
Finally, to verify the correctness of our mathematical deductions, we reconstructed according to the procedure indicated in the results, the total current expressed by the cells when they are treated with O 3 (I tot(O3) calculated) and we compared it with the experimental curve I tot(O3) .
These results show that our approach is able to identify the contribution of the different channel types to the total experimental current in an accurate and precise way.
Given the importance of the Cl − current in pulmonary alveolar cells, it becomes fundamental to know the contribution of the different types of channels to the membrane current in physiological and oxidative stress conditions. In fact, the unbalance of the chloride ions distribution leads to diseases such as cystic fibrosis (Mall and Galietta 2015) where the CFTR channel/transporter is defective. Therefore, becomes very important to find drugs able to rebalance the distribution of chloride ions, acting on the non-defective chloride channels.
Author contributions RC: conceptualization, methodology, validation, investigation, writing-reviewing and editing; MB: conceptualization, methodology, validation, investigation, and writing-original draft; MM: formal analysis, investigation and data curation; AG: investigation and resources; FC: resources and software; GV: conceptualization; writing-reviewing, editing and supervision and project administration.

Declarations
Conflict of interest Rita Canella, Mascia Benedusi, Marta Martini, Anna Guiotto, Franco Cervellati, Giuseppe Valacchi confirm that they have no conflicts of interest to disclose.
Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.