INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

SODIUM TRANSPORT in chicken hindgut (cecum, colon, coprodeum) is dependent on dietary and/or hormonal state (5, 6, 18, 26, 28). On a normal or high-NaCl diet, Na⁺ is transported across the colonocyte via both Na⁺-organic solute cotransport systems (amino acids, glucose) and electroneutral NaCl absorption (18). There are apparently only small numbers of active Na⁺ channels in chicken colonic luminal apical membrane under these conditions. However, on a chronic low-NaCl diet, the colonic apical transport mode switches. Na⁺-organic solute cotransport systems are replaced by amiloride-inhibitable Na⁺ uptake processes, many of which are electrogenic and have been assumed to be entirely due to Na⁺ channels (5, 26, 32), although there is also evidence of stimulation of electroneutral Na⁺- and Cl-linked absorption (18). The concentration of plasma aldosterone increases with chronic NaCl depletion, and aldosterone is involved in mediating changes in Na⁺ transport (27).

In contrast to the stimulation of NaCl absorption by the low-NaCl diet and the elevated blood aldosterone, sugar uptake by chicken jejunal and colonic epithelial cells is reduced (5, 6, 16, 26). The mechanism of the change in glucose absorption is unknown. Preliminary observations concluded that the effects of the low-Na⁺ diet on chicken small intestinal Na⁺-dependent glucose absorption might be due to the observed reduction in the membrane potential difference (16).

Concerning Na⁺-dependent nutrient absorption from avian colon, previous reports have shown that the avian colon transports amino acids and sugars against a concentration gradient under conditions of high- or normal NaCl intake (3, 16). The physiological significance of the presence in the colon of an active, phloridzin-sensitive sugar transport system is unknown. Because birds do not have a urinary bladder, the urine is emptied into the cloaca, and retrograde flow carries urine into the coprodeum, colon, and cecum. As a result, the epithelia of the lower intestine process the ileal and ureteral outflow. It has been suggested that the amiloride-inhibitable Na⁺ absorption of the coprodeum and colon is to conserve NaCl, whereas Na⁺-organic solute cotransport is important in counteracting the osmotic water loss into colon in chickens on a high-NaCl diet, i.e., in conserving water (21, 26).

We recently demonstrated that chicken and mammalian intestines contain both epithelial Na⁺/H⁺ exchange isoforms NHE2 and NHE3 in the brush border (BB) (10, 15, 36). However, the relative role of each in intestinal transport has only been partially defined. The ability to separate the contribution of NHE2 and NHE3 functionally, by using a pharmacological approach with (3-methylsulfonyl-4-piperidinobenzoyl)guanidine methanesulfonate (HOE-694) (7, 22, 37), allowed us to examine the effect of adaptation to a low-NaCl diet on the activity of these two epithelial exchangers in the BB of chicken colon and ileum. The results indicate that NHE2 is the major BB Na⁺/H⁺ exchanger of chicken colon, whereas in ileum NHE2 and NHE3 contribute equally to basal apical Na⁺/H⁺ exchange. In addition, a low-NaCl diet increased the amount of ileal and colonic BB membrane NHE2 while decreasing the amount of ileal and colonic Na⁺-dependent glucose absorption. The ileal but not the colonic Na⁺-dependent glucose transporter has been identified as SGLT1, and the amount of ileal SGLT1 is decreased by the low-NaCl diet.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Animals and diets. Two groups of Hubbard chickens, 10 wk old, were adapted for 2 wk to either a low- or high-NaCl diet (Table 1) (28). The low-NaCl diet consisted of low-NaCl food with a measured Na⁺ content of 0.423 mmol Na⁺/100 g diet and demineralized water containing 1 mM Ca₂Cl. The high-NaCl diet consisted of the same diet with NaCl added (1 g NaCl/100 g diet) and demineralized water containing 1 mM CaCl₂ plus 0.5 g NaCl/100 ml water and contained 18 mmol Na⁺/100 g diet. In preliminary studies, a third group of chickens fed a commercial, conventional Na⁺ diet containing 6.7 mmol Na⁺/100 g diet, was studied without a period of adaptation to a change in diet.

Plasma aldosterone analysis. Blood samples were collected from a wing vein into an ice-cold heparinized tube and centrifuged immediately at 4°C and assayed by RIA by Cerba Laboratories (Barcelona, Spain).

BB membrane vesicle preparation. After the birds were killed with intravenous urethan, the ileum and colon were removed, rinsed with ice-cold 0.9% NaCl, opened longitudinally, wrapped in aluminum foil, dropped into liquid nitrogen, and frozen at 80°C until use.

Protein and enzyme assays. Protein was measured by the Bradford method using -globulin as the standard. Citrate synthase activity, a mitochondrial marker, was assayed by the method of Srere (29). Na⁺-K⁺-ATPase and K⁺-ATPase activities were assayed by the method of Del Castillo et al. (9). Na⁺-K⁺-ATPase was defined as the difference in ATPase activity measured in the presence of 100 mM Na⁺, 20 mM K⁺, and 5 mM Mg²⁺ with or without 2 mM ouabain. K⁺-ATPase activity was defined as the difference in ATPase activity measured in the presence of 20 mM K⁺ and 5 mM Mg²⁺ with or without 2 mM ouabain (21). Sucrase activity was assayed by the method of Dahlquist (8).

Uptake studies. ²²Na or D-[¹⁴C]glucose uptake was measured at 25°C by a rapid filtration technique as described by Cano et al. (4). The uptake was initiated by adding 10 µl of membrane vesicle suspension to 90 µl of uptake buffer, and transport was carried out for ²²Na for 15 s or 1 min in ileum and 15 s or 5 min in colon and for D-[¹⁴C]glucose for 30 s or 1 min in ileum and colon, respectively (15 s was during period of linear uptake for ²²Na at up to 1 mM Na⁺; 30 s and 1 min were during time of maximal D-glucose overshoots in ileum and colon, respectively).

Antibodies. Polyclonal anti-NHE2 antibody (Ab 597) and anti-NHE3 antibody (Ab 1380) were made in rabbits as described previously (15, 33). Ab 587 and Ab 1380 were affinity purified, as previously described (15). Affinity-purified GST/NHE2-C87 or GST/NHE3-C85 fusion proteins (GST fused to COOH-terminal 87 and 85 amino acids of NHE2 and NHE3, respectively) (15, 33) were separated by 12% SDS-PAGE, transferred to a nitrocellulose membrane, and used for affinity purification of anti-NHE2 and -NHE3 antibodies. The fusion protein on the nitrocellulose was then identified by Ponceau S staining, cut out, and washed extensively with PBS. This strip of fusion protein was then exposed to blocking buffer [5% nonfat dried milk (NFDM)] in Tris-buffered saline [TBS; 150 mM NaCl and 13 mM Tris (pH 7.5)] for 1 h and incubated with the corresponding Ab 597 or Ab 1380, respectively, overnight at 4°C (33). The strip was then washed extensively (5 times, 5 min/wash) in washing buffer (1% Triton X-100 in TBS), and the bound antibody was eluted with 0.2 M glycine · HCl (pH 2.5) and immediately neutralized with 2 M Tris to pH 7.5 (11). BSA (1%) was then added to stabilize the affinity-purified antibody and dialyzed in PBS overnight. Polyclonal anti-SGLT1 antibody was prepared as described (30). Polyclonal anti-villin antibody was purchased from Immunotech (Marseilles, France).

SDS-PAGE and Western analysis. Similar amounts of protein of crude cell membranes or purified BBs from high and low-NaCl diet chickens were solubilized in Laemmli sample buffer and resolved by 8% SDS-PAGE. Proteins were electrotransferred onto nitrocellulose membranes. Western blotting was performed by blocking nitrocellulose membranes with solution containing 5% NFDM in TBS for 1 h followed by incubation with primary antibody for 1 h (diluted in blocking solution) at room temperature. Nitrocellulose membranes were then washed extensively with 0.02% Triton X-100 in TBS. After washing, the nitrocellulose membranes were incubated with horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (Jackson ImmunoResearch Laboratory) for 1 h at 1:5,000 dilution in blocking solution. Excess secondary antibody was again washed, and the bound secondary antibody was detected by enhanced chemiluminescence (Renaissance, DuPont). Simultaneously studied molecular mass standards (Bio-Rad high-range standards) were used on each gel.

Immunohistochemistry. Immunohistochemistry was performed as described (15). The chicken ileum and colon were fixed in 3% neutral buffered formaldehyde solution overnight, subsequently processed on a Technicon, and embedded in paraffin. Paraffin sections were cut at 4 µm and mounted on gelatin-coated slides. The slides were dewaxed in xylene, rehydrated sequentially in 100, 95, and 70% ethanol for 3 min each, and rinsed in PBS buffer, and endogenous peroxide was blocked by incubation for 10 min in 0.3% H₂O₂ in methanol. The slides were blocked in PBS buffer (1% NFDM) with goat serum and incubated overnight at 4°C with primary antibody. Ab 597 and Ab 1380 were centrifuged at 100,000 g at 4°C for 60 min before use; control staining was performed with secondary antibody only (goat anti-rabbit IgG, Vector Laboratories, CA), rabbit IgG, and preimmune serum from the same rabbits in which the antibodies were raised. The labeling was visualized using horseradish peroxidase Vesctastain elite ABC kit (Vector Laboratories, CA) for light microscopy. Slides were interpreted by two observers.

Calculations and statistics. Results are expressed as means ± SE. The kinetic parameters [maximal rate of enzyme reations (V_max) and Michaelis constant (K_m)] were determined using a nonlinear regression program to best fit the data (Enzfitter program). Statistical significance was evaluated by the Student's t-test for unpaired observations.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Model standardization. Low-Na⁺ diet elevates plasma aldosterone. The mean plasma aldosterone concentrations (in pg/ml) were 10 ± 1 for the chickens on a high-NaCl diet and 207 ± 19 for chickens adapted to a low-NaCl diet (n = 26 for each group, P < 0.001). In preliminary studies, chickens on a conventional diet had plasma aldosterone levels of 8 and 10 pg/ml compared with simultaneously studied chickens on a high-NaCl diet of 9, 9, and 10 pg/ml. These results showed that the low-NaCl diet caused the increase in aldosterone. In subsequent studies, chickens on high- and low-NaCl diets were compared so that both groups underwent similar times of diet adaption.

Characterization of BBMV preparations. The purification of the ileal membrane vesicle preparations was evaluated by measuring the specific activity of the apical membrane enzyme sucrase, the BLM Na⁺-K⁺-ATPase, and a mitochondrial enzyme (citrate synthase) (Table 2). Enrichment factor was calculated in reference to the homogenate specific activity. The purification of ileal BB membranes from chickens on a high-NaCl diet is based on the ~18.5-fold enrichment of sucrase, the lack of enrichment of Na⁺-K⁺-ATPase, and depletion of citrate synthase (enrichment factor <1). The purification of the colonic BBMV preparations from chickens on a high-NaCl diet was evaluated by measuring the specific activity of the colonic luminal membrane enzyme K⁺-ATPase (1, 9), BLM marker Na⁺-K⁺-ATPase and mitochondrial marker citrate synthase (Table 2). The purification of colonic BB membranes is based on the 13-fold enrichment of K⁺-ATPase specific activity compared with a threefold enrichment of Na⁺-K⁺-ATPase and a depletion of citrate synthase.

1

1

1

1

Na⁺ uptake into BBMV isolated from chicken colon or ileum. ²²Na uptake was performed into BBMV from ileum (1 min) and colon (5 min) from birds on high- and low-NaCl diets. Uptake was with an inwardly directed pH gradient [pH inside (pH_i) 5.5/pH outside (pH_o) 7.5] and in the absence of a pH gradient (pH_i = pH_o = 5.5), with both conditions studied in the absence and presence of HOE-694 (50 µM) in the transport buffer. An inwardly directed pH gradient (pH_i 5.5/pH_o 7.5) stimulated Na⁺ uptake into BBMV isolated from both colon and ileum (Fig. 1). HOE-694 partially inhibited Na⁺ uptake in the presence but not in the absence of a pH gradient (Fig. 1). A low-NaCl diet increased pH-dependent Na⁺ uptake and had no significant effect on the uptake measured either in the presence of HOE-694 or absence of a pH gradient. Na⁺/H⁺ exchange calculated from the data of Fig. 1 are shown in Table 3. Na⁺/H⁺ exchange was defined as the Na⁺ uptake in the presence minus that in the absence of the inwardly directed pH gradient, which we had previously shown for ileum (13, 21) and colon (2) to be the same as the difference of Na⁺ uptake with an acid pH_i gradient in the presence and absence of 1 mM amiloride included in the transport buffer. Both ileal and colonic BBMV had increased Na⁺/H⁺ exchange caused by the low-NaCl diet, with a similar increase in Na⁺/H⁺ exchange in colon and ileum (~81% in both). In preliminary studies, simultaneously studied chickens on a conventional diet and high-NaCl diet had similar ileal and colonic total BB Na⁺/H⁺ exchange (data not shown).

View larger version (21K): [in this window] [in a new window]   Fig. 1.   Effect of diet and HOE-694 (50 µM) on Na+ uptake into brush-border membrane vesicles (BBMV) isolated from chicken colon (A) and ileum (B). Na+ uptake (0.1 mM Na+) was measured in presence [inside pH (pHi) 5.5/outside pH (pHo) 7.5] and absence (pHi 5.5/pHo 5.5) of a transmembrane proton gradient. Composition of intravesicular and extravesicular buffers are given in MATERIALS AND METHODS. Concentration of HOE-694 was 50 µM, a concentration that has minimal effect on Na+/H+ exchanger NHE3 and totally inhibits NHE2 (33). Vesicles were preincubated with inhibitor for 10 min. Each point represents mean ± SE of 4 separate membrane vesicle preparations for ileum and 5 for colon (triplicate assay of each point). * P values are compared with data obtained with high-NaCl diet (unpaired t-test). # HOE-694 decreased Na+ uptake in presence of a proton gradient in both diets in ileum and colon, P < 0.001.

View larger version (20K): [in this window] [in a new window]   Fig. 2.   Increasing concentrations of external Na+ on colonic (A) and ileal (B) brush-border (BB) Na+ uptake (15 s) and effect of low-NaCl diet. Na+ uptake was measured in presence (, ) and absence () of proton gradient. In colon and ileum, there was no effect of low-NaCl diet on uptake with no pH gradient, and results shown are combined. Difference, uptake in presence minus that in absence of a proton gradient. , High-Na+ diet; , low-Na+ diet. Osmolarity due to different Na+ concentrations was maintained with mannitol. Each point represents mean ± SE of triplicate assays using 6 separate BB vesicle preparations. * P < 0.001 compared with data obtained with high-NaCl diet.

Presence of NHE2 in ileal and colonic BB and change with low-NaCl diet. Western analysis was performed on BBs of ileum and colon from high NaCl and low-NaCl diet birds, and anti-NHE2 Ab 597 was used to identify NHE2 (Fig. 3). NHE2 was identified as a single 87-kDa protein in colon and ileum of chicken. Of note, in rabbit ileum, NHE2 is a slightly larger protein of 95 kDa (37). Whether differences in glycosylation explain this difference in size is not known. As shown in Fig. 3A, low-NaCl diet caused an increase in amount of NHE2 both in ileal and colonic BB. In contrast, there was no increase in amount of villin, another BB protein (Fig. 3B). There was no change in amount of ileal and colonic BB NHE3, based on Western analysis (data not shown).

View larger version (77K): [in this window] [in a new window]   Fig. 3.   Low-NaCl diet increases amount of ileal and colonic BB NHE2. A: Western analysis (8% SDS-PAGE) performed on BBMV from ileum and colon of birds on a high- and low-NaCl diet using Ab 597. Data from 1 experiment are shown with low-Na+ diet increasing amount of NHE2 in ileal and colonic BB. NHE2 is shown at arrow. Molecular mass standards are on left. Similar results were obtained in 2 additional studies. B: amount of villin was identified on parallel samples electrophoresed under identical conditions using anti-villin antibody. Villin is shown as arrow. Note that, although there is more NHE2 in low-Na+ diet, there is slightly more villin in high-Na+ diet.

Immunohistochemistry of NHE2 and NHE3 distribution in ileum and colon. Anti-NHE2 Ab 597 and anti-NHE3 Ab 1380 were used for immunocytochemical analysis of ileum and colon from birds on high and low-NaCl diets. Both NHE2 and NHE3 were present in both ileum and colon in BB of epithelial cells on the villus and surface of ileum and colon, respectively. Neither were present in the BLM of these cells nor were they present in goblet cells (Fig. 4). No differences were seen in the distribution of NHE2 or NHE3 in the high- and low-NaCl diets (data not shown).

View larger version (113K): [in this window] [in a new window]   Fig. 4.   NHE2 and NHE3 are brush-border membrane proteins in chicken ileum and colon. Immunocytochemistry of chicken ileum (A and B) and colon (C and D) stained with anti-NHE2 Ab 597 (A and C) and anti-NHE3 Ab 1380 (B and D). All magnifications are ×160. Representative of 3 similar studies.

Glucose uptake into BBMV from colon and ileum: effect of a low-NaCl diet. Figure 5 shows that D-glucose uptake into BBMV isolated from both ileum and colon is Na⁺-dependent and sensitive to phloridzin. Data shown were obtained at times of peak glucose overshoots, 30 s in ileum and 1 min in colon (Ilundain, unpublished observations). Adaptation to a low-NaCl diet decreased sugar uptake by 66 ± 2% (decrease of 76% in Na⁺-dependent D-glucose uptake) in colon and 30 ± 3% (decrease of 29% in Na⁺-dependent D-glucose uptake) in ileum.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Effect of low-NaCl diet on D-glucose uptake into BBMV isolated from chick colon (A) and ileum (B). D-Glucose uptake was measured as indicated in MATERIALS AND METHODS over 30 s in ileum and 1 min in colon (time of peak overshoot values). Each point represents mean ± SE of triplicate assays using 7 separate vesicle preparations. * P < 0.001 compared with values obtained with high-NaCl diet.

Kinetics of D-glucose uptake into BBMV from colon of chickens fed either a high- or low-NaCl diet. D-Glucose uptake into colonic BBMV was measured at different external D-glucose concentrations in the presence and absence of phloridzin at a time (15 s) that D-glucose uptake was linear at all external glucose concentrations studied. In the absence of phloridzin, D-glucose uptake is curvilinear, but in the presence of phloridzin, D-glucose uptake showed a linear relationship with extravesicular D-glucose concentration (Fig. 6). The difference between total colonic BBMV D-glucose uptake and that observed in the presence of phloridzin was used to estimate the apparent K^Glc_m and V_max for Na⁺-dependent D-glucose uptake (Table 5). The results show that adaptation to a low-NaCl diet decreased the V_max for D-glucose uptake in colonic BB without affecting the K^Glc_m. In addition the low-NaCl diet did not alter the glucose permeability of the vesicles, calculated as V_max/glucose concentration (Table 5).

View larger version (17K): [in this window] [in a new window]   Fig. 6.   Effect of increasing concentrations of external D-glucose on initial rate of D-glucose uptake into colonic BBMV. A: total effect of low-NaCl diet. D-Glucose uptake was measured over 15 s in presence (, ) and absence (, ) of 0.25 mM phloridzin. B: difference, i.e., total uptake minus that in presence of phloridzin. Osmolarity was maintained with mannitol.  and , High-Na+ diet;  and , low-Na+ diet. Each point represents mean ± SE of 7 separate membrane vesicle preparations (triplicate assay of each point). * P < 0.001 compared with values obtained with high-NaCl diet.

Effect of membrane potential gradient and/or Na⁺ gradient on D-glucose uptake into BBMV isolated from colon and ileum of chickens fed either a high- or low-NaCl diet. The effect of the electrical membrane potential and/or Na⁺ concentration gradient on glucose uptake by BBMV was investigated. An inside-negative membrane potential was generated by an intravesicular-to-extravesicular K⁺ gradient in the presence of valinomycin. An inside-directed Na⁺ gradient was created by the addition of 50 mM Na⁺ to the extravesicular buffer, with the intravesicular buffer made nominally Na⁺ free. Figure 7 shows that, in both ileum and colon, a Na⁺ concentration gradient increased sugar uptake, and this uptake was further stimulated by the presence of inside-negative electrical membrane potential.

View larger version (19K): [in this window] [in a new window]   Fig. 7.   Effect of low-NaCl diet on D-glucose uptake into colonic (A) or ileal (B) BBMV in presence and absence of electrical membrane potential and/or Na+ gradient. BBMV were loaded with a pH 7.5 buffer consisting of (in mM) 140 mannitol, 50 potassium gluconate, 0.1 MgSO4, and 50 HEPES-Tris. Transport buffer contained 100 µM -methylglucose, tracer D--[14C]methylglucose, 50 mM HEPES-Tris (pH 7.5), 45 µM valinomycin, 140 mM mannitol, and 50 mM sodium gluconate (electrochemical Na+ gradient); 40 mM mannitol, 50 mM sodium gluconate, and 50 mM potassium gluconate (Na+ concentration gradient); or 140 mM mannitol and 50 mM potassium gluconate (no electrochemical Na+ gradient). Each point represents mean ± SE of triplicate assays using 6 separate vesicle preparations. * P < 0.001 as compared with values obtained with high-NaCl diet.

Effect of low-NaCl diet on amount of lleal BBMV SGLT1. Western analysis was used to determine the effect of low-NaCl diet on amount of ileal SGLT1. As shown in Fig. 8, chicken ileal SGLT1 was recognized as a 65-kDa protein. The low-NaCl diet decreased the amount of ileal BB SGLT1. The anti-SGLT1 antibody (30, 34) failed to identify any protein in colonic BB from birds on a high- or low-NaCl diet (Fig. 8).

View larger version (30K): [in this window] [in a new window]   Fig. 8.   Decrease in amount of Na+-dependent glucose transporter SGLT1 by low-NaCl diet in ileal BBMV. BBMV were prepared from chicken ileum and colon after 15 days on a high- or low-NaCl diet, and Western analysis (8% SDS-PAGE) was performed with anti-SGLT1 antibody as described (23, 24). SGLT1 was detected in ileum as a 65-kDa protein (arrow). Molecular mass standard is shown on left. Amount of SGLT1 was decreased in ileal BBMV from birds on a low-NaCl diet. No signal was detected in colon. Representative experiment shown was repeated 7 times with similar results.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

The current studies add further mechanistic information concerning the adaptation in nutrient and Na⁺ absorption of the chicken intestine to a low-NaCl diet. We show that a low-NaCl diet 1) activates ileal and colonic apical Na⁺ uptake via stimulation of BB Na⁺/H⁺ exchange, which is associated with an increase in amount of BB NHE2, and 2) decreases ileal and colonic BB Na⁺-dependent D-glucose transporter (V_max), which is associated with a decrease in SGLT1 amount in ileal BB. The identity of the colonic Na⁺-dependent glucose transporter is not known, and thus no comment can be made about effects of the low-NaCl diet on its expression. The demonstration that, in birds (chicken), NHE2 accounts for most of the colonic BB Na⁺/H⁺ exchange and all of the low-NaCl diet-induced increase in ileal and colonic BB Na⁺/H⁺ exchange represents the first identified physiological role for NHE2.

These studies extend previous studies of the effect of several weeks of a low-NaCl diet and its induction of an elevation in plasma aldosterone level on chicken colonic nutrient and Na⁺ transport (5, 6, 16, 18, 26-28). Previous studies showed that a low-NaCl diet increased Na⁺ absorption and decreased glucose and amino acid uptake in chicken cecum, colon, and coprodeum (5, 6, 16, 18, 26, 28). This was associated with structural changes consisting of more epithelial cells and increased microvillus size of each cell (19, 20). The increase in Na⁺ absorption in the colon has been assumed to be due to an increase in apical membrane Na⁺ channels, since the colon of chickens on a low-NaCl diet has an increased short-circuit current that is amiloride sensitive and is present even in the absence of luminal nutrients (5, 26, 32). However, linked NaCl absorption is also increased. This view is based on a study of chicken colon performed with the Ussing chamber-voltage clamp-ion flux technique, which showed that a low-NaCl diet was associated with increased mucosal-to-serosal and net fluxes of Na⁺ and Cl (27). The current studies show that BB vesicles from ileum and colon of chickens on low-NaCl diets have increased Na⁺/H⁺ exchange. Because BB Na⁺/H⁺ exchange is thought to be part of electroneutral NaCl absorption, this is a likely but unproven explanation for the stimulation of ileal electroneutral NaCl absorption previously described (27).

Several technical aspects of this study must be commented on. Because transport is characterized in BBMV from chickens on different diets, how comparable are the vesicles? Enrichment of ileal sucrase was comparable, and thus ileal BB enrichment compared with homogenate was similar. However, comparable colonic data are not available because of the known aldosterone-induced increase in K⁺-ATPase activity (24), and it thus remains possible that some of the results are due to diet-induced differences in the makeup of the BBMV. However, as for ileal BBMV, the colonic BBMV from chickens on different diets have the same intravesicular volumes.

Chicken ileal and colonic BBs contain NHE2 and NHE3, but the low-NaCl diet increased BB Na⁺/H⁺ exchange by affecting only NHE2. These studies used differential sensitivity to HOE-694 of NHE2 and NHE3 to separate the contribution to BB Na⁺/H⁺ exchange of these two isoforms (7, 22, 37). HOE-694 K_i for NHE2 is 5 µM and for NHE3 is 650 µM for rat and rabbit isoforms expressed in fibroblasts (7) and rat NHE3 expressed in intestinal and renal cell lines (22). Thus 50 µM HOE-694 is predicted to block all NHE2 (10× K_i) and has minimal effect on NHE3 (<1% K_i of NHE3). However, HOE-694 sensitivity of NHE2 in epithelial cells and on Na⁺/H⁺ exchangers from other species, including chicken, is unknown. Thus an assumption of these studies is that chicken NHE2 and NHE3 have similar sensitivities to HOE-694 to those of the species studied. This pharmacological approach has been used in Na⁺/H⁺ exchange measurements in rabbit ileal and kidney BB studies (37). Using 50 µM HOE-694 sensitivity to separate the contributions of NHE2 and NHE3 to chicken BB Na⁺/H⁺ exchange, we now show that NHE2 is the Na⁺/H⁺ exchanger isoform that accounts for the entire increase in chicken ileal and colonic BB Na⁺/H⁺ exchange in response to the elevation in plasma aldosterone. NHE2 also accounts for nearly all basal chicken colonic BB Na⁺/H⁺ exchange and ~50% of ileal basal Na⁺/H⁺ exchange. Whether NHE2 or NHE3 is the primary colonic BB Na⁺/H⁺ exchanger in other species, including mammals, is not known. However, it is known that both NHE2 and NHE3 are present on the intestinal BB but not BLM on the basis of immunocytochemical studies of the colons of human, rabbit, rat, and mouse (15) and as shown here in chicken ileum and colon. There are several suggestions that NHE2 is a major component of basal colonic BB Na⁺/H⁺ exchange of multiple species, although definitive studies are lacking. These include that in rabbit descending colon, the only identified BB Na⁺/H⁺ exchange isoform is NHE2, whereas NHE3 is absent (15). Also, NHE2 message is at least as high in rat colon as in the small intestine (35) and is highest in the descending colon in the human gastrointestinal tract (11). Further studies are needed to clarify the relative roles of NHE2 and NHE3 in colonic BB Na⁺/H⁺ exchange in mammalian as well as avian models.

How elevated aldosterone increases the NHE2 V_max has not been studied in detail. In the present study, we showed that the amount of NHE2 in ileal and colonic BBMV is increased by the low-NaCl diet. However, it is only speculation that this occurs by stimulation of transcription rather than an effect on NHE2 turnover or other mechanisms. Of relevance is the recent report by Goldstein et al. (14), in which a low-NaCl diet induced an increase in mRNA for the epithelial Na⁺ channel in avian colon (14). This Na⁺ channel was 60% the same as previously characterized mammalian epithelial Na⁺ channel subunits.

Previous studies have not reported the effect of a low-NaCl diet on chicken small intestinal NaCl transport. In other species, mineralocorticoids do not significantly affect active Na⁺ absorption in the small intestine; however, glucocorticoids increase ileal BB Na⁺/H⁺ exchange in multiple species. This is due to an increase in NHE3 activity and amount, with no change in NHE2 activity and amount (37, 38). It is not known whether this response of an increase in NHE2 to mineralocorticoid elevation in the avian small intestine is due to differences in the nature of mineralocorticoid vs. glucocorticoid receptors in chicken and mammalian ileum or to differences in processing of the mineralocorticoid to allow access to the glucocorticoid receptor.

Previous studies on the effect of adaptation to a low-NaCl diet on chicken intestinal nutrient transport showed that chicken jejunum and colon responded with decreased 3-O-methylglucose uptake (16, 18) and that chicken colon had decreased glucose, galactose, and lysine active absorption (17) and abolition of Na⁺-dependent electrogenic glucose and galactose absorption (5, 16, 17, 26). Similarly, the amino acid-induced increase in short-circuit current in chicken colon was abolished (4, 5, 26). At least part of the decrease in nutrient absorption was suggested as being due to a decrease in transmembrane potential difference (16). This was based on a decreased depolarization of the cells by measuring the influx of the membrane potential sensor tetraphenylphosphonium as well as no difference in abundance of jejunal SGLT1 by Western analysis. The current studies are the first to use BB vesicle techniques to probe mechanistically the low-NaCl diet-induced decrease in Na⁺-dependent D-glucose uptake. We showed that, in chicken ileum and colon, the Na⁺ concentration-dependent, phoridzin-sensitive glucose uptake mechanism is greatly reduced by a low-NaCl diet. We also showed that the inhibition of BB Na⁺-dependent D-glucose uptake induced by Na⁺ depletion could not be due to a decreased electrical membrane potential or decreased Na⁺ concentration gradient, since the imposed transmembrane electrical membrane potential or chemical Na⁺ gradient of the BBMV were equal in the high- and low-Na⁺ diet groups under the conditions of our experiments. In addition, in the absence of membrane potential, adaption to a low-Na⁺ diet decreased Na⁺-dependent glucose uptake into the BBMV, and the decrease was of the same magnitude as that observed in the presence of an electrochemical Na⁺ gradient (~45% inhibition in colon and 25% inhibition in ileum). The mechanism of decreased Na⁺-glucose transport in the ileum was related to a decrease in the amount of the Na⁺-glucose cotransport protein SGLT1, although how that occurs is not known.

The mechanism of the decrease in colonic BB Na⁺-glucose uptake is less clear. There was no SGLT1 identified using the same anti-SGLT1 antibody that recognized ileal SGLT1 (34). This antibody showed avian small intestinal SGLT1 as a 65-kDa protein, and on the same Western analysis, it showed rabbit small intestinal SGLT1 as a 75-kDa protein (34). In addition, Northern analysis and RT-PCR identified minimal chicken colonic SGLT1 (Shirazi-Beechey, unpublished observations) suggesting that it is likely that another isoform of Na⁺-glucose cotransporter is present in the chicken colon or that our antibody fails to recognize hen SGLT1. Similarly, Wright and colleagues (23) reported that human colon has only a small amount of SGLT1. The physiological role of any Na⁺-glucose cotransporter in colon has not been adequately studied. The avian colon is also a urinary conduit and, under conditions of the high-Na⁺ diet, contains a luminal glucose concentration of 1-2 mM (A. A. Ilundain, C. de la Horra, and M. L. Calonge, unpublished observations).

Thus these studies add mechanistic understanding to the change induced by several days or weeks of elevated plasma aldosterone on chicken ileal and colonic Na⁺ and nutrient transport. It appears that, to hold onto water and Na⁺, there is induction of an increase in amount of ileal and colonic BB NHE2 activity along with the previously described induction of Na⁺ channel activity, whereas there is switching away from Na⁺-dependent nutrient uptake systems, including an Na⁺-linked D-glucose uptake system. The latter, at least in ileum, is associated with a decrease in amount of SGLT1 and in colon is associated with decreased V_max Na⁺-glucose transport. In addition, the number of colonocytes and the surface area of each are expanded by an increase in amount of microvillus membrane, the goal of which is to lead to increased Na⁺ absorption. Because all transport data reported here were normalized to amount of BB protein, the changes in both surface area with elevated aldosterone and in Na⁺/H⁺ exchange activity are separate and additive. Whether a common regulatory step is involved in changing the amount of ileal NHE2 and SGLT1 in opposite directions is unknown, although this is possible, and a transcriptional mechanism must be considered. Of note, however, is the remaining unresolved question of whether all effects of a low-Na⁺ diet can be attributed to changes in plasma aldosterone levels (31).

Perspectives