Native and recombinant Slc26a3 (downregulated in adenoma, Dra) do not exhibit properties of 2Cl-/1HCO3- exchange

Abstract

The recent proposal that Dra/Slc26a3 mediates electrogenic 2Cl(-)/1HCO(3)(-) exchange suggests a required revision of classical concepts of electroneutral Cl(-) transport across epithelia such as the intestine. We investigated 1) the effect of endogenous Dra Cl(-)/HCO(3)(-) activity on apical membrane potential (V(a)) of the cecal surface epithelium using wild-type (WT) and knockout (KO) mice; and 2) the electrical properties of Cl(-)/(OH(-))HCO(3)(-) exchange by mouse and human orthologs of Dra expressed in Xenopus oocytes. Ex vivo (36)Cl(-) fluxes and microfluorometry revealed that cecal Cl(-)/HCO(3)(-) exchange was abolished in the Dra KO without concordant changes in short-circuit current. In microelectrode studies, baseline V(a) of Dra KO surface epithelium was slightly hyperpolarized relative to WT but depolarized to the same extent as WT during luminal Cl(-) substitution. Subsequent studies indicated that Cl(-)-dependent V(a) depolarization requires the anion channel Cftr. Oocyte studies demonstrated that Dra-mediated exchange of intracellular Cl(-) for extracellular HCO(3)(-) is accompanied by slow hyperpolarization and a modest outward current, but that the steady-state current-voltage relationship is unaffected by Cl(-) removal or pharmacological blockade. Further, Dra-dependent (36)Cl(-) efflux was voltage-insensitive in oocytes coexpressing the cation channels ENaC or ROMK. We conclude that 1) endogenous Dra and recombinant human/mouse Dra orthologs do not exhibit electrogenic 2Cl(-)/1HCO(3)(-) exchange; and 2) acute induction of Dra Cl(-)/HCO(3)(-) exchange is associated with secondary membrane potential changes representing homeostatic responses. Thus, participation of Dra in coupled NaCl absorption and in uncoupled HCO(3)(-) secretion remains compatible with electroneutrality of these processes, and with the utility of electroneutral transport models for predicting epithelial responses in health and disease.

Fig. 1.

Knockout of Dra abolishes cecal Cl⁻/HCO₃⁻ exchange with minimal change in short-circuit current (I_sc). A: net ³⁶Cl⁻ flux across the cecal mucosa from wild-type (WT) and Dra knockout (KO) mice (n = 3 pairs). B: I_sc measured during ³⁶Cl⁻ flux measurements of cecal mucosa from WT and Dra KO mice. C: Cl⁻/HCO₃⁻ exchange rates measured as change in intracellular pH per minute (ΔpH_i/min) during removal of apical (luminal) Cl⁻ in surface epithelium of cecal mucosa from WT and Dra KO mice (n = 4). D: pH_i measured in balanced Krebs bicarbonate Ringer (KBR) solutions (n = 4). Values are means ± SE. *Significantly different from WT littermates.

Fig. 2.

Absence of Dra does not alter Cl⁻-dependent regulation of apical membrane potential (V_a) in cecal surface epithelium. A: conventional microelectrode recording of V_a in WT cecal surface epithelium during removal and restoration of luminal Cl⁻. Representative of 14 impalements from 6 WT mice. Ap, apical; BL, basolateral. B: summary of initial V_a (left) and maximal ΔV_a during apical (luminal) Cl⁻ substitution (right) of surface epithelium at 37°C in WT and Dra KO ceca (n = 11–14 impalements from 6 WT and Dra KO pairs). C: summary of initial V_a (left) and maximal ΔV_a during apical (luminal) Cl⁻ substitution (right) of surface epithelium at 37°C in WT and putative anion transporter-1 (Pat-1) KO ceca (n = 15–18 impalements from 6 WT and Pat-1 KO pairs). Values are means ± SE. *Significantly different from WT littermates.

Fig. 3.

Chloride-dependent depolarization of cecal V_a requires expression of Cftr. A: conventional microelectrode recording of V_a in Cftr KO cecal surface epithelium during removal and restoration of apical (luminal) Cl⁻. Representative of 12 impalements from 6 Cftr KO mice. B: summary of maximal ΔV_a during apical (luminal) Cl⁻ substitution in WT, Cftr KO, Cftr KO bathed with 80 mM K⁺, and Dra/Cftr double-knockout (dKO) ceca (n = 14 impalements from 6 WT, 13 impalements from 6 Cftr KO, 12 impalements from 6 Cftr KO bathed in 80 mM K⁺ KBR, and 9 impalements from 4 Dra/Cftr dKO ceca). Values are means ± SE. #Significantly different from WT. *Significantly different from Cftr KO.

Fig. 4.

Dra-mediated Cl⁻/HCO₃⁻ exchange initiated by extracellular Cl⁻ removal from Xenopus oocytes is unaccompanied by depolarization. V_m, membrane potential. A: pH-sensitive microelectrode recording of Cl⁻/HCO₃⁻ exchange in a Xenopus oocyte previously injected with water, recorded during bath Cl⁻ removal and restoration. Top: voltage trace vs. time. Bottom: pH_i trace vs. time. Representative of 4 water-injected oocytes. B: pH-sensitive microelectrode recording of Cl⁻/HCO₃⁻ exchange in oocyte expressing mSlc26a3/Dra, recorded during bath Cl⁻ removal and restoration. Top: voltage trace vs. time. Bottom: pH_i trace vs. time. Representative of 6 mSlc26a3-expressing oocytes.

Fig. 5.

Dra-mediated Cl⁻/OH⁻ and Cl⁻/HCO₃⁻ exchanges are unaccompanied by Cl⁻-dependent currents. Two-electrode voltage clamp current-voltage relationships of Xenopus oocytes previously injected with cRNA encoding hSLC26A3/DRA (A) or mSlc26a3/Dra (B), recorded in room air during sequential exposures to baths containing Cl⁻ (ND-93), gluconate substitute, and Cl⁻ in the presence of the DRA inhibitor, niflumic acid. Current-voltage relationships of oocytes previously injected with water (C) or with cRNA encoding mSlc26a3/Dra (D), recorded in 5% CO₂/24 mM HCO₃⁻ during sequential exposures to baths containing isethionate and Cl⁻. Values are means ± SE; n = 5 for each condition.

Fig. 6.

Dra-mediated Cl⁻/HCO₃⁻ exchange triggered by bath Cl⁻ removal under voltage clamp is not accompanied by inward current. Current measured in representative Xenopus oocytes expressing mSlc26a3/Dra while clamped at −30 mV (A) or at −90 mV (B) during sequential exposures to baths containing Cl⁻, isethionate substitute, and Cl⁻. C: summary of peak change in current elicited by bath substitution from Cl⁻ to isethionate; summarized data from 5 oocytes previously injected with water, and from 9 oocytes expressing mSlc26a3/Dra, clamped at −30 mV or at −90 mV. Values are means ± SE.

Fig. 7.

Dra-mediated ³⁶Cl⁻ efflux in room air is not sensitive to human epithelial Na⁺ channel (hENaC)-dependent changes in V_m. A: V_m recordings from representative Xenopus oocytes coexpressing Dra with (black trace) or without (level gray trace) mouse ENaC α/β/γ, during sequential exposure to baths containing Na⁺, N-methyl-d-glucamine (NMDG), and again Na⁺. One of 4 similar experiments. B: ³⁶Cl⁻ efflux rate constants of oocytes previously injected with water or coinjected with cRNA encoding Dra and mouse ENaC α/β/γ, and sequentially exposed to bath solutions containing first NMDG (Na⁺-free), then Na⁺ (ND-96), and finally Na⁺ plus Dra inhibitor niflumate (100 μM). Values are means ± SE (n as indicated).