Reversible silencing of CFTR chloride channels by glutathionylation - PubMed (original) (raw)

Reversible silencing of CFTR chloride channels by glutathionylation

Wei Wang et al. J Gen Physiol. 2005 Feb.

Abstract

The cystic fibrosis transmembrane conductance regulator (CFTR) is a phosphorylation- and ATP-dependent chloride channel that modulates salt and water transport across lung and gut epithelia. The relationship between CFTR and oxidized forms of glutathione is of potential interest because reactive glutathione species are produced in inflamed epithelia where they may be modulators or substrates of CFTR. Here we show that CFTR channel activity in excised membrane patches is markedly inhibited by several oxidized forms of glutathione (i.e., GSSG, GSNO, and glutathione treated with diamide, a strong thiol oxidizer). Three lines of evidence indicate that the likely mechanism for this inhibitory effect is glutathionylation of a CFTR cysteine (i.e., formation of a mixed disulfide with glutathione): (a) channels could be protected from inhibition by pretreating the patch with NEM (a thiol alkylating agent) or by lowering the bath pH; (b) inhibited channels could be rescued by reducing agents (e.g., DTT) or by purified glutaredoxins (Grxs; thiol disulfide oxidoreductases) including a mutant Grx that specifically reduces mixed disulfides between glutathione and cysteines within proteins; and (c) reversible glutathionylation of CFTR polypeptides in microsomes could be detected biochemically under the same conditions. At the single channel level, the primary effect of reactive glutathione species was to markedly inhibit the opening rates of individual CFTR channels. CFTR channel inhibition was not obviously dependent on phosphorylation state but was markedly slowed when channels were first "locked open" by a poorly hydrolyzable ATP analogue (AMP-PNP). Consistent with the latter finding, we show that the major site of inhibition is cys-1344, a poorly conserved cysteine that lies proximal to the signature sequence in the second nucleotide binding domain (NBD2) of human CFTR. This region is predicted to participate in ATP-dependent channel opening and to be occluded in the nucleotide-bound state of the channel based on structural comparisons to related ATP binding cassette transporters. Our results demonstrate that human CFTR channels are reversibly inhibited by reactive glutathione species, and support an important role of the region proximal to the NBD2 signature sequence in ATP-dependent channel opening.

PubMed Disclaimer

Figures

F<sc>igure</sc> 1.

Figure 1.

Reversible inhibition of CFTR channel activity by GSSG. Macroscopic currents were recorded for membrane patches excised from BHK-CFTR cells as described in

materials and methods

. (A) GSSG was added at the indicated final bath concentrations. The bath chamber was washed free of PKA and GSSG and then PKA (110 U/ml each addition), and 20 mM DTT was added as indicated. Glibenclamide (300 μM), a voltage-dependent blocker of CFTR current, was added at the end of the experiment. (B) Corresponding I–V curves showing voltage-independent inhibition of CFTR current by GSSG. (C) Modest effect of GSH on CFTR current. Dotted lines indicate zero current levels.

F<sc>igure</sc> 2.

Figure 2.

CFTR inhibition by GSNO and diamide/GSH. (A) GSNO inhibition of CFTR currents. PKA inhibitory peptide (1.4 μg/ml PKI) was added at the first arrow. (B) Reversible inhibition of CFTR current by equimolar diamide/GSH. (C) CFTR channels are protected from diamide/GSH inhibition by pretreating the patch with NEM. NEM by itself stimulated the current as previously reported for CFTR (Cotton and Welsh, 1997). NEM, PKA, and PKI (but not MgATP) were washed out from the bath where indicated. (D) Mean data showing the inhibitory effects of GSNO (200 μM), GSSG (20 mM), and diamide/GSH (20 μM) on CFTR current and the rescue by subsequent addition of 20 mM DTT. Also shown is the protective effect of NEM pretreatment (0.1 mM) on the inhibition by diamide/GSH (20 μM). The results shown in Fig. 1 D were obtained following PKA inhibition with PKI (1.4 μg/ml). All data are normalized to the current recorded before the addition of the glutathione species. The number of experiments is shown above each bar.

F<sc>igure</sc> 3.

Figure 3.

pH dependence of the inhibition of CFTR current by diamide/GSH. Macroscopic currents were recorded for membrane patches excised from BHK-CFTR cells at pH 7.3 (A), pH 6.3 (B), and pH 8.3 (C). Diamide and GSH were premixed at pH 7.3 and then added to the bath where indicated. Results are representative of four experiments. CFTR currents per se were only modestly affected by changing the pH over this range; e.g., changing the bath pH from 7.3 to 8.3 resulted in a 10–20% inhibition of macroscopic current (unpublished data).

F<sc>igure</sc> 4.

Figure 4.

Glutaredoxin rescues CFTR channels from inhibition by reactive glutathione species. (A) Purified E. coli Grx1 (4 and 8 μM) reversed the inhibition of CFTR activity induced by diamide/GSH in an excised BHK-CFTR patch. (B) Purified E. coli Trx was unable to rescue CFTR activity following the addition of diamide/GSH. (C) CFTR activity also was rescued by a mutant Grx that is a specific mixed disulfide reductase (Grx3 C14S/C65Y). (D) Mean data showing the inhibitory effects of the indicated glutathione species (20 μM diamide/GSH; 200 μM GSNO; 20 mM GSSG) and the rescue effects of E. coli Grx1 (4 μM Grx1 plus 1 mM GSH) and GSH alone at high dose (20 mM). CFTR channels were preactivated with PKA and MgATP as described in

materials and methods

, and all additions were made after inhibiting the bath PKA with excess PKI inhibitory peptide (1.4 μg/ml). Results are normalized to the CFTR currents recorded before addition of the reactive glutathione species. The number of experiments is shown above each bar. For each glutathione species, the rescue by Grx or high GSH was statistically significant (P < 0.05) when compared with the current before Grx or GSH addition.

F<sc>igure</sc> 5.

Figure 5.

Reversible glutathionylation of CFTR in isolated membrane vesicles. Microsomal membranes were prepared from BHK-CFTR cells as described in

materials and methods

. Microsomes (200 mg total protein) were incubated in PBS in the absence (lane 1) or presence of 125 μM biotin GSH and/or 100 μM diamide for 10 min at 21–23°C (lanes 2–7). Where indicated, 4 μM E. coli Grx1 (plus 1 mM GSH) (lane 5), 20 mM GSH (lane 6), or 20 mM DTT (lane 7) was added for an additional 15 min at 21–23°C. Microsomes were then solubilized, and biotin-containing proteins were isolated by pull-down with streptavidin-agarose and probed for CFTR by immunoblotting as described in

materials and methods

. Total CFTR protein was assayed by immunoprecipitating CFTR from 3% of the total crude membrane lysate and blotting the immunoprecipitate with the same CFTR monoclonal antibody. This experiment was repeated four times with similar results.

F<sc>igure</sc> 6.

Figure 6.

Glutathionylated CFTR channels exhibit markedly reduced open rates. (A) Macroscopic current trace showing inhibition of CFTR current by diamide/GSH in membrane patch excised from Calu-3 epithelial cell. PKI (1.4 μg/ml) was added at the arrow. (B) Single channel record obtained from Calu-3 patch showing marked inhibition of channel opening by diamide/GSH (20 μM) and partial recovery by E. coli Grx1 (4 μM Grx1 plus 1 mM GSH). Holding potential was −80 mV. (C) Mean data showing inhibitory effects of diamide/GSH (20 μM) and rescue effects of Grx1 (+1 mM GSH) on single channel open probability (Po) and channel opening rate (n = 3 patches). The results shown in B and C were obtained without PKI addition. Nearly identical results were obtained for two additional patches that were treated with diamide/GSH after PKI addition (unpublished data).

F<sc>igure</sc> 7.

Figure 7.

CFTR channels that lack cys-1344 (C1344A-CFTR) are largely resistant to inhibition by reactive glutathione species. (A) Schematic view of CFTR depicting the 18 cysteines within the polypeptide and the specific location of cys-1344 near the ABC signature motif in NBD2. Bottom, sequences surrounding the NBD2 signature sequences in CFTR from human (h), mouse (m), and shark(s). (B) Mean data showing the inhibitory effects of diamide/GSH (20 μM) on the indicated alanine-substituted mutants and WT CFTR expressed in HEK-293T cells. Quad mutant is C128/225/343/866. Diamide/GSH was added after inhibiting the bath PKA with PKI. Data are normalized to currents measured just before the addition of diamide/GSH. (C) Representative current trace showing resistance of C1344A-CFTR to inhibition by diamide/GSH. (D) Mean data comparing the sensitivities of WT CFTR and C1344A-CFTR to inhibition by diamide/GSH (20 μM), GSNO (200 μM), and GSSG (20 mM).

F<sc>igure</sc> 8.

Figure 8.

Cys-1344 is sufficient for reactive glutathione species to inhibit CFTR currents. Cys-free CFTR and cys-1344-only CFTR were expressed in HEK-293T cells as described in the text. (A) Representative current trace showing lack of inhibition of cys-free CFTR by diamide/GSH. To confirm that the observed currents were CFTR-mediated, PKA and MgATP were washed out of the bath and 1.5 mM MgATP was readded where indicated. (B) Representative current trace showing inhibition of the current mediated by the cys-1344-only construct by diamide/GSH. The currents mediated by both constructs were blocked by glibenclamide in a voltage-dependent manner, although they were inhibited by PKI to a smaller degree than that observed for WT channels (possibly because of the altered conformations of these ER processing mutants). (C) Mean data showing percent current remaining after the addition of 20 μM diamide/GSH (cys-free, n = 7; cys-1344-only, n = 8).

F<sc>igure</sc> 9.

Figure 9.

An R domain deletion mutant is highly sensitive to diamide/GSH inhibition, whereas channels that are locked open by AMP-PNP are protected. (A) Inhibitory effect of 20 μM diamide/GSH on macroscopic current mediated by ΔR-S660A-CFTR in patch excised from transfected HEK-293T cell. (B) Mean data for ΔR-S660A-CFTR showing inhibition by diamide/GSH (20 μM) and recovery by E. coli Grx1 (4 μM) plus 1 mM GSH. Currents are normalized to control currents before diamide/GSH. Currents were activated with 1.5 mM MgATP in the absence of PKA for the experiments summarized in A and B. (C) Slow inhibition of CFTR currents by diamide/GSH following addition of 2 mM AMP-PNP to BHK-CFTR patch. Inset, exponential fit of current trace following addition of diamide/GSH to this patch. (D) Mean data comparing the exponential time constants for current inhibition by 20 μM diamide/GSH in the presence and absence of 2 mM AMP-PNP. Note that the time constants in the absence of AMP-PNP are overestimates due to the very rapid inhibition observed under the control condition.

F<sc>igure</sc> 10.

Figure 10.

Schematic views of the regions proximal to the signature sequences in related ABC transporters. (A) Schematic view of ATP binding pockets sandwiched between NBD dimers based on structures of Rad50 (Hopfner et al., 2000) and MalK (Chen et al., 2003). (B) ATP binding pocket in the MalK dimer of the bacterial maltose transporter (adapted from Chen et al., 2003). Asterisks denote the positions corresponding to cys-1344 in CFTR NBD2.

Similar articles

Cited by

References

    1. Aleksandrov, L., A. Mengos, X.-B. Chang, A. Aleksandrov, and J.R. Riordan. 2001. Differential interactions of nucleotides at the two nucleotide binding domains of the cystic fibrosis transmembrane conductance regulator. J. Biol. Chem. 276(16):12918–12923. - PubMed
    1. Aracena, P., G. Sanchez, P. Donoso, S.L. Hamilton, and C. Hidalgo. 2003. S-glutathionylation decreases Mg2+ inhibition and S-nitrosylation enhances Ca2+ activation of RyR1 channels. J. Biol. Chem. 278(44):42927–42935. - PubMed
    1. Barrett, W.C., J.P. DeGnore, S. König, H.M. Fales, Y.-F. Keng, Z.-Y. Zhang, M.B. Yim, and P.B. Chock. 1999. Regulation of PTP1B via glutathionylation of the active site cysteine 215. Biochemistry. 38:6699–6705. - PubMed
    1. Berger, A.L., M. Ikuma, J.F. Hunt, P.J. Thomas, and M.J. Welsh. 2002. Mutations that change the position of the putative γ-phosphate linker in the nucleotide binding domains of CFTR alter channel gating. J. Biol. Chem. 277(3):2125–2131. - PubMed
    1. Bushweller, J.H., F. Åslund, K. Wüthrich, and A. Holmgren. 1992. Structural and functional characterization of the mutant Escherichia coli glutaredoxin (C14S) and its mixed disulfide with glutathione. Biochemistry. 31:9288–9293. - PubMed

Publication types

MeSH terms

Substances

Grants and funding

LinkOut - more resources