Formaldehyde impairs transepithelial sodium transport

Unsaturated oxidative formaldehyde is a noxious aldehyde in cigarette smoke that causes edematous acute lung injury. However, the mechanistic effects of formaldehyde on lung fluid transport are still poorly understood. We examined how formaldehyde regulates human epithelial sodium channels (ENaC) in H441 and expressed in Xenopus oocytes and exposed mice in vivo. Our results showed that formaldehyde reduced mouse transalveolar fluid clearance in vivo. Formaldehyde caused a dose-dependent inhibition of amiloride-sensitive short-circuit Na+ currents in H441 monolayers and of αβγ-ENaC channel activity in oocytes. α-ENaC protein was reduced, whereas phosphorylation of the extracellular regulated protein kinases 1 and 2 (ERK1/2) increased significantly post exposure. Moreover, both α- and γ-ENaC transcripts were down-regulated. Reactive oxygen species (ROS) was elevated significantly by formaldehyde in addition to markedly augmented membrane permeability of oocytes. These data suggest that formaldehyde contributes to edematous acute lung injury by reducing transalveolar Na+ transport, through decreased ENaC activity and enhanced membrane depolarization, and by elevating ROS production over long-term exposure.


Formaldehyde decreases mouse alveolar fluid clearance in vivo.
To investigate whether formaldehyde could inhibit alveolar fluid transport in the mouse lungs to develop lung edema, AFC was measured by instilling 5% BSA saline. As shown in Fig. 1, the in vivo AFC rate was reduced to 10.3% in the presence of formaldehyde, significantly lesser than controls (17.8 ± 1.7%, N = 6, P < 0.01). We next studied whether formaldehyde inhibits ENaC activity. To address this issue, we applied a specific ENaC inhibitor, amiloride (1 mM). Our data showed that amiloride reduced AFC to 8.5 ± 1.2% (P < 0.01 versus Control, N = 6). Coadministration of formaldehyde (FA/Amil) did not further reduce fluid resolution (8.3 ± 1.0%, P > 0.05 versus formaldehyde group, N = 6). These in vivo data suggest that formaldehyde could decrease amiloride-sensitive fraction of AFC associated with ENaC and formaldehyde might decrease transalveolar fluid clearance contributed by ENaC.

Effects of formaldehyde on amiloride-sensitive short-circuit currents in intact H441 monolayers.
Amiloride-sensitive component of AFC reflecting ENaC-mediated fluid transport and ENaC has been reported to contribute to ~60% of AFC 23,24 . We postulated that ENaC was a target of formaldehyde-induced lung edema. To examine the effects of formaldehyde on electrogenic transepithelial Na + transport, we measured the amiloride-sensitive short-circuit currents in confluent H441 monolayers, which were mounted in an 8-chamber Ussing chamber system. H441 cells are derived from human Clara cells found in the bronchiolar epithelium. This cell line has been applied to study lung ENaC in last two decades 25 . Compared with primary cells, there are not between-donor variances since lung ENaC are regulated by sex hormones, aging, pollutants, and other noxious insults. Although H441 are cancerous cells, the airway epithelial cells in smokers are neoplasm. When formaldehyde was pipetted to the apical side of the monolayers, short-circuit current levels were inhibited in a dose-dependent manner, and the current was markedly inhibited by 100 μ M amiloride, a specific inhibitor of sodium channels ( Fig. 2A). We set the basal amiloride-sensitive current as 100%, with amiloride-sensitive currents defined as the difference between the total current and the amiloride-resistant current. Following treatment of H441 monolayers with 50, 100, 200, or 500 μ M formaldehyde, the amiloride-sensitive short-circuit currents decreased to 97.9% (± 0.6%), 91.1% (± 2.3%), 78.2% (± 6.2%), and 46.4% (± 9.6%), respectively. The IC 50 value was 165.6 μ M (± 0.5 μ M), calculated by fitting the dose-response curve with the Boltzmann equation 26 (Fig. 2B). These findings suggested that formaldehyde treatment decreased ENaC activity.
Formaldehyde inhibits expression of the ENaC α-subunit protein in H441 cells. Per the genetic engineered mice, 'scnn1a' , encoding ENaC α subunit, is critical for removing fluid from the air spaces. To determine whether formaldehyde exposure altered expression of the ENaC α -subunit at the protein level, we treated H441 cells with 200 μ M formaldehyde for different periods. As shown by immunoblotting analysis (Fig. 3A), formaldehyde exposure decreased α -ENaC protein levels, whereas expression of the internal standard β -actin was unchanged. The specific band about 97 kD for the α -ENaC protein could be seen according to the manufacturer's manual and the full-length blots/gels are presented in Supplementary Figure 1. Using β -actin levels as reference, α -ENaC protein expression was significantly reduced by 12 h, and although levels had risen by 48 h, they remained significantly lower than controls (Fig. 3B). We also examined γ -ENaC protein expression, but although a slight decrease was detected in treated cells, the levels were not significantly different from control levels (data not shown). We were unable to examine β -ENaC protein expression in cell lysates, due to the lack of an anti-β ENaC antibody suitable for western blot method.
Formaldehyde inhibits mRNA expression of ENaC αand γ-subunits in H441 cells. Based on the reduction of ENaC protein level, we predicted a decrease in ENaC subunit at the transcriptional level. To examine , and both (FA/Amil) for exposed mice. Reabsorption rate of instillate was computed as the percentage of instilled volume after 60 min (% 60 min).
Average AFC values are presented as mean ± SE, One-way ANOVA. **P < 0.01, compared with control group. N = 6.
whether formaldehyde inhibits ENaC expression in human alveolar epithelial cells at the transcriptional level, H441 cells were treated with 200 μ M formaldehyde for different periods, and then α -ENaC and γ -ENaC expression was evaluated by real-time PCR. Figure 4A shows that formaldehyde inhibited α -ENaC mRNA expression rapidly at 2 h. By 4 h, α -ENaC mRNA levels had decreased to 15.7% of control cell levels. γ -ENaC mRNA expression was also suppressed quickly, falling to 13.9% of control cell levels by 4 h (Fig. 4B). These data demonstrate that formaldehyde inhibits α -ENaC and γ -ENaC mRNA expression in H441 cells in a time-dependent manner.
Formaldehyde increases protein phosphorylation. Phosphorylation of ENaC is the most common post-translational modification by noxious insults. We tested the possibility for formaldehyde to alter the balance

Figure 5. Effects of formaldehyde on ERK1/2 phosphorylation in H441 cells. (A)
Representative western blot of phosphorylated ERK1/2 in protein extracted from H441 cells exposed to 200 μ M formaldehyde for 0-90 min. Blots were immunostained with total ERK1/2 as a loading control. (B) Graphical representation of data obtained from three sets of western blots and quantified using gray analysis (p-ERK1/2/ERK1/2). Data are shown as the mean ± SE, *P < 0.05, **P < 0.01, compared with control.
Scientific RepoRts | 6:35857 | DOI: 10.1038/srep35857 Formaldehyde increases oxidative stress in H441 cells. Cigarette smoke increases oxidative stress levels in the lung 27 , and based on that study, we speculated that potent oxidative formaldehyde, a component of cigarette smoke, might increase oxidative stress in H441 cells. Therefore, H441 cells were treated with 200 μ M formaldehyde for 0-24 h, and intracellular oxidative levels were examined using the dichlorofluorescein assay. Figure 7 demonstrated that cells exposed to formaldehyde exhibited a time-dependent increase in ROS levels (A-D), with significant differences detected at 6 and 24 h post exposure compared to the unexposed controls ( Fig. 7E).

Formaldehyde inhibits human αβγ-ENaC activity in oocytes.
To examine the effects of formaldehyde on human α β γ -ENaC, oocytes expressing α β γ -ENaC were superfused with 200 μ M formaldehyde. The inward and whole cell currents were decreased by formaldehyde and reached a stable plateau within a few minutes (Fig. 8A). The inhibited current at − 120 mV was approximately 15% of the basal level, a statistically significant change ( Fig. 8B, P < 0.01). In the presence of amiloride (10 μ M), α β γ -ENaC activity was almost inhibited completely. These results suggest that formaldehyde can rapidly inhibit human α β γ -ENaC activity. We speculated that the rapid decrease of oocyte current and the short circuit current is associated with the specific inhibition of formaldehyde on ENaC-associated function, for only α β γ -ENaC channel is expressed in occytes and the amiloride-sensitive short circuit currents which mainly reflect ENaC activity in H441 monolayers are affected. Whereas the slow effect of formaldehyde on the oxidative stress shows the overall trend of all channels in H441 cells, including non-ENaC components.

Formaldehyde depolarizes cells by increasing membrane permeability. Incubation in formalde-
hyde for 24 h had little effect on the shape of oocytes expressing human α β γ -ENaC (Fig. 9A). However, analysis of amiloride-resistant currents revealed a significant increase in the currents in formaldehyde-pretreated oocytes compared to the control group (Fig. 9B). These data indicate that formaldehyde leads to an augmentation in membrane permeability of oocytes, but without causing visible structural damage.

Discussion
Alveolar fluid clearance appears to play an important role in lung fluid balance in both physiological and pathological conditions, as evidenced by the study of Matthay's group, which showed that patients with acute lung injury and acute respiratory distress syndrome (ALI/ARDS) have dysfunctional alveolar fluid clearance 28 . The increased permeability of capillary endothelium, resulting from injury to the lung microvascular circulation, may lead to increased transvascular fluid filtration into the perimicrovascular interstitium 14 . If the epithelial layer is also damaged, the fluid will move though the epithelial layer into the airspace 29 . While cystic fibrosis transmembrane conductance regulator deficiency had been shown to contribute to the physiopathology of cigarette-induced diseases such as chronic bronchitis and COPD 30,31 , most studies confirm that Na + reabsorption is also a key factor in maintaining fluid balance in the lung, and ENaC plays a crucial role in Na + reabsorption 32-34 . The human lung adenocarcinoma cell line NCI-H441 is one of the classic cell lines used in researching ENaC activity and function. In the present study, H441 cells were used to examine the effects of formaldehyde, a component of cigarette smoke, on ENaC function. We found that formaldehyde could inhibit the amiloride-sensitive short-circuit current levels of H441 cell monolayers in a dose-dependent manner, suggesting that formaldehyde may inhibit Na + transport by suppressing lung ENaC. We chose 200 μ M formaldehyde in the following experiments according to the IC 50 value calculated in our ussing chamber results and the previous studies 35 , which amounts to the tobacco smoke for 3~4 cigarettes 36 . We also studied the effects of formaldehyde on mouse alveolar fluid clearance, with results suggesting that formaldehyde reduced alveolar fluid clearance in mice.
Due to the importance of ENaC in lung fluid clearance and the major functional role of its α -subunit 18 , we examined whether formaldehyde had any effect on α -ENaC. Using H441 cells, we discovered that formaldehyde exposure decreased α -ENaC expression at the protein and transcriptional levels, and that γ -ENaC mRNA expression was also inhibited. Furthermore, we explored the influence of formaldehyde on protein phosphorylation of the ERK pathway. ERK protein, consisting of ERK1 and ERK2, is a member of the MAPK family, which plays a key role in signal transduction from surface receptors to the nucleus 37 . Previous studies demonstrated that the activation of ERK1/2 inhibited ENaC activity by phosphorylating residues in the C-termini of the β andγ subunits, enhancing the docking of the ubiquitin ligase (Nedd4-2) with these subunits [38][39][40] . It has also been reported that transforming growth factor-β 1 reduced α -ENaC mRNA expression on apical membranes through the ERK1/2 pathway 41 . Additionally, formaldehyde induces apoptosis of lung epithelial cells through the p38MAPK pathway 42 . In this study, the phosphorylation of ERK1/2 was enhanced rapidly and strongly by exposure to formaldehyde, whereas treatment with the ERK1/2 inhibitor PD98059 inhibited this formaldehyde-induced phosphorylation of ERK1/2, suggesting that formaldehyde reduces ENaC activity via enhancing ERK1/2 phosphorylation in H441 cells.
Previous studies suggested the relevance of the ROS in the apoptosis induced by cigarette smoke extract and ROS production as a common mechanism of ENaC regulation by insulin 27,43 . Studies also have shown that cigarette smoke extract regulates lung ENaC via oxidant signaling, affecting alveolar fluid balance 44 . Because formaldehyde is one of the substances in cigarette smoke that can induce oxidative stress in the lung, we examined the role of formaldehyde on oxidative stress, using a fluorescent marker of oxidation. The results showed an  increase in fluorescence in cells exposed to formaldehyde, suggesting that formaldehyde can increase intracellular oxidation levels. The apical membrane ENaC expression is known to be regulated by aldosterone and the amiloride-sensitive short-circuit currents mainly reflects ENaC activity, is also a function of the ion selectivity and permeability properties of the paracellular and transcellular pathways 45,46 . As we know, oocytes have little intrinsic channels and Xenopus laevis oocyte expression system has been widely used to elucidate the electrophysiological features of different kinds of channels by injecting the corresponding cRNAs. The biophysical properties and pharmacological profile of ENaC expressed in oocytes are similar to those of the native channel 47 . Another advantage of the oocyte expression system is to eliminate other native ion transport pathways, enable us to test the direct effects of formaldehyde on human lung ENaC. Our voltage clamp results using oocytes expressing α β γ -ENaC showed that formaldehyde exposure markedly augmented the membrane permeability of oocytes without apparent damage.

Conclusions
These data suggest that formaldehyde sharply reduces ENaC activity via enhancing ERK1/2 phosphorylation and membrane depolarization, thus reducing transalveolar Na + transport, followed by decreasing the transcription and translation of ENaC subunits, as well as elevating cellular ROS products over long-term exposure.

Materials and Methods
Animals. All experiment methods involving mice and Xenopus laevis were carried out in accordance with the guidelines and regulations of Animal Care and Use Committee and all experimental protocols were approved by China Medical University and the University of Texas Health Science Center at Tyler, respectively. Animals were kept under pathogen-free conditions. Please see following relevant paragraphs for experimental details.
Air-Liquid Cell Culture. The human club cell line NCI-H441 was cultured as described previously 48,49 , in a humidified atmosphere of 5% CO 2 and 95% O 2 at 37 °C. RPMI-1640 medium (ATCC, Manassas, VA) was supplemented with 10% fetal bovine serum (Gibco, Waltham, MA), 2 mM L-glutamine, 10 mM HEPES, 1 mM sodium pyruvate, 4.5 g/l glucose, 1.5 g/l sodium bicarbonate, 100 IU/ml penicillin and 100 μ g/ml streptomycin, and the culture medium was changed every other day. Cells were maintained in serum-free medium overnight after reaching 90% confluency, and then the medium was changed before exposure of the cells to formaldehyde. To analyze protein phosphorylation, cells were pretreated with 100 μ M of the mitogen-activated protein kinase (MAPK) inhibitor PD98059 (Beyotime, Shanghai) for 30 min prior to formaldehyde exposure. For Ussing chamber assays, cells were grown in Costar Snapwell culture cups, until reaching confluency at 24 h, and then medium and non-adherent cells in the apical compartment were removed to adapt the cells to air-liquid interface culture. An epithelial volt/ohm meter (WPI, Sarasota, FL) was used to measure transepithelial resistance, and confluent filters with resistance > 500 Ωcm 2 were used for measuring short-circuit current levels. Ussing Chamber Assay. Measurements of transepithelial resistance and short-circuit currents in H441 monolayers were performed as described previously 51 . H441 monolayers were mounted in Ussing chambers (Physiologic Instruments, San Diego, CA) and bathed on both sides with a solution of 120 mM NaCl, 25 mM NaHCO 3 , 3.3 mM KH 2 PO 4 , 0.83 mM K 2 HPO 4 , 1.2 mM CaCl 2 , 1.2 mM MgCl 2 , and 10 mM HEPES, supplemented with either 10 mM mannitol (apical compartment) or 10 mM glucose (basolateral compartment) 24 . The osmolality of each solution, as measured by a freezing depression, was between 290 and 300 mOsm/kg. Both sides of the bath solutions (pH 7.4) were bubbled with 95% O 2 and 5% CO 2 continuously at 37 °C. The short-circuit current levels were measured with 3 M KCl, connected on either side to Ag-AgCl electrodes filled with 4% agarose salt bridges. H441 monolayers were short-circuited to 0 mV, and a 10 mV pulse of 1 s was given every 10 s to monitor transepithelial resistance. Data were collected using the Acquire and Analyse program version 2.3.
Real-time PCR. For real-time PCR, H441 cells were seeded in RPMI-1640 medium containing 10% FBS, in 6-well plates, and grown to about 80-90% confluence. After exposure to formaldehyde for different time periods, cells were washed with cold PBS. Total RNA was extracted using TRIzol reagent (Invitrogen, Waltham, MA), according to the manufacturer's instructions, and RNA concentration was measured by spectrophotometry at 260 nm. The following primer pairs were used for real-time PCR: α -ENaC forward (5′ -AAC AAA TCG GAC TGC TTC TAC-3′ ) and α -ENaC reverse (5′ -AGC CAC CAT CAT CCA TAA A-3′ ), γ -ENaC forward (5′ -GCA CCG TTC GCC ACC TTC TA-3′ ) and γ -ENaC reverse (5′ -AGG TCA CCA GCA GCT CCT CA-3′ ), and GAPDH forward (5′ -AGA AGG CTG GGG CTC ATT TG-3′ ) and GAPDH reverse (5′ -AGG GGC CAT CCA CAG TCT TC-3′ ). Total RNA (1 μ g) was used as a template for each reaction, and all samples were run in triplicate in a final volume of 20 μ l. Reverse transcription was performed using a single cycle of 95 °C for 15 min, followed by 40 cycles of 95 °C for 1 min, 55 °C for 0.5 min, 72 °C for 0.5 min. Relative expression of ENaC mRNA was calculated using the 2 −△(△CT) comparative method, with normalization of each sample against expression of the endogenous reference gene GAPDH.
Detection of Intracellular ROS. The fluorogenic substrate 2′ ,7′ -dichlorofluorescein diacetate is a cell-permeable dye, which, after entering into a cell, can be hydrolyzed into non-fluorescent 2′ ,7′ -dichlorofluorescein by intracellular esterases and then oxidized to fluorescent 2′ ,7′ -dichlorofluorescein by ROS within the cytosol; the resulting change in fluorescence intensity can be used to measure intracellular generation of ROS. To assay for ROS, H441 cells were cultured in 24-well plates, as described above. After exposure to 200 μ M formaldehyde for 24 h, the cells were washed twice with PBS and then incubated in medium containing 10 μ M 2′ ,7′ -dichlorofluorescein diacetate for 30 min at 37 °C in the dark. Immediately after incubation, the cells were washed three times with serum-free medium to remove extracellular 2′ ,7′ -dichlorofluorescein diacetate, and fluorescence intensity was monitored using a Leica DMI 3000B inverted fluorescence microscope.

Expression of human ENaC in oocytes.
Oocytes were surgically removed from appropriately anesthetized adult female Xenopus laevis (Xenopus Express, Brooksville, FL), and cRNAs for α , β , and γ human ENaC were prepared as described previously 52  Measurement of in vivo alveolar fluid clearance in mice. AFC was measured as previously described 54 .
In brief, BALB/c male mice (20-30 g) were anaesthetized with diazepam (17.5 mg kg −1 , intraperitoneally) followed 6 min later by ketamine (450 mg kg −1 , intraperitoneally) and were placed on a heating pad. The trachea was exposed and cannulated with a trimmed 18-gauge intravenous catheter, which was then connected to a mouse respirator (HX-300, Chengdu Taimeng Technology Co. Ltd, Chengdu, China). Mice were ventilated with 100% O 2 with a 200 μ l tidal volume (8-10 ml kg −1 ) at a 160 breaths min −1 . Once stable anaesthesia was obtained, mice were positioned in the left decubitus position, and 300 μ l of isosmolar NaCl or 200 μ M formaldehyde containing 5% fatty acid-free bovine serum albumin was instilled via the tracheal cannula, followed by 100 μ l of room air to clear dead space. After instillation, mice were ventilated for a 60 min period, and then the alveolar fluid was aspirated. All reagents were added to the AFC instillate from stock solutions directly before instillation, in a minimal volume of solvent (1-10 μ l ml −1 ). The protein concentrations in aspirated solutions were measured by Bradford Method. We estimated AFC (% AFC 60 ) by the changes of concentration for bovine serum albumin as water was absorbed after 60 min. AFC was calculated as the follow equation: AFC = [(Vi − Vf)/Vi] × 100, where V is the volume of the instilled bovine serum albumin solution (i) and the final alveolar fluid (f). Vf = Vi × Pi/Pf, where P represents the protein concentration in the instilled bovine serum albumin solution (i) and the final alveolar fluid (f). Statistical analysis. Data were expressed as the mean ± SE. One-way ANOVA computations were used to analyze the difference of the means for normally distributed data. Mann-Whitney test was applied for nonparametric data. Significance levels were *P < 0.05 and **P < 0.01.