Abstract
Cystic fibrosis, an autosomal recessive disease frequently diagnosed in the Caucasian population, is characterized by deficient Cl- transport due to mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene. A second major hall-mark of the disease is Na+ hyperabsorption by the airways, mediated by the epithelial Na+ channel (ENaC). In this study, we report that in human airway epithelial CF15 cells treated with the CFTR corrector miglustat (n-butyldeoxynojyrimicin), whole-cell patch-clamp experiments showed reduced amiloride-sensitive ENaC current in parallel with a rescue of defective CFTR Cl- channel activity activated by forskolin and genistein. Similar results were obtained with cells maintained in culture at 27°C for 24 h before electrophysiology experiments. With monolayers of polarized CF15 cells, short-circuit current (Isc) measurements also show normalization of Na+ and Cl- currents. In excised nasal epithelium of cftrF508del/F508del mice, like with CF15 cells, we found normalization of amiloride-sensitive Isc. Moreover, oral administration of miglustat (6 days) decreased the amiloride-sensitive Isc in cftrF508del/F508del mice but had no effect on cftr-/- mice. Our results thus show that rescuing the trafficking-deficient F508del-CFTR by miglustat down-regulates Na+ absorption. A miglustat-based treatment of CF patients may thus have a beneficial effect both on Cl- and Na+ transports.
The cystic fibrosis disease is the most common lethal genetic disorder in the Caucasian population with also a worldwide distribution. CF is characterized by a defective cAMP-regulated Cl- conductance due to mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene (Riordan, 1993). One of these mutations, F508del, is the most common in CF patients, leading to abnormal trafficking of CFTR protein, which is retained in the ER (Cheng et al., 1990; Kartner et al., 1991; Pind et al., 1994). This molecular mechanism makes F508del a prototype of class II (trafficking-deficient) mutations (Welsh and Smith, 1993). CF tissues are also characterized by enhanced Na+ absorption mediated by the epithelial Na+ channel ENaC (for review, see Boucher, 2007). Abnormal Na+ transport by CF airway epithelia has been demonstrated by many in vivo and in vitro observations in human and mice, and an increased amiloride-sensitive transepithelial potential is used as a diagnostic criterion in CF (Knowles et al., 1981, 1983; Boucher et al., 1986; Grubb and Boucher, 1997; Mall et al., 1998). ENaC is expressed in a variety of epithelial tissues including airways, renal collecting duct, urinary bladder, colon, and sweat and salivary glands (Canessa et al., 1994; Barbry and Ladzunski, 1996).
It is now well documented that CFTR also regulates many transport proteins and cellular functions due to large and dynamic macromolecular complexes that contain CFTR, signaling molecules and transport proteins (for review, see Kunzelmann et al., 2000; Guggino and Stanton, 2006). In favor of this concept, several studies showed that the functional interaction between CFTR and ENaC regulates both epithelial Cl- and Na+ conductances (Guggino and Stanton, 2006). However, it remains unclear why CF tissues display such a high ENaC activity. In addition, no studies have examined the effect of F508del-CFTR correction on Na+ hyperabsorption in native tissues. Finally, other studies argued against a regulation of ENaC by CFTR after coexpression in Xenopus laevis oocytes (Nagel et al., 2005).
We recently showed that miglustat rescues the trafficking-deficient F508del-CFTR to the plasma membrane in human airway epithelial cells but also in the intestine of F508del-CFTR mice (Norez et al., 2006; Antigny et al., 2008). Miglustat is now evaluated in CF patients within a pilot phase 2a clinical trial (http://clinicaltrials.gov/). In this report, we addressed the question whether miglustat, by rescuing F508del-CFTR abnormal trafficking, also down-regulates ENaC-dependent sodium hyperabsorption. To this aim, we have studied endogenous CFTR and ENaC channels in miglustat-corrected human airway epithelial CF15 cells and in cftrF508del/F508del mice. We demonstrate that the rescue of endogenous F508del-CFTR by miglustat or by low temperature in human airway and excised nasal epithelium of cftrF508del/F508del mice is paralleled by a down-regulation of Na+ absorption.
Materials and Methods
Cell Culture. The human nasal epithelial JME/CF15 cell line, derived from a F508del-CFTR homozygous patient (Jefferson et al., 1990), was grown at 37°C in 5% CO2 under standard culture conditions, in Dulbecco's modified Eagle's medium-Ham's F-12 (3:1) nutritive mix supplemented by 10% fetal bovine serum, 100 IU/ml penicillin and 100 μg/ml streptomycin, 5 μg/ml insulin, 5 μg/ml transferrin, 5.5 μM epinephrine, 180 μM adenine, 2 nM 3,3′,5-triiodo-l-thyronine sodium salt, and 1.1 μM hydrocortisone (Cao et al., 2005; Norez et al., 2006). All culture media and antibiotics were from Invitrogen (Carlsbad, CA). Fetal bovine serum was from PerbioScience (Brebières, France). Hormones and growth factors were from Sigma-Aldrich (St. Louis, MO). Cells were seeded in 35-mm plastic dishes for whole-cell patch-clamp recordings and on 12-mm snap wells (diameter, 1.13 cm2; pores, 3 μm; Corning Life Sciences, Acton, MA) for Ussing chamber experiments. Medium was renewed at 2-day interval.
Patch-Clamp Experiments. Perforated whole-cell patch-clamp experiments were performed on CF15 cells at room temperature. Currents were recorded with an RK-400 patch-clamp amplifier (Biologic, Grenoble, France). I-V relationships were built by clamping the membrane potential to -20 mV and by pulses from -140 to +100 mV (20-mV increments). Pipettes with resistance of 3 to 4 MΩ were pulled from borosilicate glass capillary tubing (GC150-TF10; Clarke Electromedical Instruments, Pangbourne, UK) using a two-step vertical puller (Narishige, Tokyo, Japan). They were filled with the following solution: 20 mM NaCl, 100 mM l-aspartic acid, 100 mM CsOH, 1 mM MgCl2, 20 mM CsCl, 4 mM EGTA, and 10 mM HEPES, pH 7.2. Amphotericin B (100 μg/ml) was dissolved ex temporane. Pipettes were connected to the head of the patch-clamp amplifier through an Ag-AgCl pellet. Seal resistances ranging from 6 to 20 GΩ were obtained. Results were analyzed with the pClamp 6.0.2 package software (pClamp; Molecular Devices, Sunnyvale, CA). The external bath solution contained 150 mM NaCl, 6 mM CsCl, 1 mM CaCl2, 1 mM MgCl2, 10 mM glucose, and 10 mM HEPES, pH 7.4. The liquid potential was corrected before seal establishment. Pipette capacitances were electronically compensated in cell-attached mode. To standardize experiments, recordings were performed only when the input resistance had a value ≤15 MΩ. The mean value of access resistances was 11.7 ± 1.6 MΩ (n = 42). Membrane capacitances were measured in the whole-cell mode by fitting capacitance currents, obtained in response to a hyperpolarization of 6 mV, with a first order exponential and by integrating the surface of the capacitance current. Mean values of membrane capacitance were 30.7 ± 8.2 pF (n = 42). For graphic representations, I-V relationship was normalized to 1 pF to remove variability due to differences in cell sizes.
Animals. Rotterdam homozygous F508del-CFTR mice (cftrtm1Eur) and their litter mate controls (FVB inbred, 14–17 weeks old, weight between 20 and 30 g) were kept on solid food in a pathogen-free environment. cftr-/- mice (cftrtm2Cam) were backcrossed for 12 generations into the FVB background (Scholte et al., 2004). Animals were anesthetized by an i.p. injection of a cocktail containing ketamine (10 mg/ml), xylazine (1.5 mg/ml), and diazepam (0.6 mg/ml). Nasal epithelia were dissected away from mice and mounted in a mini-Ussing chamber (exposed tissue area, 1.13 mm2) (De Jonge et al., 2004).
Short-Circuit Current Measurements. For short-circuit current (Isc) measurements, we seeded CF15 cells on semipermeable membrane snap well inserts (exposed surface area, 1.13 cm2). When cells were forming an impermeable monolayer (transepithelial resistance ≥ 500 Ω/cm2), short-circuit current recordings were performed. Snap wells were mounted in a vertical Ussing chamber (Harvard Apparatus Inc., Holliston, MA). Cells were bathed in Meyler buffer (apical and basolateral side) containing the following: 120 mM NaCl, 1.2 mM CaCl2, 1.2 mM MgCl2, 0.8 mM K2HPO4, 3.3 mM KH2PO4, 25 mM NaHCO3, and 10 mM d-glucose, pH 7.4 (gassed with 95% O2-5% CO2 at 37°C). Current magnitude was referred to the apical side of the monolayer. Miglustat was added directly to the culture medium (100 μM, 2 h, 37°C). Ex vivo short-circuit currents across isolated nasal epithelium were recorded first under control conditions using continuous oxygenated (95% CO2-5% O2) and temperature-controlled (37°C) Meyler solution, and measurements were repeated on the same tissue after 2-h incubation in Meyler solution containing 100 μM miglustat. For other technical details, see elsewhere (Noel et al., 2006).
Oral Administration of Miglustat to Mice and Isc Measurements. To evaluate the effect of in vivo miglustat treatment on amiloride-sensitive current in nasal epithelium, we administrated 1200 mg/kg/day miglustat by gavage to cftrF508del/F508del or cftr-/- (FVB) mice. Control groups received vehicle, i.e., PBS solution only. After 6 days, we dissected the nasal epithelium from mice of the two groups and recorded the amiloride-sensitive Isc ex vivo.
Pharmacological Agents. The specific CFTR inhibitor CFTRinh-172 (Ma et al., 2002) was from Calbiochem (San Diego, CA). Forskolin was from LC Laboratories (Woburn, MA). Miglustat was purchased from Toronto Research Chemicals (Toronto, ON, Canada). All other chemicals were obtained from Sigma-Aldrich. All chemicals were dissolved in DMSO (final concentration in DMSO < 0.1%) except miglustat, which was dissolved in water for all in vitro and ex vivo experiments and in PBS for oral administration. The currents were not altered by DMSO alone.
Data Analysis. All the data are presented as mean value ± S.E.M., where n refers to the number of experiments and N to the number of animals. The unpaired Student's t test was used to compare sets of data. All graphs are plotted with GraphPad Prism 4.0 for Windows (GraphPad Software Inc., San Diego, CA). Values of P < 0.05 were considered as statistically significant: *, P < 0.05; **, P < 0.01; and ***, P < 0.001. N.S. difference was P > 0.05.
Results
Miglustat Reduces Amiloride-Sensitive Na+Current. Perforated whole-cell patch-clamp experiments were performed to measure the impact of miglustat treatment on ENaC and CFTR currents in human airway epithelial CF cells (cells maintained 2 h at 37°C in a culture medium containing 100 μM miglustat; Fig. 1). JME/CF15 cells derived from the nasal airway epithelium of a CF patient (homozygous F508del-CFTR; Jefferson et al., 1990) were not responsive to forskolin (fsk)/genistein (gst) stimulation but displayed a significant amiloride-sensitive Na+ current (Tong et al., 2004; Cao et al., 2005). In the first series of experiments, we identified and characterized ENaC currents in several bath conditions (as indicated on the top of each panel) for control CF15 cells cultured at 37°C (Fig. 1, left traces), for temperature (27°C)-corrected cells (Fig. 1, middle traces) and for miglustat-corrected cells (Fig. 1, right traces). For each cell for which a perforated whole-cell experiment was possible, we first recorded basal currents (Fig. 1A) and then added 100 μM amiloride in the bath solution (Fig. 1B). By subtracting the residual current in the presence of amiloride from the basal current, we obtained the mean value for amiloride-sensitive current normalized to the cell capacitance and calculated the mean current densities (pA/pF; Fig. 2). At +100 mV, we obtained mean values of 7.8 ± 2.3 pA/pF for control cells (n = 12), 0.52 ± 0.35 pA/pF for low temperature-corrected cells (n = 7, P < 0.05), and 1.7 ± 0.8 pA/pF for miglustat-corrected cells (n = 10, P < 0.05). Then, on the same cell, we added the cocktail containing 10 μM forskolin plus 30 μM genistein to activate CFTR currents (Fig. 1C). Under this condition, we only observed activation of linear currents with miglustat-corrected (n = 10) and low temperature-corrected (n = 7; Figs. 1C and 2) cells. In the presence of fsk/gst in the bath, the corresponding experimental reversal potential Erev was -33.6 ± 3.4 mV for miglustat and -32.6 ± 2.9 mV for low temperature-corrected cells, both values being near the theoretical Nernst potential for the Cl- ion (ECl- =-33 mV) showing stimulation of chloride-selective currents. We determined the current density, after adding amiloride, for the Cl- currents recorded with miglustat-treated cells (current density at +100 mV, 11.6 ± 3.2 pA/pF, n = 10) and with temperature-corrected CF15 cells (current density at +100 mV, 16.5 ± 4.9 pA/pF, n = 7). No activation of Cl- current was recorded in untreated CF15 cells (current density of 1.75 ± 0.27 pA/pF, n = 12). To prove that the Cl- current activated after amiloride in the presence of fsk/gst was due to F508del-CFTR, we perfused 10 μM of the specific CFTR inhibitor CFTRinh-172 (Ma et al., 2002). This maneuver fully inhibited Cl- currents in miglustat- and temperature-corrected cells (Fig. 1D).
Activation of CFTR Does Not Influence ENaC Currents in CF15 Cells. In a second series of experiments, we wished to evaluate the effect of a preactivation of CFTR by fsk/gst on the amiloride-sensitive current on CF15 cells in the same experimental conditions as in Fig. 1. To that end, we reversed the protocol for channel activation, i.e., we activated CFTR first and then measured amiloride-sensitive ENaC currents. Figure 3A shows spontaneous control currents in resting CF15 cells. Adding fsk/gst in the experimental chamber activated a linear Cl- current only in miglustat-corrected cells (Fig. 3B, right traces, current density at +100 mV of 13.1 ± 3.1 pA/pF, n = 4, data not shown). As expected, no Cl- current was activated in untreated cells (Fig. 3B, left traces). The Cl- current density at +100 mV was 6.8 ± 0.9 pA/pF in the control condition and 5.8 ± 0.8 pA/pF in the presence of fsk/gst in the bath (N.S.; data not shown). After adding amiloride to the bath to block the activity of ENaC, the residual current in miglustat-corrected cells (Fig. 3C, right traces) was inhibited by CFTRinh-172 (Fig. 3D, right traces) indicating that F508del-CFTR channels were active. We measured the corresponding amiloride-sensitive ENaC current and calculated the amiloride-sensitive current density at +100 mV (Fig. 4). For miglustat-corrected cells, we found no significant difference between the amiloride-sensitive current with fsk/gst in the bath (1.76 ± 0.3 pA/pF at +100 mV, n = 4, N.S.) or without (1.7 ± 0.8 pA/pF at +100 mV, n = 10). Likewise, no significant effect of fsk/gst was noted on the magnitude of the amiloride-sensitive current in untreated CF15 cells (7.8 ± 2.3 pA/pF in the control condition, n = 12; 6.7 ± 1.7 pA/pF in the presence of fsk/gst, n = 9, N.S.) (Fig. 4). In addition, we performed several additional experiments to determine whether miglustat could by itself activate membrane conductances in CF cells. When perfusing miglustat in the experimental chamber bathing CF15 cells (cultured at 37°C), we were not able to record any conductances (data not shown). Moreover, using iodide efflux methods, we did not detect any stimulation of efflux in the presence of this agent. Finally, with CF15 cells treated for 2 h by this corrector, no modulation of either volume- or calcium-dependent-iodide efflux was noted (data not shown; see also Norez et al., 2006). Taken together, these experiments show that the iminosugar miglustat is not a channel activator but rather a F508del-CFTR corrector.
Miglustat Reduces Amiloride-Sensitive Isc in Polarized CF15 Cells. We performed Ussing chamber experiments on CF15 cells that we seeded on semipermeable snap well membrane in control conditions and after 2-h incubation in culture medium containing miglustat (Fig. 5). The mean value of transepithelial resistance was 531 ± 26 Ω/cm2 (n = 4) for untreated cells monolayers and increased to 658 ± 39 Ω/cm2 (n = 4, P < 0.05) for miglustat-corrected cells monolayers. Adding amiloride to the apical compartment induced a change of Isc, i.e., inhibition of apical Na+ absorption (Fig. 5A). We found ΔIsc of -2.05 ± 1.1 μA/cm2 for miglustat-corrected CF15 cells monolayers (n = 4) and -7.43 ± 0.7 μA/cm2 (n = 4) for control monolayers (P < 0.01; Fig. 5B). Then, addition of 10 μM fsk (basolateral side) and 30 μM gst (both sides) stimulated a glibenclamide-sensitive Isc, i.e., activation of apical F508del-CFTR-dependent Cl- secretion, for miglustat-corrected monolayers (Fig. 5B) but not for control monolayers (Fig. 5A). The ΔIsc was 8.3 ± 0.6 μA/cm2 for miglustat-corrected monolayers (n = 4) and 0.18 ± 0.09 μA/cm2 for control monolayers (n = 4, P < 0.001) (Fig. 5C).
Miglustat Reduces Amiloride-Sensitive Isc in Ex Vivo Experiments on Nasal Epithelium of cftrF508del/F508del Mice. To further explore the potential effect of miglustat on ENaC, we conducted electrophysiological experiments with nasal epithelium dissected from cftrF508del/F508del mice (Fig. 6). For each experiment, we recorded amiloride-sensitive ΔIsc before and after a 2-h incubation of the tissue with miglustat. In preliminary experiments, we determined that 2-h incubation in Meyler buffer without drug did not modify the amiloride response (data not shown). Compared with the untreated condition, amiloride-sensitive Isc was reduced after a 2-h incubation in miglustat-supplemented Meyler buffer. We found ΔIsc of -8.5 ± 1.9 μA/cm2 in the control condition and ΔIsc of -2.8 ± 0.5 μA/cm2 in miglustat (n = 8, P < 0.05; Fig. 6). With cftr+/+ mice, we found no significant difference between amiloride-sensitive Isc in the control condition (-0.6 ± 1.27 μA/cm2) and after miglustat (-0.77 ± 1.37 μA/cm2, n = 7, N.S.; Fig. 6).
Effect of Oral Administration of Miglustat on Ex Vivo Bioelectrics of cftrF508del/F508del and cftr-/-Mice. We administered 1200 mg/kg/day miglustat to cftrF508del/F508del and cftr-/- mice by gavage for 6 days. This concentration has been applied previously to demonstrate therapeutic benefits of miglustat in a mouse model of Sandhoff disease (Andersson et al., 2004). The control group received PBS only. On day 6 (i.e., after 12 applications), we dissected nasal epithelium from the different mice groups and recorded the amiloride-sensitive ΔIsc. We found significantly reduced amiloride-sensitive ΔIsc for nasal epithelium of cftrF508del/F508del mice (ΔIsc = -12.3 ± 6.2 μA/cm2, n = 10 mice) who received miglustat, compared with the PBS group (ΔIsc = -28.2 ± 3.5 μA/cm2, n = 12 mice) (P < 0.05; Fig. 7). To learn whether the inhibition of ENaC-mediated Na+ absorption in nasal epithelium is a consequence of the F508del-cftr rescue by miglustat in this tissue or is due to a cftr-independent effect, we have repeated the in vivo study with cftr-/- mice. However, no effect was noted between the two groups of cftr-/- mice (PBS, ΔIsc = -27.3 ± 17.0 μA/cm2, n = 5 mice; miglustat, ΔIsc = -24.0 ± 4.3 μA/cm2, n = 5 mice; N.S.; Fig. 7). Because there is no effect on KO mice, the effect of miglustat is due to F508del-cftr rescue; therefore, this corrector has no direct effect on ENaC activity.
Discussion
Nowadays, the majority of clinical treatments of CF targets the secondary manifestations of the pulmonary disease (inhaled antibiotics and recombinant human DNase). However, CFTR-directed treatments will probably arise in the near future due to our expanded knowledge of transepithelial ion transport pharmacology and molecular biology. One approach for correcting the basic defect in CF (also called protein therapy) aims at creating conditions to restore cAMP-dependent chloride transport and, hence, rehydration, of the airway surface in priority. However, it remains uncertain whether such a therapy is able to restore all pleiotropic functions of CFTR (Vankeerberghen et al., 2002). In the present study, we addressed one particular aspect of this question by analyzing the consequence of CFTR correction for the activity of ENaC of treating airway epithelial cells with miglustat, an agent that we showed able to restore functional and mature F508del-CFTR in epithelial cells (Norez et al., 2006; Antigny et al., 2008) and that is now evaluated in a phase 2a clinical trial for homozygous F508del patients (http://clinicaltrials.gov/). The major findings of the present study are summarized hereafter. In CF15 human airway epithelial cells incubated for 2 h at 37°C in the presence of 100 μM miglustat, a cAMP-dependent and CFTRinh-172-sensitive F508del-CFTR current was restored in parallel to the reduction in amplitude of the amiloride-sensitive ENaC Na+ current. In miglustat-treated cells, the magnitude of the amiloride-sensitive current was similar in the presence or absence of forskolin/genistein (to open CFTR channels), arguing that the activation of F508del-CFTR by forskolin/genistein is not required for down-regulation of ENaC. Finally, a CFTR-dependent normalization of amiloride-sensitive Na+ absorption in response to miglustat was observed both in human airway epithelial cells and in nasal epithelia of cftrF508del/F508del mice.
CFTR is a pleiotropic ion channel, i.e., apart from its ability to transport chloride ions as an ionic channel, it also regulates many other transport proteins and cellular functions due to large and dynamic macromolecular complexes that contain CFTR, signaling molecules, and transport proteins (Vankeerberghen et al., 2002; Guggino and Stanton, 2006). Abnormal Na+ transport in CF-affected airway epithelia has been suggested by many in vivo and in vitro observations in humans and mice, showing increased amiloride-sensitive transepithelial potentials in CF (Knowles et al., 1981, 1983; Boucher et al., 1986; Grubb and Boucher, 1997; Mall et al., 1998). Although not completely solved, it becomes apparent that interaction between CFTR and ENaC may involve PDZ-domain proteins and kinases (Guggino and Stanton, 2006). It is particularly important that the functional and reciprocal interaction between CFTR and ENaC regulates both epithelial Cl- and Na+ conductances (Stutts et al., 1995, 1997). However, it remains unclear why CF airway epithelia display such a high ENaC activity. Despite numerous studies, the molecular mechanism (direct or indirect) is still unknown. A potential direction of investigation to clear these points will have to address protein-protein interactions between CFTR and ENaC, which have been studied only in few reports. Berdiev et al. (2007) recently demonstrated a direct physical interaction between CFTR and the three ENaC subunits by fluorescence resonance energy transfer analysis. In the same study, these experiments were confirmed by coimmunoprecipitation. Nevertheless, these results argue that CFTR and α- and β-rENaC interact in a complex. However, to our knowledge, no results concerning physical interactions between ENaC subunits and CFTR mutant proteins, and in particular the F508del mutant, have yet been shown.
We also found that the CFTR activators forskolin/genistein on whole-cell current, examined under control condition and in the presence of amiloride, had no influence on amiloride-sensitive ENaC currents in CF15 cells. This outcome is at variance with previous studies showing that the CFTR regulation of ENaC in human intestinal (Mall et al., 1998) and colonic epithelial cells (Ecke et al., 1996) required active CFTR channels. The mechanism of reciprocal regulation of CFTR and ENaC is currently under investigation. It was initially observed in MDCK cells expressing both ENaC and CFTR and was subsequently demonstrated in X. laevis (Stutts et al., 1995; Mall et al., 1996). ENaC inhibition by CFTR was demonstrated when α, β, and γ ENaC subunits were coexpressed in oocytes of X. laevis with wild-type CFTR but not with F508del-CFTR (Stutts et al., 1995; Mall et al., 1996). In these studies, ENaC was inhibited during stimulation by agonist raising intracellular cAMP. A few studies showed that ENaC inhibition by CFTR also takes place in cells expressing both proteins endogenously (Ecke et al., 1996; Letz and Korbmacher, 1997) and was operational in normal human airways but not in CF patient tissues (Mall et al., 1998). It is now admitted that these findings may explain the typical enhanced amiloride-sensitive Na+ conductance and increased reabsorption of electrolytes observed in CF airways, two phenomena that are leading to highly viscous mucus and reduced mucociliary clearance (Zhang et al., 1996).
Note that it has been recently shown that overexpression of βENaC in mice led to a CF-like phenotype even in the presence of functional CFTR channels (Mall et al., 2004). Thus, inhibition of ENaC activity alone might already be of therapeutic value in CF. However, despite the fact that F508del-CFTR-mediated chloride secretion can be restored by a number of physical or pharmacological maneuvers in vitro and for some, in vivo (reviewed in Becq, 2006; MacDonald et al., 2007), a parallel change in amiloride-sensitive Na+ transport has not been frequently reported. The protein repair agent 4-phenylbutyrate (Buphenyl), which has been clinically evaluated in F508del-homozygous CF patients, partially restored CFTR function but had no effect on nasal amiloride-sensitive potential (Rubenstein and Zeitlin, 1998). In contrast, curcumin not only corrected CFTR functions but also affected the amiloride-sensitive response in cftrF508del/F508del mice (Egan et al., 2004). Thus, these results suggest that a single pharmacological agent should be, in principle, capable of sufficient correction of the defects in CF cells to produce clinical benefits. However, a preliminary phase 1 clinical trial with oral curcumin was rather disappointing and did not show correction of CFTR (http://www.cff.org). Oral administration of miglustat in cftrF508del/F508del mice resulted in reduction of amiloride-sensitive Isc. Therefore, our study demonstrates potential normalization of cAMP-dependent Cl- secretion and amiloride-sensitive Na+ absorption by a treatment with a single agent, the CFTR corrector miglustat. Moreover, the persistence of ENaC inhibition in cftrF508del/F508del animals receiving miglustat, together with a lack of effect on cftr-/- mice, offer direct proof that the rescue of F508del-cftr and ENaC down-regulation are linked.
How does miglustat affect both CFTR and ENaC transports in CF cells? Earlier, we provided evidence that miglustat prevents, at least in part, the interaction of the mutant channel with the ER-resident molecular chaperone calnexin (Norez et al., 2006). Preventing the calnexin interaction with the mutant protein in the ER has been regarded as one of the major mechanism of rescue (Egan et al., 2004; Norez et al., 2006). During an extensive study to better understand the mechanism of action of miglustat on epithelial CF cells, we also observed that this agent has no direct effect either on the Cl- channel activity of CFTR or on the activity of other non-CFTR Cl- channels. Finally, as far as the literature showed, in epithelial CF cells, the three ENaC subunits are all expressed and located at the apical plasma membrane. However, because F508del-CFTR is retained in the ER, the equilibrium between Cl- secretion and Na+ absorption is affected due to the absence of negative control of ENaC by CFTR as proposed by several authors (Stutts et al., 1995; Mall et al., 1996, 1998; Letz and Korbmacher, 1997; Kunzelmann et al., 2000; Berdiev et al., 2007). Moreover, the ENaC conductance is not inhibited by F508del-CFTR (Mall et al., 1996, 1998), and overexpression of the β-subunit of ENaC produces CF-like lung disease in a mouse model (Mall et al., 2004). Therefore, we propose that miglustat indirectly affects the transport of Na+ in CF cells through its effect as an α1,2-glucosidase inhibitor to perturb the F508del-CFTR/calnexin molecular interaction in the ER. Further studies will be needed to understand how rescuing F508del-CFTR from its intracellular retention re-establish the control of ENaC activity and thus normalize the transport of Na+ in CF cells.
In summary, in this report, we demonstrate that the rescue of endogenous F508del-CFTR by miglustat (or by low temperature) in human airway and nasal epithelial cells of cftrF508del/F508del mice is accompanied by a down-regulation of Na+ transport. Because the balance between CFTR-dependent Cl- secretion and ENaC-dependent Na+ reabsorption regulates the net amount of salt and water in airway periciliary fluid and thus the ability to clear bacteria and other noxious agents from the lungs, our findings predict that miglustat may not only ameliorate chloride transport but also sodium hyperabsorption.
Acknowledgments
We thank Nathalie Bizard for cell culture maintenance and James Habrioux for excellent assistance.
Footnotes
-
This study was supported by specific grants from Vaincre La Mucoviscidose and MucoVie66. S.N. was supported by a studentship from MucoVie66.
-
Part of this work were previously presented as an abstract as follows: Noel S, Wilke M, De Lange HR, and Becq F (2006) Rescue of F508del-CFTR processing defect by miglustat down-regulates sodium absorption in homozygous F508del-CFTR mice and human nasal cell (Abstract number 114). Pediatr PulmonolSuppl. 29:247.
-
Article, publication date, and citation information can be found at http://jpet.aspetjournals.org.
-
doi:10.1124/jpet.107.135582.
-
ABBREVIATIONS: CF, cystic fibrosis; CFTR, CF transmembrane conductance regulator; F508del, deletion of phenylalanine at 508 position of CFTR protein; ER, endoplasmic reticulum; ENaC, epithelial sodium channel; miglustat, n-butyldeoxynojyrimicin; Isc, short-circuit current; PBS, phosphate-buffered saline; CFTRinh-172, 3-[(3-trifluoromethyl)phenyl]-5-[(4-carboxyphenyl)methylene]-2-thioxo-4-thiazolidinone; DMSO, dimethylsulfoxide; fsk, forskolin; Gst, genistein.
- Received December 18, 2007.
- Accepted February 27, 2008.
- The American Society for Pharmacology and Experimental Therapeutics