Lipopolysaccharide hyperpolarizes guinea pig airway epithelium by increasing the activities of the epithelial Na+ channel and the Na+-K+ pump

Michael W. Dodrill, Jeffrey S. Fedan

Abstract

Earlier, we found that systemic administration of lipopolysaccharide (LPS; 4 mg/kg) hyperpolarized the transepithelial potential difference (Vt) of tracheal epithelium in the isolated, perfused trachea (IPT) of the guinea pig 18 h after injection. As well, LPS increased the hyperpolarization component of the response to basolateral methacholine, and potentiated the epithelium-derived relaxing factor-mediated relaxation responses to hyperosmolar solutions applied to the apical membrane. We hypothesized that LPS stimulates the transepithelial movement of Na+ via the epithelial sodium channel (ENaC)/Na+-K+ pump axis, leading to hyperpolarization of Vt. LPS increased the Vt-depolarizing response to amiloride (10 μM), i.e., offset the effect of LPS, indicating that Na+ transport activity was increased. The functional activity of ENaC was measured in the IPT after short-circuiting the Na+-K+ pump with basolateral amphotericin B (7.5 μM). LPS had no effect on the hyperpolarization response to apical trypsin (100 U/ml) in the Ussing chamber, indicating that channel-activating proteases are not involved in the LPS-induced activation of ENaC. To assess Na+-K+ pump activity in the IPT, ENaC was short-circuited with apical amphotericin B. The greater Vt in the presence of amphotericin B in tracheas from LPS-treated animals compared with controls revealed that LPS increased Na+-K+ pump activity. This finding was confirmed in the Ussing chamber by inhibiting the Na+-K+ pump via extracellular K+ removal, loading the epithelium with Na+, and observing a greater hyperpolarization response to K+ restoration. Together, the findings of this study reveal that LPS hyperpolarizes the airway epithelium by increasing the activities of ENaC and the Na+-K+ pump.

  • endotoxin
  • lung
  • airway epithelium
  • Na+ transport
  • transepithelial voltage

the airway epithelium forms a barrier between the airway lumen and the lung. It performs many functions, one of which is the regulation of the composition and depth of the airway surface liquid through the transepithelial transport of ions to maintain proper function of the mucociliary escalator (41). The epithelial transport of Na+ is tightly regulated via the regulation of the open probability (48), membrane expression, and internalization (58) of the epithelial sodium channel (ENaC) and the turnover and membrane expression of the Na+-K+ pump (21). The regulation of ENaC activity is not fully understood, but factors that have been identified to regulate ENaC include IL-4 (19), TNFα (11), IL-1β (67), aldosterone (38), the serine proteases furin and prostasin (51), the ubiquitin ligases neuronal precursor cell expressed developmentally downregulated (Nedd)4–2 (56), cAMP (65), the thickness and osmolarity of the airway surface liquid (44), and pH (10). The transport of Na+ in the lung is also upregulated by catecholamines (1), steroids (2), and thyroid hormone (5), and inhibited by hypoxia (40, 47, 68). The Na+-K+ pump, which is the rate-limiting step in transepithelial Na+ transport (21), is regulated by aldosterone and nitric oxide (62).

LPS, which accumulates in the mammalian host upon infection with gram-negative bacteria, stimulates the expression of Toll-like receptors (TLR)2 and 4 in mouse lung epithelium (52) and the production of inflammatory cytokines via the TLR4-mediated activation of p38 (49), ERK1/2, and JNK in human alveolar epithelial cells (23). Both TLR4 and TLR2 are expressed in the human airway epithelium (23, 27, 54). The airway epithelium plays an important role in the immunological defense of the host by responding to and secreting inflammatory mediators (63). Many of these factors, including TNFα (11, 18), IL-4 (19), TGF-β (64), IL-1β (67), and nitric oxide (15), regulate the rate of airway epithelial Na+ transport.

Normal ENaC function is vital, and the severe consequences of improper ENaC function are manifested in cystic fibrosis, Liddle's syndrome, and pseudohypoaldosteronism type I. Mice deficient in αENaC are unable to clear liquid from their lungs and die after birth (31). ENaC hyperactivity in cystic fibrosis, due to defective Cl transport, and in Liddle's syndrome, resulting from the inability to ubiquitinate ENaC due to a mutated PY motif in βENaC, leads to a depletion of the airway surface liquid, plugging of the airways, and susceptibility to lung infection, whereas ENaC hypoactivity in pseudohypoaldosteronism type I causes lung edema (50, 51, 56).

LPS modifies the reactivity of the airways to contractile and relaxing agents and affects the bioelectric behavior of the airway epithelium. Endotoxins cause attacks in some types of asthma (25) and are thought to be the causative agent for byssinosis (9). Lung injury from endotoxin-induced inflammation has also been identified as a trigger for acute respiratory distress syndrome (25). Earlier, we observed that administration of LPS in vitro to the guinea pig isolated, perfused trachea (IPT) inhibited the contractile response to methacholine and histamine (17). These effects of LPS on the reactivity to methacholine and histamine involve nitric oxide, which does not seem to play any role in the absence of LPS (17). Systemic LPS potentiated, in the IPT, the relaxant effect of hyperosmolar solution applied apically to the epithelium, which is mediated by the release of epithelium-derived relaxing factor (EpDRF), but had no effect on the reactivity of the airway smooth muscle to methacholine (33). It has been suggested that the regulation of the airway smooth muscle tone by EpDRF is linked to epithelial ion transport (14). Indeed, systemic LPS increased basal Vt, increased the hyperpolarization component of the biphasic methacholine concentration-response curve for bioelectrical responses to methacholine, and potentiated hypertonic NaCl-induced depolarization (33). The effects of systemic LPS require the presence of LPS binding protein, needed for the activation of TLR4 by LPS (61). Therefore, we earlier investigated the in vitro effects of LPS and cytokines on the reactivity of the airway smooth muscle and epithelium to methacholine and hyperosmolarity in the absence of LPS binding protein in the IPT and the Ussing chamber (32). Incubating tracheas with LPS, IL-1β, IL-4, IL-13, IFNγ, and TNFα alone or in combination caused effects similar to those that in vivo LPS treatment had on mechanical responses of the airways to methacholine and hyperosmolar solutions, but had varying effects on Vt and bioelectric responses of the epithelium that did not entirely reproduce the effects of LPS given in vivo (32).

In this study, we investigated the effects of systemically administered LPS on Na+ transport in the guinea pig airway epithelium. The effects of LPS on ENaC and Na+-K+ pump function were investigated using the IPT and the Ussing chamber preparations. We found that LPS hyperpolarizes the airway epithelium and that the effect of amiloride on Vt is larger in the epithelium from LPS-treated animals compared with controls (33), i.e., amiloride offsets the effect of LPS on Vt. Therefore, we hypothesized that LPS hyperpolarizes the epithelium by increasing the transepithelial transport of Na+. Using the IPT, we measured the effects of LPS on the activities of ENaC and the Na+-K+ pump in the epithelium. LPS increased the activities of both ENaC and the Na+-K+ pump. Additionally, the effects of LPS on the activity of the Na+-K+ pump were measured in the Ussing chamber using a Na+-loading technique, and the increase in Na+-K+ pump activity by LPS was confirmed. The effects of LPS on the proteolytic activation of ENaC in the epithelium were measured using apical trypsin in the Ussing chamber. The findings indicated that LPS does not increase ENaC activation via proteolytic cleavage. Together, the results suggest that LPS increases airway epithelial Na+ transport by increasing the activities of both ENaC and the Na+-K+ pump and that the effect of LPS on ENaC activity does not involve proteolytic activation.

MATERIALS AND METHODS

Animals.

These studies were conducted in facilities accredited fully by the Association for the Assessment and Accreditation of Laboratory Animal Care International, and the research protocol was approved by the National Institute for Occupational Safety and Health Institutional Animal Care and Use Committee. Male Hartley guinea pigs (Crl:Ha 600–700 g) from Charles River Laboratories (Wilmington, MA), monitored free of endogenous viral pathogens, parasites, and bacteria, were used in all experiments. The animals were acclimated before use and were housed in filtered ventilated cages on Alpha-Dri virgin cellulose chips and hardwood Beta chips as bedding, provided HEPA-filtered air, Teklad 7906 diet and tap water ad libitum, under controlled light cycle (12-h light) and temperature (22–25°C) conditions. The animals were anesthetized with pentobarbital sodium (65 mg/kg ip) and killed by thoracotomy and bleeding before removing the trachea.

IPT.

The IPT, described previously (16, 43), can be used to measure simultaneously the effects of agents on airway diameter and Vt (14). Briefly, 18 h after the in vivo administration of LPS (4 mg/kg ip) or a volume equivalent of saline, the animal was anesthetized, and a 4.2-cm section of trachea was excised, cleaned in a dish filled with modified Krebs-Henseleit (MKH) solution aerated with 95% O2/5% CO2, and mounted on a perfusion holder. The holder was placed in a bath of MKH (basolateral bath), while the lumen was perfused with MKH from a separate (apical) bath of MKH at a rate of 20 ml/min. Both baths were maintained at 37°C and aerated with 95% O2/5% CO2. Agents could be added to either the apical or the basolateral bath. As described previously (14), indwelling cannulas inserted at the inlet and outlet ends of the tracheal lumen had side holes connected to the positive and negative sides, respectively, of a differential pressure transducer. Changes in the diameter of the trachea could be detected by measuring changes in the difference between the inlet minus the outlet pressures (▵P), but such measurements were not taken for this study. Vt was measured under open-circuit conditions using Ag/AgCl electrodes with 9% saline bridges in 4% agar. The apical electrode consisted of the side holes of the outlet cannula and the basolateral electrode consisted of a J-tube placed in the basolateral bath 1 cm from the wall at the center of the trachea. Both electrodes were matched to within 2 mV and connected to a voltage/current clamp amplifier (DVC-1000; World Precision Instruments, Sarasota, FL); the offset was adjusted to 0 mV.

Ussing chamber.

The Ussing chamber (CHM8 chamber, World Precision Instruments) was used to measure the effects of LPS treatment and other agents on Vt and transepithelial resistance (Rt) of the tracheal epithelium. A 4-cm section of the trachea was excised from the animal, cleaned in MKH, cut along the smooth muscle band, and mounted over the aperture (0.125 cm2) between the hemichambers. The apical and basolateral faces of the trachea were each bathed with 5 ml of MKH at 37°C and aerated with 95% O2/5% CO2. Vt and Rt were measured under open-circuit conditions using Ag/AgCl electrodes (0.9% NaCl in 4% agar). One pair of electrodes was used to measure voltage; the other was used to apply 5 μA current pulses for 5 s every 50 s to measure a voltage deflection for the calculation of Rt from Ohm's law. Each pair of electrodes was matched to less than 2 mV, and the offset was adjusted to 0.

Materials.

LPS from Salmonella enterica serotype typhimurium (phenol-extracted) was prepared in saline. MKH solution (pH 7.4, osmolarity of 281 ± 5 mosM and temperature of 37°C) contained 113 mM NaCl, 4.8 mM KCl, 2.5 mM CaCl2, 1.2 mM KH2PO4, 1.2 mM MgSO4, 25 mM NaHCO3, and 5.7 mM glucose, and was saturated with 95% O2/5% CO2. K+-free MKH lacked KCl and KH2PO4, but contained 119 mM NaCl. Reduced Na+ MKH was prepared by omitting NaCl and replacing it with N-methyl-d-glucamine (NMDG)-Cl (113 mM). Amiloride and amphotericin B from Streptomyces were prepared in dimethyl sulfoxide (2.6 mM and 13.5 mM, respectively). Ouabain and trypsin type IX-S from porcine pancreas were dissolved in saline. All reagents were from Sigma-Aldrich (St. Louis, MO).

Data analysis.

The unpaired or paired Student's t-tests were used where appropriate with a significance threshold of 5% (SigmaStat version 3.1, Systat Software). Otherwise, the Wilcoxon signed rank test or the Mann-Whitney rank sum test were used where appropriate. Error bars represent the standard errors of the mean. P < 0.05 was regarded as significant.

RESULTS

LPS hyperpolarizes the airway epithelium by increasing the transport of Na+.

Basal Vt measured in the IPT was significantly greater in tracheas from LPS-treated animals (−34.2 ± 2.7 mV) compared with controls (−5.5 ± 0.7 mV) (Fig. 1). The ENaC blocker amiloride (10 μM) (30) was added apically to determine the role of epithelial Na+ transport in the hyperpolarization. Amiloride inhibited Vt in tracheas from both saline- and LPS-treated animals (Fig. 1). Vt in the presence of amiloride was −4.5 ± 0.9 mV in saline-treated animals and −13.1 ± 1.3 mV in the LPS-treated group. The effect of amiloride, or ▵Vt, was significantly greater (Mann-Whitney rank sum test) in tracheas from LPS-treated animals (21.1 ± 2.8 mV) compared with saline-treated animals (1.0 ± 0.3 mV). These findings indicate that LPS hyperpolarizes the airway epithelium, at least in part, by increasing Na+ transport via ENaC.

Fig. 1.

LPS increases epithelial sodium channel (ENaC) activity in the airway epithelium. The isolated, perfused trachea (IPT) was used to measure amiloride-sensitive transepithelial potential difference (Vt) in tracheas from saline- and LPS-treated guinea pigs. Basal Vt was greater in tracheas from LPS-treated animals. Amiloride (10 μM) inhibited Vt (▵Vt) in both treatment groups, but to a larger extent after LPS treatment. The effect of amiloride on Vt was greater in tracheas from LPS-treated animals compared with controls. Therefore, the LPS-induced hyperpolarization is due, in part, to an increase in Na+ transport. Saline, n = 6; LPS, n = 5. *P < 0.05 compared with saline-treated controls (−Amiloride; Wilcoxon signed rank test). **P < 0.05 compared with saline-treated controls (−Amiloride). ***P < 0.05 compared with LPS-treated (−Amiloride).

LPS increases the activity of ENaC.

The main route for Na+ transport across the airway epithelium is via ENaC in the apical membrane and the Na+-K+ pump in the basolateral membrane (6). LPS could increase transcellular Na+ transport by increasing the open probability and/or abundance of ENaC (4) and/or by increasing the activity and/or abundance of the Na+-K+ pump (22). To measure the effects of LPS on ENaC activity in the IPT, the Na+-K+ pump was short-circuited with basolateral amphotericin B (7.5 μM), thereby removing the contribution of the Na+-K+ pump to Vt (Fig. 2). Concurrently with the administration of amphotericin B, Na+ was replaced in the basolateral MKH with NMDG-Cl to establish a large Na+ gradient across the apical membrane (Fig. 2). Consequently, any effect of LPS on the activity of ENaC would be measurable as a change in Vt. The response to NMDG-Cl/amphotericin B was complex and consisted of an initial hyperpolarization followed by a decrease to a steady state. The transient hyperpolarization was not different between treatment groups (Fig. 3A). To understand the transient hyperpolarization response, we investigated if it was due to permeabilization or sodium replacement. Nystatin (7.5 μM), a similar pore-forming agent, caused similar transient Vt responses after apical (n = 6) and basolateral (n = 1) application in the Ussing chamber (data not shown). This effect was not observed when amphotericin B was added alone, but was caused by the removal of basolateral sodium whether before (n = 2), concurrently with (n = 3), or after (n = 2) the addition of amphotericin B (data not shown). Stable Vt was greater in the tracheas from LPS-treated animals, indicating that LPS increases the activity of ENaC (Fig. 3B).

Fig. 2.

LPS increases ENaC activity. Representative tracings from tracheas from a saline- and an LPS-treated animal demonstrate the experimental design. The contribution of the Na+-K+ pump to Vt was removed before measuring ENaC activity in the IPT by short-circuiting the basolateral membrane with amphotericin B (AB; 7.5 μM). To create a large gradient for Na+ across the apical membrane, Na+ was concurrently removed from the modified Krebs-Henseleit (MKH) of the basolateral bath by replacing the NaCl with NMDG-Cl (113 mM). The left arrows indicate an artifact due to turning off the preamplifier while the basolateral bath was drained and replaced with NMDG-Cl- and amphotericin B-containing MKH. The right arrows indicate the point of Vt measurement. Amiloride (A; 10 μM) and trypsin (T; 100 U/ml) were added to the apical bath. Saline, n = 7; LPS, n = 5.

Fig. 3.

Summary of the effects of LPS on ENaC activity in the IPT preparation. See Fig. 2 for the protocol of the experiments whose results are summarized in this figure. A: the maximum, transient hyperpolarization effect of basolateral Na+ removal and amphotericin B addition on Vt before Vt stabilized was not different between the epithelia of tracheas from saline- and LPS-treated animals. B: the stable Vt response to Na+ removal and amphotericin B addition was greater in tracheas from LPS-treated animals compared with controls (*P < 0.05, t-test). C: the Vt response to amiloride was greater in tracheas from LPS-treated animals compared with control (*P < 0.05, Mann-Whitney rank sum test). D: Vt in the presence of trypsin was not different between tracheas from saline- or LPS-treated animals. E: the ▵Vt response to trypsin was not different between tracheas from saline- and LPS-treated animals (Mann-Whitney rank sum test). Saline, n = 8; LPS, n = 5.

To confirm the role of ENaC in the LPS-induced hyperpolarization under these conditions, apical amiloride (10 μM) was subsequently added. The effect of amiloride was greater in the tracheas from LPS-treated animals compared with controls (Fig. 3C), as it had been in the presence of normal basolateral MKH. Finally, apical trypsin (100 U/ml) was added to activate ENaC. This concentration is within the range reported by the literature that has been used to activate ENaC directly. Hughey et al. (28) and Sheng et al. (55) used a smaller concentration of 28.2 U/ml to activate ENaC expressed in Xenopus oocytes, whereas Bridges et al. (7) used a larger concentration of 336 U/ml. Vt in the presence of trypsin was not different between tracheas from control and LPS-treated guinea pigs (Fig. 3D). Trypsin also had little effect (Fig. 3E) in the presence of amiloride, demonstrating that amiloride had effectively inhibited ENaC. This finding further supports the conclusion that ENaC plays a major role in the LPS-induced hyperpolarization of the airway epithelium.

LPS does not affect the proteolytic activation of ENaC.

The finding that LPS increases the activity of ENaC led us to ask if the mechanism involves increases in the proteolytic activation of ENaC. It has been recently discovered that endogenous serine proteases, including furin and prostasin, activate ENaC via the cleavage of an inhibitory domain from the channel's α- and γ-subunits (28, 30, 34). It is thought that the membrane-bound pool of ENaC exists in two states: an uncleaved, near-silent, immature state and a cleaved, activated, or mature state (29). The regulation of these serine proteases is not fully understood, but it is thought that ENaC activity can be increased in renal and respiratory epithelia via protease activation or disinhibition, thereby shifting the balance of membrane ENaC from the near-silent state to the mature state (50, 51, 57). Trypsin has been used to activate ENaC in ENaC-transfected patch-clamped oocytes (3, 9, 10, 56), cystic fibrotic human bronchial epithelial cells (7), human airway epithelial cells (13, 60), and increase amiloride-sensitive Na+ transport in rat (60) and mouse trachea (57).

To determine the effect of LPS on the balance of near-silent vs. mature ENaC, we administered apical trypsin (100 U/ml) in the Ussing chamber while measuring Vt. A greater Vt response to trypsin would reflect a greater pool of near-silent ENaC. If LPS increases cleavage-mediated activation of ENaC, we hypothesized that the Vt response to trypsin should be reduced compared with controls. LPS increased basal Vt, as we found earlier, and basal equivalent transepithelial current (Isc) (Table 1), but did not affect the basal resistance of the epithelium. Trypsin increased Vt and Rt, but did not affect Isc, in tracheas from both control and LPS-treated animals. The trypsin-induced changes in Vt and Rt (▵Vt and ▵Rt), and Isc, were, however, not different between the two groups. As well, the %▵Vt/%▵Rt ratio was not different between the untreated and LPS-treated groups. These results indicate that the effects of LPS on the Vt, Rt, and Isc responses to trypsin cannot be explained by an increase in transepithelial ion transport. The small increases in Rt by trypsin are surprising and are not understood at present, but have been observed by Swystun et al. (60).

View this table:
Table 1.

Effects of trypsin on Vt, Rt, and Isc in tracheas from untreated and LPS-treated animals

Greater abundance of mature ENaC after LPS should affect responses to amiloride after trypsin treatment. To investigate this possibility, apical amiloride was administered in the Ussing chamber to tracheas from saline- and LPS-treated animals before and after addition of trypsin (Fig. 4). Apical amiloride was added first to measure basal ENaC activity and washed out, apical trypsin was added to measure channel activation, and, last, apical amiloride was administered in the presence of trypsin. As in the IPT (Fig. 1), depolarization by amiloride was greater in the tracheas from LPS-treated animals compared with controls (Fig. 5), but the Vt responses to trypsin were not different (Fig. 5). This finding supports the evidence that the numbers of membrane-bound near-silent channels in the airway epithelium are not affected by LPS treatment. The Vt responses to apical amiloride in the presence of trypsin remained greater in tracheas from LPS-treated animals compared with controls (Fig. 5), indicating that ENaC activity remains elevated after the activation of the near-silent pool of ENaC. The Vt responses to amiloride in the absence and presence of trypsin were not different from each other within both the saline- and LPS-treated groups, suggesting that the numbers of the membrane-bound near-silent ENaC substrates were small compared with the proteolytically activated pools. Together, these data suggest that trypsin had no significant effect on ENaC activity because most of the membrane-bound population of ENaC is in the proteolytically activated state. Even if LPS would lead to activation or disinhibition of channel-activating proteases, the number of membrane-bound near-silent ENaC appears to be sufficiently small for them to have no effect on Vt. Therefore, proteolytic activation of ENaC appears not to be involved in the hyperpolarizing effect of LPS in the epithelium, and the abundance and/or ratio of silent and active ENaC is seemingly not affected.

Fig. 4.

Representative tracings from tracheas from a saline-treated and an LPS-treated animal demonstrate the experimental design in which trypsin was used in the Ussing chamber to assess the effects of LPS on proteolytic activation of ENaC. Amiloride (A; 10 μM) was added to the apical bath to measure ENaC activity. W, both the apical and basolateral baths were washed with fresh MKH. Trypsin (T; 100 U/ml) was added to the apical bath to activate ENaC.

Fig. 5.

The stimulation of ENaC activity by LPS does not involve channel-activating proteases. See Fig. 5 for the protocol for which results are summarized in this figure. The effects of amiloride (10 μM), trypsin (100 U/ml), and amiloride (10 μM) in the presence of trypsin are expressed as the change in Vt after the administration of the agent. *P < 0.05 (t-test). Saline and LPS, n = 11.

LPS increases Na+-K+ pump activity.

The finding that LPS increases the activity of ENaC raised the question of whether LPS also affected Na+-K+ pump activity. Did increased ENaC activity and intracellular Na+ concentrations stimulate the Na+-K+ pump, or did LPS stimulate the activity of the Na+-K+ pump? To address that question, the IPT was used to measure the contribution of the Na+-K+ pump to Vt in tracheas from saline- and LPS-treated animals after negating the involvement of ENaC. Amiloride was applied to the apical bath, and apical amphotericin B (7.5 μM) was administered to short-circuit ENaC and remove its contribution to Vt (Fig. 6A). Vt at this point, due entirely to the Na+-K+ pump, was more than doubled in tracheas from the LPS-treated group compared with controls, indicating that LPS increases Na+-K+ pump turnover (Fig. 6B). Finally, apical trypsin was added to confirm that ENaC was inhibited by amiloride, short-circuited by amphotericin B, and did not contribute to Vt. Even though Vt in the presence of trypsin was doubled in the tracheas from LPS-treated animals compared with controls (Fig. 6C), there was no significant effect of trypsin on Vt (Fig. 6D). Together, these findings indicate that LPS increases the turnover of the Na+-K+ pump in addition to its effects on ENaC.

Fig. 6.

LPS stimulates the activity of the Na+-K+ pump in the IPT: amphotericin B approach. A: representative tracings of Vt responses in tracheas from saline- and LPS-treated animals demonstrate the experimental design. Amiloride (A; 10 μM) was added to the apical bath to remove the contribution of ENaC from Vt. Amphotericin B (AB; 7.5 μM) was added to the apical bath to short-circuit the apical membrane of the epithelium. Trypsin (T; 100 U/ml) was added to the apical bath to verify that ENaC had been inhibited by amiloride and short-circuited by amphotericin B. B: the effect of amphotericin B on Vt, measured at the points indicated by the single-headed arrows in A, was greater in the epithelium of tracheas from LPS-treated animals compared with controls. C: Vt measured in the presence of apical trypsin, at the points indicated by the double-headed arrows in A, was greater in tracheas from LPS-treated tracheas. D: the Vt response to the apical administration of trypsin (100 U/ml) was not different between groups. Since trypsin had no effect on Vt, the greater Vt was not due to ENaC. Saline, n = 6; LPS, n = 5. *P < 0.05 compared with saline-treated control (t-test).

Additional evidence was sought to determine whether LPS increases Na+-K+ pump activity by adapting a Na+-loading approach, used previously in rabbit arteriole myocytes (26), to measure tracheal epithelial Na+-K+ pump activity in the Ussing chamber. Because it has an absolute requirement for extracellular K+, the pump was inhibited by removing K+ from the apical and basolateral MKH solution (Fig. 7A). Being electrogenic, the inhibition of the Na+-K+ pump resulted in a depolarization of the epithelium during accumulation of intracellular Na+. The effect of K+ removal on Vt was greater in tracheas from LPS-treated animals compared with controls (Fig. 7B), reflecting the greater contribution of pump current in the tracheas from LPS-treated animals. Basolateral K+ was then restored to the MKH solution to stimulate pump activity and the epithelium repolarized (Fig. 7A). Apical K+ was subsequently restored and had no effect on Vt (Fig. 7A). The repolarization response after basolateral and apical K+ restoration was greater in the tracheas from LPS-treated animals compared with controls (Fig. 7B). This finding provides additional evidence that LPS increases the activity of the Na+-K+ pump. Finally, the Na+-K+ pump inhibitor ouabain (10 μM) was added basolaterally to confirm the contribution of the Na+-K+ pump to Vt, whereupon Vt was decreased. Even though the Vt responses to ouabain were not different between the tracheas from saline- and LPS-treated animals, they were, however, both as large as the responses to K+ removal, indicating that pump activity was restored by K+ restoration (Fig. 7B). Together, the results indicate that LPS increases the activity of the Na+-K+ pump.

Fig. 7.

LPS stimulates the activity of the Na+-K+ pump in the Ussing chamber: Na+-loading approach. A: representative tracings demonstrate the Na+-loading technique to measure the effects of LPS on the activity of the Na+-K+ pump in the Ussing chamber. At this compressed time scale, the upward deflections in Vt resulting from application of the current pulses are compressed together; thus, the lower edge of the signal represents Vt and the upper edge of the signal represents the deflections in Vt due to the applied current pulses. K+-free, K+ was removed from both the apical and basolateral baths by omitting KCl and KH2PO4 from the MKH solution and increasing the NaCl concentration to 119 mM. K+ removal inhibited the Na+-K+ pump, Na+ accumulated within the epithelium, and the epithelium depolarized. Arrows indicate noise in the signal during K+ removal and NaCl addition. Basolateral K+ was restored by readmitting KCl (B K+) to the basolateral bath. Apical K+ was restored by adding KCl (A K+) to the apical bath. K+ restoration allowed Na+-K+ pump activity to resume and the epithelium to repolarize. A greater Vt response to K+ restoration was interpreted as indicative of increased Na+-K+ pump activity. Ouabain (O; 10 μM) was administered to the basolateral bath to confirm that the K+-restoration maneuver restored the activity of the Na+-K+ pump. B: the effects of K+ removal, K+ restoration, and ouabain (10 μM) are expressed as the difference in Vt before and after the intervention. The effect of K+ restoration takes into account responses to both the basolateral and apical additions of KCl, the latter of which were negligible. Saline and LPS, n = 5. *P < 0.05 (t-test).

DISCUSSION

The first major finding of this study is that, 18 h after its systemic administration, LPS causes hyperpolarization of the tracheal epithelium by increasing the transport of Na+. This was demonstrated by the greater inhibition of Vt by amiloride in the epithelia in IPT and Ussing experiments from LPS-treated animals compared with controls. The second major finding, using exogenous trypsin, was that LPS increased Na+ transport in a manner that seems to be independent of proteolytic cleavage of ENaC. Third, two different approaches for measuring the activity of the Na+-K+ pump revealed that it was increased by LPS. Together, these results indicate that LPS upregulates the activities of both ENaC and the Na+-K+ pump.

LPS plays a role in airway responsiveness in several diseases that are marked by dysregulation of airway tone. Earlier, the effects of intraperitoneally administered LPS on guinea pig airway smooth muscle reactivity and Vt were investigated in the IPT, and it was found that LPS reduced the reactivity to inhaled methacholine and potentiated the EpDRF-mediated relaxation responses to hyperosmolar solutions in vitro (33). These effects were associated with a potentiation of the hyperpolarization portion of the biphasic Vt response to methacholine and a potentiation of hyperosmolar NaCl-induced depolarization of the airway epithelium (33). Since the main route of transepithelial Na+ transport is through the ENaC/Na+-K+ pump axis, this study investigated the effects of LPS on the activities of ENaC and the Na+-K+ pump in functional experiments.

Using amiloride, the role of Na+ transport in the effects of LPS on Vt were determined in the IPT. Amiloride inhibited Vt in both treatment groups, yet inhibited Vt in tracheas from the LPS-treated group to a greater extent. The Vt response to Na+ removal was different from what was predicted. In an earlier report of the permeabilization of the basolateral membranes of rat primary alveolar epithelial cell monolayers, a similar Na+ removal maneuver (replacement of 116 mM of a total of 141 mM NaCl with 116 mM NMDG-Cl in the basolateral Ringer solution) done concurrently with the administration of basolateral amphotericin B at the same concentration (7.5 μM) resulted in a lasting and stable increase in transmonolayer current (40). In our experiment, Na+ removal hyperpolarized the epithelium only transiently before Vt became depolarized (Fig. 2), and LPS had no affect on the magnitude of the transient hyperpolarization (Fig. 3A). Another study using apical nystatin, a similar pore-forming antifungal agent, to short-circuit ENaC in cultured human bronchial epithelial cell monolayers in the Ussing chamber without removing Na+ reported a similar effect, i.e., a transient increase in short-circuit current before returning to baseline (37). The authors (37) concluded that the Na+-K+ pump decreased short-circuit current by clearing an intracellular Na+ load. We tested nystatin (7.5 μM) and observed similar transient Vt responses. In contrast, however, the transient Vt responses did not occur upon the addition of amphotericin B without the removal of basolateral Na+, but occurred upon basolateral Na+ removal whether Na+ was removed before, concurrently with, or after the addition of amphotericin B. That is, this complex response was attributable to the replacement of Na+ by NMDG, not to amphotericin B per se.

Apically applied trypsin increased Vt and Rt, but had no effect on Isc in tracheas from saline- and LPS-treated animals (Table 1). Vt was increased by the same proportion as Rt in both treatment groups, and trypsin had no effect on Isc. These findings indicate that trypsin had no effect on the activity of ENaC and are compatible with the very small effect of amiloride on the tracheal epithelium from control animals (Fig. 1). We considered the possibility that protease-activated receptors (PAR) may be involved in the effects of trypsin. Trypsin is an agonist of PARs, which stimulate an increase in the intracellular concentration of Ca2+ and PKC activation. The activation of PAR receptors has been found to inhibit the Ca2+ response of guinea pig tracheal epithelial cells in culture to LPS (45). Small concentrations of trypsin (1–10 ng/ml) below those that cleave ENaC directly were found to activate ENaC via an indirect, receptor-mediated mechanism (3). However, PAR-2 overexpression or knockdown had no influence on the effects of trypsin in that study. Additionally, Swystun et al. (60) found that apical trypsin increased Na+ and Cl transport, and Isc and Rt in the airway epithelium. Due to the lack of effect of ouabain on the trypsin-induced effect on Rt in that study, the authors concluded that Rt was increased due to a decrease in paracellular conductance. We are unaware of any literature linking PARs and ENaC and are left with a lack of understanding about why trypsin increases Rt (until detailed future experiments can be performed).

The larger effect of amiloride on Vt could be explained two ways: ENaC activity could be stimulated by a reduction in intracellular Na+ by the Na+-K+ pump (22), or increased ENaC activity could stimulate the Na+-K+ pump by elevating intracellular Na+ concentrations (42). Since amiloride-sensitive Vt was increased by LPS regardless of whether or not the Na+-K+ pump was short-circuited with amphotericin B, ENaC activity was stimulated by LPS independently of the intracellular Na+ concentration. ENaC is regulated by many complex mechanisms, including open probability, the number of mature and near-silent channels in the membrane, protein abundance, localization, and channel kinetics and degradation. We have examined the effects of LPS on ENaC message and abundance (data not shown) and observed that it had no effect on the abundance or cleavage of the α-, β-, or γ-subunits of ENaC in the tracheal epithelium.

Several studies have investigated the effects of Pseudomonas aeruginosa on the transcription or activity of ENaC. Intratracheal instillation of P. aeruginosa inhibits distal alveolar fluid clearance with a decrease in the transcription of βENaC, but without an effect on αENaC or α1-Na+-K+-ATPase transcription (46). In a similar study, the intratracheal instillation of P. aeruginosa in mice reduced the transcription of αENaC, but had no effect on the transcription of β- or γENaC or α1-Na+-K+-ATPase (12). In another study, after dog bronchial epithelium was treated with a culture of P. aeruginosa, short-circuit current, active Na+ absorption, and Cl fluxes were decreased (59). In contrast to our findings with LPS, these studies, the results of which no doubt involve P. aeruginosa LPS, indicate that the infection of the lung with P. aeruginosa inhibits ENaC activity. They differ from our study, where we injected extracted LPS intraperitoneally, in that the bacterium was instilled into the lung. The instillation resulted in inflammation and edema, which were absent in the guinea pig lung 18 h postinjection (33). LPS also inhibits transepithelial Na+ transport in vitro. Amiloride-sensitive short-circuit current is inhibited by LPS in the mouse trachea in the Ussing chamber (35). Although the ionic basis of the effect was not examined, we observed that incubation of guinea pig trachea in vitro with LPS reduced Vt (32).

Since the Na+-K+ pump has been identified as the rate-limiting step for Na+ transport across the airway epithelium (39), another major emphasis of this study was to investigate the effects of LPS on its activity. The experimental design whereby the contribution of the Na+-K+ pump to Vt was measured in the IPT after short-circuiting ENaC using amphotericin B was adapted from one used in the Ussing chamber (11, 40). Using this technique to isolate Na+-K+ pump activity, Vt was twice as large in tracheas from LPS-treated animals compared with controls, indicating initially that LPS increases the activity of the Na+-K+ pump. Additional support for this notion was provided using the Na+-loading approach that has been employed previously to measure pump activity in smooth muscle (8). The magnitude of the repolarization after loading the epithelium with Na+ reflects the activity of the Na+-K+ pump and was increased by LPS.

The mechanisms through which LPS affects the activities of the Na+-K+ pump cannot be determined from the results of this study, but could involve the expression and/or abundance of the pump and/or its regulation. In our laboratory, we have determined that the expression of the α1-subunit of Na+-K+-ATPase is increased by LPS (data not shown). Several other factors that increase transepithelial Na+ transport by increasing the activity of the Na+-K+ pump could contribute to the observed effects of LPS. FXYD5, found in the airway epithelium, belongs to the FXYD family of proteins, which has been described as an auxiliary γ-subunit of the Na+-K+ pump (22). Each isoform has a unique tissue distribution and regulatory effects on Na+-K+ pump activity (20). The airway epithelium responds to an increase in intracellular Na+ by increasing the expression of FXYD5, which, in turn, binds to and increases the activity of the Na+-K+ pump (42). In the cystic fibrotic airway epithelium, the abnormally high intracellular Na+ concentration stimulates the activity of the Na+-K+ pump by increasing the expression of FXYD5 (42). After the permeabilization of the apical membrane with amphotericin B in our experiments, the intracellular concentration of Na+ in the epithelium of tracheas from both saline- and LPS-treated animals would approximate that of the apical MKH bath, i.e., 119 mM. Therefore, it is convenient to speculate that LPS increases the expression of FXYD in the airway epithelium, which would increase both the response to intracellular Na+ and the basal activity of the Na+-K+ pump.

We also earlier found that the cyclooxygenase inhibitor, indomethacin, inhibited Vt to a significantly greater extent in the tracheas from LPS-treated animals compared with controls (33). Even though the effect of indomethacin (33) was not as great as that of amiloride (Fig. 1), it indicates that cyclooxygenases also play a role in the LPS-induced hyperpolarization of the airway epithelium, suggesting a possible avenue of exploration in future experiments.

Hypoxia and reactive oxygen species have been found to reduce Na+-K+ pump activity. These factors, present in lung inflammation, cause edema by activating AMPK, which activates PKC, which, in turn, stimulates the endocytosis of the Na+-K+ pump from the basolateral membrane (24, 40, 62, 66). The LPS-stimulated upregulation of nitric oxide synthases has been reported to reduce the activity of the Na+-K+ pump in the guinea pig kidney (53) and to reduce the protein expression of the α1-subunit of the Na+-K+ pump in the rat sciatic nerve (36). The effects of LPS appear to be mediated via an interaction between LPS and the cell because they were reversed in the latter study by polymyxin B, which binds to the lipid A portion of LPS. In our guinea pig experimental model, however, the effects of systemic LPS occur in the absence of inflammation in the alveoli, bronchioles, or bronchi (33), and, no doubt, reflect the concerted actions of proinflammatory mediators whose abundance is stimulated by LPS.

In conclusion, the systemic administration of LPS hyperpolarizes the airway epithelium, at least in part, by stimulating the transport of Na+. LPS increases the activities of both ENaC and the Na+-K+ pump. The mechanisms through which Na+ transport is increased cannot be determined from the present experiments, yet it seems that proteolytic activation of ENaC is not involved. In a separate investigation, we observed that the expression and abundance of Na+-K+-ATPase but not ENaC are increased following LPS treatment (data not shown).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s). The findings and conclusions in this report are those of the authors and do not necessarily represent the views of the National Institute for Occupational Safety and Health. The mention of brand name does not constitute product endorsement.

ACKNOWLEDGMENTS

We thank Janet Thompson for technical assistance.

REFERENCES

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.
  67. 67.
  68. 68.
View Abstract