INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

CONGENITAL CHLORIDE DIARRHEA (CLD) is a rare autosomal disease characterized by a watery stool with high chloride concentration and metabolic alkalosis (24). Previous perfusion studies have demonstrated that the basic defect in these patients is in the ileal and colonic luminal membrane Cl/HCO exchange process (3). Recently, a gene family of anion exchangers (AE) including AE1, AE2, bAE3, and cAE3 has been described (1). Chow et al. (9) have reported that in the rabbit ileum, AE2 protein is localized to the apical membrane of the ileal enterocytes and suggested AE2 to be involved in the luminal chloride absorption. In this regard, a gene, downregulated in adenoma (DRA), was recently described that is expressed in the normal differentiated epithelium in human colon, but its expression is significantly decreased or lost in colonic adenoma and adenocarcinama (37). By genetic and physical mapping, recent studies (15) have implicated that the DRA gene but not any of the AE isoforms is a positional candidate for CLD. Moreover, these studies (16) identified three mutations of the DRA gene in CLD patients and suggested that DRA may be involved in intestinal chloride transport. The cDNA sequence of DRA has been shown to exhibit a high homology with two sulfate transporters, the diastrophic dysplasia sulfate transporter (DTDST) and rat liver sulfate anion transporter (Sat-1), but not with any member of the AE family (37). Furthermore, early in vitro studies showed that DRA was capable of transporting sulfate and oxalate (37). However, recent studies (22, 23) have also shown that human and mouse DRA are able to transport chloride as well.

Previous studies from our laboratory, utilizing purified apical plasma membrane vesicles, have demonstrated the presence of Cl/HCO (OH) exchange process in the human colon (21). Additionally, we have previously shown that AE2 and bAE3 but not AE1 and cAE3 are expressed throughout the length of the human intestine (38). Our immunoblotting studies demonstrated that the protein products of AE2 and bAE3 are localized to basolateral membranes of the epithelial cells in all the regions of the human intestine (2). Recent studies (39) from our laboratory have also demonstrated and characterized a sulfate/hydroxyl exchange process in the proximal colonic apical membrane vesicles and showed that this exchanger has very low affinity for chloride compared with sulfate. The results of these studies strongly suggested that the described sulfate/hydroxyl exchange activity was distinct from the previously described Cl/HCO (OH) exchange process. In light of the aforementioned, it is not yet clear whether DRA protein product is in fact the apical anion exchanger responsible for the Cl/HCO exchange across the apical membranes of the human colonic epithelium or whether it is mainly a sulfate and oxalate transporter that could also transport chloride.

These intriguing findings warrant the development of a suitable in vitro model to study the sulfate and chloride uptake, their substrate specificity, and their regulation with various hormones and to examine the role of DRA in intestinal electrolyte transport. The colonic adenocarcinoma Caco-2 cell line has been shown to be a good model to study the mechanisms of electrolyte transport in the intestine (40). Additionally, thyroxine has previously been shown to alter the level of expression and the function of a number of electrolyte transporters in various tissues and cell lines (6, 8, 10). Therefore, our current studies were undertaken to examine whether Caco-2 cell line could serve as a suitable model to study the apical chloride and sulfate uptake mechanisms and their substrate specificity. Also, we examined the possible role of thyroxine in regulation of the expression of human DRA mRNA along with chloride and sulfate uptake activities in Caco-2 cells.

Our current studies demonstrated that Caco-2 cells could serve as an experimental model to study the apical intestinal SO/OH and Cl/OH exchange activities. Our data demonstrated that the two transporters are distinct based on the following: 1) in contrast to the Cl/OH exchange process, the SO/OH exchange activity was highly sensitive to inhibition by DIDS (the AE inhibitor); 2) DIDS competitively inhibited the Cl/OH exchange, whereas it inhibited the SO/OH exchange in a mixed-type manner; 3) chloride inhibited the SO/OH exchange competitively with a high inhibition constant (K_i) for Cl compared with the Michaelis-Menten consant (K_m) for SO, indicating lower affinity of chloride for this sulfate transporter; and 4) thyroxine significantly reduced the level of expression of human DRA mRNA along with SO/OH exchange activity, whereas the Cl/OH exchange activity was not altered by thyroxine. These data indicate that DRA may be primarily responsible for sulfate transport although capable of transporting chloride and that it appears to be distinct from the previously described Cl/HCO (OH) exchanger in the intestinal luminal membrane.

	MATERIAL AND METHODS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

Cell culture

³⁵SO and ³⁶Cl uptake

Designing of PCR primers and PCR technique

Isolation of RNA

Generation of cRNA probes and RNase protection assay

For the RNase protection assay, total RNA (20-30 µg) was coprecipitated with 10⁷ counts/min of high specific activity ³²P-labeled DRA riboprobe and 10⁵ counts/min of low specific activity ³²P-labeled GAPDH riboprobe. Samples were then resuspended in a hybridization buffer containing 75% formamide, 400 mM NaCl, 1 mM EDTA, and 40 mM PIPES, pH 6.4, and were hybridized at 45°C for 12-18 h. Samples were then diluted in 10 vol of 300 mM NaCl, 5 mM EDTA, and 10 mM Tris pH 7.5, and 1,400 units of T1 ribonuclease were added to each sample. After a 45-min incubation at 37°C, samples were added to a stop solution containing 4 M LiCl and 5 µg tRNA and precipitated with 2 vol of ethanol. Precipitates were resuspended in a small volume of dye solution (xylene cyanole and bromophenol blue in 90% formamide and 10 mM EDTA, pH 7.5). The double-stranded ³²P-cRNA fragments that were protected from the RNase digestion were heated at 95°C for 5 min and analyzed by electrophoresis on a denaturing polyacrylamide gel containing 8 M urea. The gels were dried and exposed to storage phosphor screen overnight and then analyzed utilizing a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). The relative abundance of DRA was calculated by comparing the number of counts of radioactivity of DRA band in response to thyroxine treatment or vehicle alone after normalization to the number of counts in the GAPDH band. Because GAPDH mRNA is expressed in very high abundance compared with DRA, ~20-fold lower specific activity of ³²P-labeled GAPDH compared with DRA probes were synthesized by including more cold CTP in the transcription reaction. This was specifically done to keep the density of GAPDH band in a readable range on final scan while enabling the detection of DRA band. The procedure of labeling was exactly the same every time, and once the probes were synthesized, the batch of probes was simultaneously utilized for hybridization with equal amounts of RNA.

Statistical analysis

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

The OH gradient-stimulated ³⁵SO uptake in Caco-2 cells

Time course of OH gradient-stimulated ³⁵SO uptake. Previous studies have shown the protein product of DRA gene to be a sulfate and oxalate transporter. Additionally, Caco-2 cells have previously been utilized to characterize electrolyte transport in the human intestine. Therefore, to establish an in vitro cellular model to study the role of DRA in anion transport in the human intestine and its regulation by the hormones, we first examined the OH gradient-dependent apical ³⁵SO uptake in monolayers of 5-7 days postconfluent well-differentiated Caco-2 cells. As shown in Fig. 1, OH gradient-driven ³⁵SO uptake in Caco-2 cells was linear as a function of time for up to 7 min and was significantly inhibited by 0.3 mM DIDS (the AE inhibitor). Therefore, a 2-min incubation time was used in all subsequent experiments.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Time course of OH gradient-stimulated 35SO uptake in Caco-2 cells. Caco-2 cells, 5-7 days postconfluent, were incubated for 1 h at room temperature in the HEPES/KOH medium adjusted to pH 8.5. Cells were then washed and incubated with the uptake buffer, pH 7, containing 35SO (50 µM) as described in MATERIALS AND METHODS, in the absence () or the presence () of 0.3 mM DIDS. Uptake was then terminated by 2 washes with ice-cold PBS buffer at indicated time points (x-axis). Results are means ± SE of 6 uptake determinations performed on 2 separate occasions.

Kinetics of SO/OH exchange process in Caco-2 cells. To further characterize the apical SO/OH exchange process in Caco-2 cells, we examined the kinetic parameters of the exchanger by measuring sulfate uptake in the presence of increasing concentrations of extracellular SO (50 µM to 1.5 mM). Figure 2A shows that the apical OH gradient-stimulated ³⁵SO uptake in Caco-2 cells exhibited saturation kinetics in the presence of increasing concentrations of sulfate. Lineweaver-Burk plot (Fig. 2B) demonstrated a straight line with an apparent K_m of 0.2 ± 0.08 mM for sulfate and a V_max of 1.1 ± 0.2 nmol · mg protein¹ · 2 min¹. These data indicate a carrier-mediated process for the OH gradient-driven apical sulfate uptake in Caco-2 cells.

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Kinetics of SO/OH exchange process in Caco-2 cells. Postconfluent Caco-2 cells were incubated for 1 h at room temperature in the HEPES/KOH medium adjusted to pH 8.5. Cells were then washed and incubated with the 35SO uptake buffer, described in MATERIALS AND METHODS, in the presence of increasing concentrations of extracellular SO (50-1,500 µM). A: Michaelis-Menton plot of SO uptake as a function of the SO concentration; B: Lineweaver-Burk plot for data. Results are means ± SE of 10 uptake determinations performed on 4 separate occasions.

³⁵SO uptake inhibition by DIDS. As shown above, the SO/OH exchange process was highly sensitive to inhibition by DIDS. To further examine the characteristics of the exchanger, we studied the effect of DIDS on kinetic parameters of the apical ³⁵SO uptake in Caco-2 cells. Lineweaver-Burk plot, as shown in Fig. 3, demonstrated that ³⁵SO uptake was inhibited by 50 µM DIDS in a mixed-type manner, with an increase in the apparent K_m for sulfate from 0.22 to 0.73 mM and a decrease in the V_max from 1.57 to 0.23 nmol · mg protein¹ · 2 min¹. These experiments were repeated using different concentrations of DIDS (0.015, 0.025, and 0.05 mM; not shown) and demonstrated the same pattern of inhibition with a K_i of 0.9 ± 0.3 µM.

View larger version (14K): [in this window] [in a new window]   Fig. 3.   Effect of DIDS on the kinetics of SO/OH exchange activity in Caco-2 cells. Postconfluent Caco-2 cells were preincubated for 1 h at room temperature in HEPES/KOH medium adjusted to pH 8.5. OH gradient-stimulated SO uptake in Caco-2 cells, in the presence () or the absence () of 0.05 mM DIDS, was determined as described in Fig. 2 legend. Lineweaver-Burk plot for data is shown. Results are means ± SE of 4 uptake determinations performed on 2 separate occasions.

Effect of anions on ³⁵SO uptake. Recent studies (22, 23) have suggested that the DRA protein product is a Cl/HCO (OH) exchanger. Previously, DRA was shown to be a sulfate and oxalate transporter. To examine the anion specificity of the apical SO/OH exchanger in Caco-2 cells, we investigated the effect of various anions on the sulfate uptake in Caco-2 cells. As shown in Fig. 4, 5 mM cis-concentration of butyrate, formate, lactate, succinate, and nitrate failed to significantly inhibit ³⁵SO (50 µM) uptake in Caco-2 cells. On the other hand, 5 mM cis-concentration of oxalate and chloride inhibited ³⁵SO (50 µM) uptake by ~90 and ~30%, respectively. These findings suggest that oxalate and chloride but not butyrate, formate, lactate, succinate, or nitrate could serve as alternative substrates for the SO/OH antiporter in Caco-2 cells. These results also indicate that the exchanger has a markedly higher affinity for oxalate compared with chloride.

View larger version (16K): [in this window] [in a new window]   Fig. 4.   Effects of anions on OH gradient-driven 35SO uptake in Caco-2 cells. Caco-2 cells preincubated with the HEPES medium, pH 8.5, were incubated with 35SO uptake buffer, containing 50 µM of 35SO, in the presence of 5 mM concentration of different anions. Results are a percentage of uptake in the presence of each anion compared with control (100%) and are means ± SE of 6 uptake determinations (*P < 0.05) from 3 separate occasions.

The mechanism(s) of ³⁵SO uptake inhibition by oxalate and chloride. Because oxalate and chloride significantly inhibited the apical OH gradient-stimulated ³⁵SO uptake in Caco-2 cells, we further investigated the mechanisms by which oxalate and chloride inhibited the ³⁵SO uptake. Figure 5 shows the effect of 0.5 mM concentration of oxalate on the kinetics of ³⁵SO uptake, which demonstrates a competitive inhibition. The experiment was repeated using different concentrations of oxalate (0.3, 0.5, and 1 mM; not shown) and showed the same type of inhibition with a K_i of 0.2 ± 0.03 mM for oxalate. On the other hand, as shown in Fig. 6, ³⁵SO uptake was also competitively inhibited by 10 mM concentration of chloride. Repeating the experiment with different concentrations of chloride (5, 10, and 25 mM; not shown) revealed the same pattern of inhibition with a K_i of 5.9 ± 1 mM for chloride. These findings suggest that although the SO/OH exchanger could use oxalate and chloride as alternative substrates in addition to sulfate, it appears that the antiporter has a markedly higher affinity for both sulfate and oxalate compared with chloride.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Effect of oxalate on the kinetics of SO/OH exchange process in Caco-2 cells. OH gradient-stimulated SO uptake in Caco-2 cells in the presence () or the absence () of 0.5 mM concentration of oxalate was determined as described in Fig. 2 legend. Lineweaver-Burk plot for data is shown. Results are means ± SE of 8 uptake determinations performed on 3 separate occasions.

View larger version (15K): [in this window] [in a new window]   Fig. 6.   Effect of Cl on 35SO in Caco-2 cells. pH gradient-driven SO uptake in postconfluent Caco-2 cells was measured in the presence () or the absence () of 10 mM concentration of Cl as described in Fig. 2 legend. Lineweaver-Burk plot for data is shown. Results are means ± SE of 4 uptake determinations performed on 2 separate occasions.

OH gradient-stimulated ³⁶Cl uptake in Caco-2 cells

Time course of OH gradient-stimulated ³⁶Cl uptake. We have previously characterized Cl/HCO (OH) exchange process in the human colonic proximal and distal apical plasma membrane vesicles. To investigate whether Caco-2 cells could also serve as a model to study the chloride uptake, we examined the effect of outwardly directed OH gradient on the time course of ³⁶Cl uptake in Caco-2 cells. As shown in Fig. 7, OH gradient-dependent ³⁶Cl uptake in Caco-2 cells was linear as a function of time up to 7 min and it was inhibited by DIDS. It should be noted that in contrast to SO/OH exchange activity, OH gradient-stimulated ³⁶Cl uptake was much less sensitive to inhibition by DIDS. These data suggest that in Caco-2 cells the two processes may be occurring via two distinct transporters.

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Time course of OH gradient-stimulated 36Cl uptake in Caco-2 cells. Postconfluent Caco-2 cells were incubated for 1 h at room temperature in HEPES/KOH medium adjusted to pH 8.5. Cells were then washed and incubated with uptake buffer, pH 7, containing 36Cl as described in MATERIALS AND METHODS, in the absence () or the presence () of 0.3 mM DIDS. Uptake was then terminated by 2 washes with ice-cold PBS buffer at indicated time. Results are means ± SE of 6 uptake determinations performed on 3 separate occasions.

The OH gradient-stimulated ³⁶Cl uptake inhibition by DIDS. Because the SO/OH exchange process was shown to be inhibited by DIDS in a mixed-type manner and Cl/OH exchange appeared to be more resistant to inhibition by DIDS, we examined the mechanism of Cl/OH exchange inhibition by DIDS in Caco-2 cells. As shown in Fig. 8, 0.35 mM DIDS altered the apparent K_m for chloride, whereas no significant changes occurred in the V_max. The data also indicated a competitive mode of inhibition. These experiments were repeated using different concentrations of DIDS (0.2, 0.5, and 1 mM; not shown) and demonstrated the same pattern of inhibition with a K_i of 0.3 ± 0.1 mM.

View larger version (14K): [in this window] [in a new window]   Fig. 8.   Effect of DIDS on kinetics of Cl/OH exchange activity in Caco-2 cells. OH gradient-stimulated 36Cl uptake was determined as described in MATERIALS AND METHODS in presence of increasing concentrations of Cl (2.7-50 mM). Uptake was measured in the presence () or the absence () of 0.35 mM DIDS. Lineweaver-Burk plot for data is shown. Results are means ± SE of 6 uptake determinations performed on 2 separate occasions.

The effect of thyroxine on OH gradient-stimulated ³⁵SO and ³⁶Cl uptake in Caco-2 cells

1

1

View larger version (13K): [in this window] [in a new window]   Fig. 9.   Effect of thyroxine on OH gradient-stimulated 35SO and 36Cl uptake in Caco-2 cells. Postconfluent Caco-2 cells were serum-starved for 24 h and then incubated for 24 h with 100 nM of thyroxine or vehicle alone (NaOH). 35SO (A) and 36Cl (B) uptakes were determined as described in MATERIALS AND METHODS. Data are a percentage of uptake compared with control (100%). Results are means ± SE of 8 uptake determinations (*P < 0.05) performed on 4 separate occasions.

View larger version (13K): [in this window] [in a new window]   Fig. 10.   Effect of thyroxine on the kinetics of 35SO uptake in Caco-2 cells. Postconfluent Caco-2 cells were treated with thyroxine () or vehicle (NaOH) alone () as described in Fig. 9. 35SO uptake was measured in the presence of increasing concentration of SO. Michaelis-Menton plot of 6 uptake determinations performed on 3 separate occasions is shown.

The effect of thyroxine on the relative abundance of human DRA mRNA

View larger version (34K): [in this window] [in a new window]   Fig. 11.   Effect of thyroxine on the relative abundance of DRA mRNA in Caco-2 cells. A: representative RNase protection assay autoradiogram for human DRA and GAPDH (internal control). Lanes A, C, and E: protected fragments for human DRA and GAPDH (indicated by arrows) after hybridization to total RNA extracted from Caco-2 cells incubated for 12, 20, and 24 h with vehicle alone (NaOH), respectively. Lanes B, D, and F: protected fragments for human DRA and GAPDH from total RNA extracted from Caco-2 cells incubated with 100 nM thyroxine for 12, 20, and 24 h, respectively. B: percentage of relative abundance of human DRA mRNA in 24-h thyroxine treated Caco-2 cells compared with control (cells treated with vehicle alone). Results are means ± SE (*P < 0.05) from 4 independent experiments.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIAL AND METHODS RESULTS DISCUSSION REFERENCES

In our current studies, we have established an in vitro cellular model to study sulfate and chloride transport and have defined the possible role of DRA in the anion transport in the human intestine. Our results demonstrated that the Caco-2 cell line could serve as a suitable model to study sulfate and chloride transport. The data of our studies demonstrated and characterized the presence of an apical SO/OH exchange process in Caco-2 cells. The findings consistent with a carrier-mediated activity for the OH gradient-stimulated ³⁵SO uptake are: 1) ³⁵SO uptake exhibited saturation kinetics; 2) ³⁵SO uptake was significantly inhibited by DIDS (the AE inhibitor); and 3) oxalate and chloride (but not butyrate, formate, lactate, succinate, and nitrate), competitively inhibited the SO/OH exchange process, indicating the presence of an anion antiporter that could use oxalate and chloride as alternative substrates in addition to sulfate. Furthermore, the results of the present studies indicated that the apical SO/OH and Cl/OH exchange in Caco-2 cells are distinct processes based on the following: 1) SO/OH exchange activity was more sensitive to inhibition by DIDS compared with Cl/OH exchange process; 2) SO/OH exchange process was inhibited by DIDS in a mixed-type manner, whereas Cl/OH exchange was competitively inhibited by DIDS; 3) incubation with 100 nM thyroxine inhibited the OH gradient-stimulated ³⁵SO but not ³⁶Cl uptake in Caco-2 cells; and 4) in parallel to reduced SO uptake, thyroxine treatment also reduced the relative abundance of DRA mRNA.

Previous studies have shown that postconfluent differentiated Caco-2 cells possess many of the functional and structural characteristics of the native enterocyte including similar transport mechanisms and regulatory pathways (27). Therefore, Caco-2 cells have been previously used as a model to characterize the electrolyte transport in the small intestine and colon (40). Silberg et al. (37) have recently shown that the DRA gene is expressed in postconfluent but not preconfluent Caco-2 cells. Consistent with the previous studies of Silberg et al., our results showed that 5-7 days postconfluent but not preconfluent Caco-2 cells expressed the DRA gene (data not shown) and possess apical SO/OH and Cl/OH exchange activities. Therefore, these cells have been utilized here as an in vitro cellular model to study the regulation of the DRA expression by thyroxine and delineate the interactions between sulfate, chloride, and oxalate transport in the human intestine.

The apical SO/OH exchange in the polarized monolayers of Caco-2 cells appeared to be a carrier-mediated process. The SO/OH exchanger in Caco-2 cells exhibited kinetic characteristics similar to SO/OH or SO/HCO exchangers in other systems. For instance, K_m of 0.2 mM for SO uptake in Caco-2 cells is comparable to SO/OH exchanger in rabbit ileal BBM (0.475 mM; Ref. 34), SO/HCO exchanger in rabbit ileal BLM (0.122 mM;Ref. 20), rat liver lysosomal sulfate transporter (0.213 mM; Ref. 8), and human proximal colon BBM (0.8 mM; Ref. 39). Furthermore, the SO/OH exchanger in Caco-2 cells appears to be highly sensitive to DIDS. Our results showed that the effect of DIDS on ³⁵SO uptake occurs in a linear mixed type of inhibition with a K_i of 0.9 ± 0.3 µM for DIDS. This K_i value is also comparable with the K_i of DIDS inhibition of SO/HCO exchanger in the rabbit ileum BLM (6 µM; Ref. 20). Although the mechanism of inhibition by DIDS of the SO/OH exchange process needs more detailed characterization, this kind of inhibition suggests that DIDS and SO have different binding sites on the exchanger and that DIDS, upon binding to its site, may alter the affinity of the transporter for substrates probably via inducing conformational changes in the antiporter.

The apical SO/OH exchange process in Caco-2 cells appeared to be specific for SO, Cl, and oxalate but not for the other anions such as nitrate and short chain fatty acids. Our kinetic studies demonstrated that both Cl and oxalate inhibited the OH gradient-stimulated ³⁵SO uptake in Caco-2 cells in a competitive manner. The K_i values for Cl and oxalate suggest that although the exchanger can transport Cl and oxalate in addition to SO, it has higher affinity for SO (K_m = 0.2 mM) and oxalate (K_i = 0.2 mM) compared with Cl (K_i = 5.9 mM). In agreement with the previous studies of Silberg et al. (37), which demonstrated that DRA gene product is a sulfate and oxalate transporter, our results indicate that DRA may be responsible for the SO/OH exchange process in Caco-2 cells. On the other hand, previous studies in the human proximal colon and rabbit ileum (18, 21) have shown that nitrate and bromide could substitute for chloride in the Cl/HCO (OH) exchange processes. Because the SO/OH exchanger in the current study exhibited lower affinity for chloride and nitrate did not alter the OH gradient-stimulated ³⁵SO uptake, the data of the current study suggest that SO/OH and Cl/HCO (OH) exchange activities in the human intestine may be distinct processes.

We have previously characterized a Cl/HCO₃ (OH) exchange process in the human small intestine and proximal colon (21, 29). These studies demonstrated that Cl uptake into the human colonic apical membrane vesicles was stimulated in the presence of a pH gradient. The OH gradient-stimulated Cl uptake into these vesicles was further stimulated in the presence of a HCO gradient. The data of these studies clearly demonstrated that in the human small intestine and colon, Cl/HCO and Cl/OH exchange activities were mediated via the same transporter. In the present study, we intended to examine whether Caco-2 cells could also serve as a suitable model to study chloride transport. Our data demonstrated the presence of an outwardly OH gradient-stimulated ³⁶Cl uptake that was linear as a function of time and could be inhibited by DIDS. Because Caco-2 cells were derived from human colon adenocarcinoma and when confluent demonstrate the characterstics of differentiated enterocytes (27), it is most likely that the OH gradient-driven ³⁶Cl uptake into these cells represents the activity of the same transporter that is also responsible for Cl/HCO exchange process. In this regard, previous studies by Rajendran and Binder (28) have shown the presence of two distinct transporters, Cl/OH and Cl/HCO exchangers in rat distal colonic apical membrane vesicles. In these studies, the pH gradient-dependent ³⁶Cl uptake was not (unlike our previous studies with human colonic apical membrane vesicles) affected by imposing a HCO gradient. Additionally, in these studies with rat colonic apical membrane vesicles, bumetanide was shown to preferentially inhibit the Cl/OH but not the Cl/HCO exchange process (28). In contrast, in our current studies, bumetanide inhibited both ³⁶Cl and ³⁵SO uptake to the same extent (~30-40%, data not shown) ruling out the possibility that, similar to rat distal colon, Caco-2 cells may also possess two different Cl/HCO₃ and Cl/OH exchange processes with one that could take ³⁵SO as a substrate.

Similar to our previous findings in the human proximal colonic apical membrane and to other intestinal AEs such as of rabbit and human ileum (13, 18, 21, 29), the Cl/OH exchange process in Caco-2 cells appeared to be relatively less sensitive to inhibition by DIDS. It has previously been suggested that the possible explanation for the poor inhibition may be a result of competition with the substrate (5). In agreement with that, our findings showed that Cl/OH exchange process in Caco-2 is competitively inhibited by DIDS with a K_i of 0.3 mM. The K_i value for Cl uptake inhibition by DIDS is also comparable with other systems such as Cl/HCO exchange in the rabbit ileal basolateral membrane, with a K_i of 0.28 mM (19). The fact that SO/OH and Cl/OH exchange activities in Caco-2 cells have different sensitivities and mechanisms of inhibition by DIDS with different K_i values further supports the notion that they are mediated via distinct transporters. Recent studies (22, 23) have shown that DRA is capable of transporting Cl and suggested that DRA is the intestinal apical Cl/HCO exchanger. The data in the current study showed that SO uptake is competitively inhibited by Cl with a K_i of 5.9 mM indicating that SO/OH exchanger is able to transport Cl but with a low affinity compared with SO.

The thyroid hormone thyroxine has been shown to have a widespread effect on membrane transport of amino acids (12), glucose (35), and ions (6, 8, 10) in various tissues and cell lines. For instance, Cano et al. (6) demonstrated a stimulation of renal Na⁺/H⁺ exchanger by transcripitional activation in OK cell line by thyroxine. Furthermore, Chou et al. (8) have reported that the lysosomal sulfate transport in the rat liver is decreased by thyroxine treatment. Additionally, thyroxine has been previously shown to play an important role in the developmental changes of a number of intestinal digestive enzymes and transporters (14). In this regard, Chow et al. (10) have previously shown that thyroxine treatment to suckling rats was followed by a 60% decrease in the expression of AE2 in the small intestine. The novel finding in the current study is that, in Caco-2 cells, 24-h treatment with