Sulfate secretion and chloride absorption are mediated by the anion exchanger DRA (Slc26a3) in the mouse cecum

Abstract

Inorganic sulfate (SO₄²⁻) is essential for a multitude of physiological processes. The specific molecular pathway has been identified for uptake from the small intestine but is virtually unknown for the large bowel, although there is evidence for absorption involving Na⁺-independent anion exchange. A leading candidate is the apical chloride/bicarbonate (Cl⁻/HCO₃⁻) exchanger DRA (down-regulated in adenoma; Slc26a3), primarily linked to the Cl⁻ transporting defect in congenital chloride diarrhea. The present study set out to characterize transepithelial ³⁵SO₄²⁻ and ³⁶Cl⁻ fluxes across the isolated, short-circuited cecum from wild-type (WT) and knockout (KO) mice and subsequently to define the contribution of DRA. The cecum demonstrated simultaneous net SO₄²⁻ secretion (-8.39 ± 0.88 nmol·cm⁻²·h⁻¹) and Cl⁻ absorption (10.85 ± 1.41 μmol·cm⁻²·h⁻¹). In DRA-KO mice, SO₄²⁻ secretion was reversed to net absorption via a 60% reduction in serosal to mucosal SO₄²⁻ flux. Similarly, net Cl⁻ absorption was abolished and replaced by secretion, indicating that DRA represents a major pathway for transcellular SO₄²⁻ secretion and Cl⁻ absorption. Further experiments including the application of DIDS (500 μM), bumetanide (100 μM), and substitutions of extracellular Cl⁻ or HCO₃⁻/CO₂ helped to identify specific ion dependencies and driving forces and suggested that additional anion exchangers were operating at both apical and basolateral membranes supporting SO₄²⁻ transport. In conclusion, DRA contributes to SO₄²⁻ secretion via DIDS-sensitive HCO₃⁻/SO₄²⁻ exchange, in addition to being the principal DIDS-resistant Cl⁻/HCO₃⁻ exchanger. With DRA linked to the pathogenesis of other gastrointestinal diseases extending its functional characterization offers a more complete picture of its role in the intestine.

Keywords: PAT1; Slc26a6; epithelial ion transport; large intestine.

Fig. 1.

Unidirectional sulfate fluxes, J^SO₄ (A and B), and chloride fluxes, J^Cl (C and D), measured simultaneously across isolated, short-circuited segments of cecum taken from wild-type (WT) and DRA-knockout (DRA-KO) mice following application of mucosal DIDS (500 μM). E and F compare the responses of short-circuit current (I_sc) and transepithelial conductance (G_t) between groups. M, mucosal; S, serosal. Each data point represents mean ± SE from n = 12 and 6 WT and DRA-KO mice, respectively. *Statistically significant change from the preceding control period (0–30 min).

Fig. 2.

Unidirectional J^SO₄ (A) and J^Cl (B) measured simultaneously across isolated, short-circuited segments of cecum taken from PAT1-knockout (PAT1-KO) mice following application of mucosal DIDS (500 μM). C displays the responses of I_sc and G_t. Each data point represents mean ± SE from n = 6 PAT1-KO mice. *Statistically significant change from the preceding control period (0–30 min).

Fig. 3.

Unidirectional J^SO₄ (A and B) and J^Cl (C and D) measured simultaneously across isolated, short-circuited segments of cecum taken from WT and DRA-KO mice following application of serosal DIDS (500 μM). E and F compare the responses of I_sc and G_t between groups. Each data point represents mean ± SE from n = 5 and 6 WT and DRA-KO mice, respectively. *Statistically significant change from the preceding control period (0–30 min).

Fig. 4.

Unidirectional and net J^SO₄ measured across isolated, short-circuited segments of cecum taken from WT (A) and DRA-KO (B) mice following bilateral replacement of chloride in the buffer. Values represent means ± SE from n = 6 mice for each genotype. *Statistically significant difference from the corresponding flux measured in standard buffer. For comparison, the accompanying flux data in standard buffer (solid bars) has been reproduced from Table 1. In Cl⁻-free buffer (shaded bars), I_sc and G_t were −1.16 ± 0.11 μeq·cm⁻²·h⁻¹ and 8.38 ± 0.23 mS/cm² for WT mice (n = 12), and −2.56 ± 0.10 μeq·cm⁻²·h⁻¹ and 7.17 ± 0.37 mS/cm² for DRA-KO mice (n = 12), respectively.

Fig. 5.

Unidirectional and net J^SO₄ (A and B) and J^Cl (C and D) measured simultaneously across isolated, short-circuited segments of cecum taken from WT and DRA-KO mice following bilateral replacement of HCO₃⁻/CO₂ in the buffer. Values represent means ± SE from n = 6 mice for each genotype. *Statistically significant difference from the corresponding flux measured in standard buffer. For comparison, the accompanying flux data in standard buffer (solid bars) has been reproduced from Table 1. In HCO₃⁻/CO₂-free buffer (shaded bars), I_sc and G_t were −1.09 ± 0.08 μeq·cm⁻²·h⁻¹ and 14.90 ± 0.47 mS/cm² for WT mice (n = 12), and −3.02 ± 0.16 μeq·cm⁻²·h⁻¹ and 12.08 ± 0.53 mS/cm² for DRA-KO mice (n = 12), respectively.

Fig. 6.

Unidirectional J^SO₄ (A and B) and J^Cl (C and D) measured simultaneously across isolated, short-circuited segments of cecum taken from WT and DRA-KO mice following application of serosal bumetanide (100 μM). E and F compare the responses of I_sc and G_t between groups. Each data point represents mean ± SE from n = 6 WT and n = 6 DRA-KO mice. *Statistically significant change from the preceding control period (0–30 min).

Fig. 7.

A simple model illustrating the proposed mechanisms of transepithelial sulfate (SO₄²⁻) and chloride (Cl⁻) secretion (A) and absorption (B) by the mouse cecum. 1: For secretion, SO₄²⁻ from the serosal bath is accumulated in the cell via basolateral DIDS-sensitive anion exchange driven by either intracellular Cl⁻ or HCO₃⁻ (indicated by A⁻). At the apical membrane, SO₄²⁻ exits via DRA in exchange for mucosal HCO₃⁻. 2: The baseline secretory flux of Cl⁻ was supported by an anion exchanger (AE) at the basolateral membrane coupled to a Na⁺-HCO₃⁻ cotransporter (NBC), which provides HCO₃⁻ to drive Cl⁻ into the cell. A potential apical pathway for Cl⁻ exit was suggested to be via a conductive channel such as the CFTR. 3: SO₄²⁻ absorption was mediated by a DIDS-sensitive apical mechanism independent of either DRA or PAT1 and may reflect the activity of DTDST. Intracellular SO₄²⁻ exits via another DIDS-sensitive mechanism at the basolateral membrane tentatively identified as SAT1. 4: DRA is exclusively responsible for mucosal Cl⁻ uptake driven by the energy in the outward HCO₃⁻ gradient that is largely being supplied from the serosal bath by basolateral NBC. There are a number of pathways whereby Cl⁻ can then enter the serosal bath as described in the text, including DIDS-sensitive AE.