The gastric H,K-ATPase blocker lansoprazole is an inhibitor of chloride channels

Abstract

1. It was postulated that swelling dependent chloride channels are involved in the proton secretion of parietal cells. Since omeprazole, lansoprazole and its acid activated sulphenamide form AG2000 are structurally related to phenol derivatives known to block swelling dependent chloride channels, we set out to test, whether these substances--which are known to block the H,K-ATPase--could also lead to an inhibition of swelling-dependent chloride channels. Swelling-dependent chloride channels--characterized in many different cell types--show highly conserved biophysical and pharmacological features, therefore we investigated the effect of omeprazole, lansoprazole and its acid activated sulphenamide form AG2000 on swelling-dependent chloride channels elicited in fibroblasts, after the reduction of the extracellular osmolarity. 2. Omeprazole, lansoprazole and its acid activated sulphenamide form AG2000 are able to block swelling-dependent chloride channels (IClswell). 3. Lansoprazole and its protonated metabolite AG2000 act on at least two different sites of the IClswell protein: on an extracellular site which seems to be in a functional proximity to the nucleotide binding site, and on an intracellular site which allows the formation of disulfide-bridges. 4. The inhibition of the proton pump and the simultaneous blocking of chloride channels by omeprazole, lansoprazole and its acid activated sulphenamide form AG2000, as described here could be an effective mode to restrict proton secretion in parietal cells.

Figure 1

Relative permeabilities of swelling-dependent chloride currents for SCN⁻, I⁻, Cl⁻ and gluconate (gluc⁻) anions. (a) The cells were held at 0 mV, and voltage ramps were made between −120 and +100 mV for 500 ms. (b) The summary of the reversal potentials (E_rev) in the presence of the different anions is given.

Figure 2

Lansoprazole and AG2000 are able to block IClswell in a dose-dependent manner. (a) At a concentration of 0.1 mM, lansoprazole led to a slight inhibition of IClswell in four out of eight experiments. In the presence of 0.5 mM, IClswell is blocked by 78.8±3.4% (n=7). Each concentration was tested individually, and therefore the original readings of two different cells are shown. The holding potential for both experiments was 0 mV. Repetitive voltage steps were made up to a potential of +40 mV for 400 ms. (b) The current-density/voltage relation is given for the current under isotonic (iso) conditions, after reduction of the extracellular osmolarity (hypo), under hypotonic conditions in the presence of 0.5 mM lansoprazole (lanso 0.5 mM) and 180 s after washing out the drug (wash). (c) The same experimental procedure as described in (a) is used for AG2000. At a concentration of 20 μM, AG2000 led to a moderate block of IClswell in five out of six experiments. At a concentration of 100 μM, AG2000 blocks IClswell by 70.4±9.5% (n=5). (d) According to (b), the current-density/voltage relation is given in the absence and presence of 100 μM AG2000. Note that the effect of AG2000 (100 μM) is not reversible, since 180 s after washing out the drug a further reduction of the current can be measured.

Figure 3

Lansoprazole and AG2000 as well as omeprazole block IClswell at different concentrations. (a) The molecular structures of omeprazole and lansoprazole are given. The substituted benzimidazole lansoprazole is rearranged after protonization to form a sulfenic acid that is further rearranged to a sulphenamide. The sulphenamide of lansoprazole is termed AG2000, and it interacts covalently with sulfhydryl groups of the membrane spanning H,K-ATPase. (b) The half-maximal concentration required to block IClswell (IC₅₀) is ≈22 μM for AG2000 and ≈160 μM for lansoprazole (the holding potential is 0 mV and repetitive voltage steps to +40 mV are made). Omeprazole is significantly less active in blocking IClswell compared with lansoprazole.

Figure 4

At the maximal concentration tested, lansoprazole and AG2000 block IClswell with different time constants. The block with lansoprazole is significantly faster (τ is 57.9±8.6 s; n=7) compared to AG2000 (τ is 104.6±5.2 s; n=5).

Figure 5

Intracellular (pipette) effect of lansoprazole (160 μM) and AG2000 (22 μM). The swelling-dependent chloride current (hypo (control)) can be blocked by intracellular AG2000, but not by lansoprazole (lanso).

Figure 6

Dose-response curve of TDP on swelling-dependent chloride currents. No block can be observed at different concentrations tested.

Figure 7

Effect of lansoprazole (lanso) and AG2000 in the presence of 0.1 mM TDP. (a) The block effected by lansoprazole (160 μM) is reduced in the presence of 0.1 mM TDP. (b) The block of AG2000 (22 μM) remains unaffected by the presence of 0.1 mM TDP.

Figure 8

The effect of DTT in the presence and in the absence of AG2000. (a) The block effected by AG2000 (control) can be significantly reduced in the presence of 1 mM intracellular (pipette) DTT, but can not be reduced by adding DTT to the extracellular (bath) side (the cells were preincubated at least 1 min with DTT before adding AG2000). (b) DTT at a concentration of 1 mM is not able to block swelling-dependent chloride channels when added to either side.