Experimental models of hypertension, such as spontaneously hypertensive rats (SHR), show alterations in cellular sodium transport that affects Na+-coupled cotransport processes and has been involved in the pathogenesis of this disease. The objective of the present study was to analyze the kinetic properties of the sodium-dependent glucose transport in the jejunum and ileum of SHR and its genetic control, Wistar-Kyoto (WKY) rats, as well as the regulation of the transporter, SGLT1. In hypertensive rats, the increased systolic blood pressure was accompanied by an enhancement of serum aldosterone levels compared with WKY rats, but no alterations were found in their body weight or serum glucose/insulin levels. The values for d-glucose maximal rate of transport (Vmax) were 42 and 60% lower, respectively, in the jejunum and ileum of SHR than those from WKY rats. On the other hand, the values for the Michaelis constant (Km) were similar in both animal groups, as was the diffusive component of transport (Kd). Immunoblotting and Northern blot analysis revealed the existence of a lower abundance of SGLT1 protein and mRNA in SHR. Moreover, hypertensive rats showed a decrease in the molecular mass of SGLT1 that could not be explained in terms of different glycosylation and/or phosphorylation levels or an alternative splicing in the expression of the protein. These findings demonstrate that SGLT1 is regulated at a transcriptional level in the intestine of hypertensive rats, and suggest that this transporter might participate in the dysregulation of sodium transport observed in hypertension.
- Northern blot
- spontaneously hypertensive rats
- Western blot
arterial hypertension is a widespread health problem that affects many individuals worldwide. This disorder is a major risk factor for a variety of important diseases including stroke, myocardial infarction, congestive heart failure, and end-stage renal disease. Although the precise mechanisms involved in the onset of hypertension remain undefined, the application of recent molecular approaches to this disease has allowed its association with a group of alterations, including oxidative stress (57), vascular inflammatory processes (53), and ion transport defects (47), among others. Alterations in the cellular sodium transport (56) are of particular importance because they might be crucial not only in the pathogenesis of hypertension but also in other associated pathologies, such as diabetes and nephropathy (60).
Several reports have found a relationship between hypertension and altered sodium transport. These reports have described inhibition of active sodium transport (26), and increased passive membrane permeability for sodium in essential hypertension (5). In spontaneously hypertensive rats (SHR), a classical experimental model for the study of hypertension, stimulation of Na+/H+ exchanger has been shown compared with their normotensive genetic control, Wistar-Kyoto (WKY) rats (58). Furthermore, similar changes in sodium transport have been found in other hypertensive models, such as Dahl salt-sensitive rats (30, 32), Milan hypertensive rats (48), and fructose-fed rats (9). Most of these studies have been conducted in the kidney, due to the role of renal sodium transport in blood pressure regulation, and very little attention has been paid to the intestine. Nevertheless, an increase in the small intestinal sodium absorption is also present in SHR compared with WKY rats, which suggests a possible role of the small intestine in the pathophysiology of arterial hypertension (39).
The relationship between glucose homeostasis and sodium balance in the organism has been the object of many studies that intend to contribute to the knowledge of morbid disorders including hypertension, diabetes, and obesity, among others (16). Abnormal glucose metabolism with decreased insulin sensitivity has been demonstrated in some hypertensive models, such as Dahl salt-sensitive (54), SHR (45), Milan hypertensive (15), and fructose-fed rats (31). All of these reports demonstrate the association between sodium balance, glucose metabolism, and hypertension.
d-Glucose is normally transported across the intestinal brush-border membranes by a sodium transporter protein, the Na+-glucose cotransporter SGLT1. SGLT1 also functions as a water channel (17) and may, therefore, be important in the regulation of extracellular fluid volume in hypertensive rats. With this background, our interest was addressed to study whether intestinal SGLT1 is involved in the pathogenesis of arterial hypertension. Previous studies in our laboratory have shown a reduction in Na+-dependent d-glucose uptake across jejunal (52) and ileal (43, 51) epithelia from SHR compared with WKY rats, thus suggesting that SGLT1 might be linked to arterial hypertension. In addition, it has recently been reported that both oxidative stress status (24) and cytokines (21) are involved in the regulation of Na+-glucose cotransporter, which may be relevant in the pathology of SHR (53, 57). Therefore, in the present study we decided to investigate the molecular mechanisms involved in the regulation of the Na+-glucose transporter in arterial hypertension, and we evaluated the kinetic parameters of SGLT1-mediated sugar transport and measured protein and mRNA levels of this cotransporter in the jejunum and ileum of hypertensive rats.
MATERIALS AND METHODS
Animals and serum determinations.
Male SHR and WKY rats were obtained at the age of 7–8 wk from Harlan-Iberica (Barcelona, Spain) and housed in standard temperature and humidity conditions with free access to water and a commercial rat chow (Panlab, Barcelona, Spain). To control that diastolic and systolic blood pressure levels were between those expected in each group, measurements were made weekly by the indirect method of tail-cuff occlusion in conscious animals, using an electrosphygmomanometer and physiograph recorder (Letica, Barcelona, Spain). The mean of three or four successive measurements was used as the estimate of blood pressure, and body weight was then determined. At the age of 12–14 wk, rats were fasted for 18 h and then anesthetized with pentobarbital sodium (50 mg/kg ip) and killed by decapitation in the early morning. Blood was obtained from cardiac puncture, centrifuged at 3,000 g for 10 min, and serum aliquots were stored at −70°C until assay. The jejunum and ileum were quickly excised and washed in ice-cold saline buffer and then divided into two pieces, one to isolate enterocytes and the second one to obtain brush-border membrane vesicles (BBMVs) from the mucosa. Serum glucose was determined using a commercial kit (Cima Diagnostics, Sevilla, Spain), and insulin and aldosterone levels in serum were measured by radioimmunoassay (Immunotech, Barcelona, Spain). All procedures were approved by the Ethical Committees of the University of Seville and the University of Barcelona, in accordance with the Spanish legislation for the use and handling of experimental animals.
Enterocytes were isolated from the jejunum and ileum as described by Ferrer et al. (18) with some modifications. Pieces of the jejunum and ileum were placed in a medium containing (in mM) 80 NaCl, 3 K2HPO4, 20 Tris·HCl, 37 mannitol, 0.1 EGTA, 27 trisodium citrate, and 1 mg/ml BSA at pH 7.4. To ensure that cells were obtained from the whole villus including the crypts, the tissue pieces were shaken for 90 min and filtered through nylon gauze (50-μm pore size). The filtered suspension was centrifuged three times at 800 g at 4°C. The enterocytes obtained from both segments of the small intestine were used to extract total RNA.
Preparation of BBMVs.
The mucosa was scraped from the underlying layer in both jejunum and ileum with a glass slide and used for BBMV preparation by a MgCl2 precipitation method, as previously described in detail (51). The final pellets containing purified BBMVs were resuspended in a medium comprised of 300 mM mannitol, 0.1 mM MgSO4, 0.02% (wt/vol) NaN3, and 20 mM HEPES/Tris, pH 7.4, and homogenized with a 25- and 29-gauge needle. The vesicles were immediately frozen and stored in liquid nitrogen until use (for a period no longer than 7 days). To obtain suitable amounts of BBMVs for kinetic experiments, the mucosa (jejunum or ileum) from two rats was pooled for each BBMV preparation. Moreover, BBMV preparation from WKY and SHR mucosa was always run in parallel on the same day.
Protein and enzyme activity determinations.
Protein determination was carried out by the method of Bradford (6) using a bovine gamma globulin as a standard. Membrane preparations from the jejunum and ileum (n = 5–6 preparations in each group) were routinely evaluated by measuring the specific activities of marker enzymes at 37°C. Sucrase (α-d-glucohydrolase, EC 188.8.131.52), as the marker enzyme for BBMVs, was estimated by the method of Dahlqvist (14), and the ouabain-sensitive Na+-K+-activated ATPase (Na+-K+-ATPase, EC 184.108.40.206), was measured as a basolateral membrane marker enzyme according to Colas and Maroux (13).
The uptake of d-glucose into BBMVs was measured by a rapid filtration technique (51). Briefly, 5 to 10 μl of BBMVs (equivalent to 100–200 μg of protein) were combined with 100 μl of incubation media containing 0.1 mM MgSO4, 0.02% (wt/vol) NaN3, 20 mM HEPES/Tris, pH 7.4, and 100 mM of either NaSCN or KSCN. d-glucose concentrations were 0.01, 0.05, 0.1, 0.5, 1, 5, 10, 20, 50, and 100 mM, and the osmolarity was kept constant (320 mosmol/kg) by adjusting the total sugar concentration with the required amounts of mannitol. BBMVs (n = 4 preparations in each group) were incubated for 5 s at 37°C, and detection of sugar uptake was performed by liquid scintillation adding traces of [d-14C]glucose (specific activity = 306 mCi/mmol) into the incubation media.
SDS-PAGE and Western blot analysis.
Similar amounts of protein (30 μg) of BBMVs from SHR and WKY rats (n = 5 BBMV preparations in each group) were solubilized in Laemmli sample buffer and resolved by 8% SDS-PAGE (35). Proteins were electrotransferred onto nitrocellulose membranes and immunoblotted as previously described (1) by using antisera raised in rabbits against a synthesized peptide corresponding to residues 564–575 of the deduced amino acid sequence of rabbit intestinal SGLT1 (25), diluted at 1:5,000. In experiments carried out in parallel, nitrocellulose membranes were incubated with the same antibody previously adsorbed with the antigenic peptide (1 mg/ml). The anti-SGLT1 antibody was detected by the enhanced chemiluminiscence (ECL) method according to the supplier's protocol and using a peroxidase-conjugated anti-rabbit IgG as a secondary antibody (1:3,000 dilution). To reject the possibility of having altered results due to manipulation, SGLT1 antibody was stripped off the membranes by washing with PBS-Tween 20 for 30 min at room temperature. Membranes were later incubated with a mouse anti-actin monoclonal antibody following the same protocol.
Deglycosylation and dephosphorylation assays.
N-linked deglycosylation of SGLT1 was achieved by treatment with peptide-N4-(N-acetyl-β-glucosaminyl) asparagines amidase-F from Flavobacterium meningosepticum (PNGase-F) in situ in BBMVs. Previously, BBMVs (equivalent to 50 μg protein) were solubilized in SDS in 50 mM NaCl, 5 mM EDTA, 10% (vol/vol) β-mercaptoethanol, 20 mM Tris·HCl, pH 7.5. The potential inhibitory effect of SDS on the activity of the enzyme was neutralized by sequestering free SDS in an excess of Nonidet P-40. Finally, 2,500 units of PNGase-F were added to the mixture in a buffer containing 50% glycerol, 50 mM NaCl, 5 mM EDTA and 20 mM Tris·HCl, pH 7.5. Incubation time was 1 h at 37°C. Deglycosylated samples were solubilized in Laemmli sample buffer and separated by SDS-PAGE on 8% minigels as described in SDS-PAGE and Western blot analysis.
To perform the dephosphorylation of the transporter, BBMVs equivalent to 50 μg protein were treated with 2,000 units of lambda protein phosphatase (λ-PPase) in a buffer containing 5 mM DTT, 0.1 mM EDTA, 2 mM MnCl2, 0.01% (vol/vol) polyoxyethyleneglycol dodecyl ether (Brij 35), 50 mM Tris·HCl, pH 7.5 for 1 h at 30°C. Dephosphorylated samples were solubilized in Laemmli sample buffer and separated by SDS-PAGE on 8% minigels as described above.
Sequential removal of N- and O-linked glycosylation followed by dephosphorylation was assayed as follows: 2.5 mU of O-glycopeptide endo-d-galactosyl-N-acetil-α-galactosamino hydrolase from Diplococcus pneumoniae (O-glycosidase) and 50 mU of neuraminidase from Arthrobacter ureafaciens were added in the buffer also containing PNGase-F. Deglycosylation and dephosphorylation treatment followed by Western blot analysis were then carried out in the same conditions as above.
RNA extraction and Northern blot assays.
Whole RNA was extracted from isolated enterocytes (n = 5 animals per group) as described by Chomczynski and Sacchi (12). RNA was obtained after phenol extraction and alcoholic precipitation and measured by spectrophotometric analysis at 260 and 280 nm to evaluate the purity and concentration of RNA. Samples were electrophoresed in a formaldehyde-agarose gel (15 μg total RNA/lane), stained with ethidium bromide to verify RNA integrity and equivalence of loading and transferred to a nylon membrane (NytranN, 0.45 μm; Schleicher and Schuell, Dassel, Germany). Membranes were prehybridized (1 h, 68°C) and hybridized (1 h, 68°C) using a commercial solution (ExpressHyb hybridization solution; Clontech, Palo Alto, CA). Blots were washed with 2 × SSC/0.05% SDS at room temperature (5 × 10 min) and 0.1 × SSC/0.1% SDS at 50°C (3 × 20 min). Specific RNA was detected using a 3.1-kb EcoRI fragment from pMJC424 plasmid encoding a rabbit jejunal SGLT1 (provided by Dr. E. M. Wright). Probes were labeled with [α-32P]dCTP by random priming (Random primer DNA labeling mix; Biological Industries, Kibbutz, Israel). Membranes were stripped off and rehybridized with 1) a plasmid encoding for the 18S ribosomal protein and 2) a 1.1 kb PstI fragment encoding a chicken GAPDH. Autoradiographies were carried out at −80°C and autoradiograms were quantified by scanning densitometry.
RT was carried out in a final volume of 33 μl. Ready-To-Go You-Prime First-Strand Beads (Amersham Biosciences, Madrid, Spain) were used according to the supplier's protocol and using 5 μg of primer random p(dN)6 (Roche, Madrid, Spain) and 2 μg of whole RNA as a template. After RT, cDNA was purified using a commercial kit (GFX DNA purification kit, Amersham Biosciences). PCR was performed in a final volume of 50 μl in the presence of 25–150 ng cDNA as a template, 1.5 units Taq DNA polymerase in the supplier's buffer (Amersham Biosciences), 1.5 mM MgCl2, 30 μM dNTPs, and 0.15 μM each of the following sets of forward (f) and reverse (r) primers: SG624f: 5′-CCAGTCCTATGAGCGCATTCGCAAT-3′; SG624r: 5′-TCTCGGAAAGCAAACCCAGTCAGGA-3′ (position 220 and 843); SG781f: 5′-TCCTGACTGGGTTTGCTTTCCGAGA-3′; SG781r: 5′-CAGAAAATAGCCAGCAGGAAGACGG-3′ (positions 819 and 1599); SG1348f: 5′-TCCTGACTGGGTTTGCTTTCCGAGA-3′; SG1348r: 5′-CAGGCAAAATAGGCGTGGCAGAAGA-3′ (positions 819 and 2166). After an initial denaturation at 94°C for 4 min, 35 cycles of denaturation (94°C for 40 s), annealing (56°C for 40 s) and elongation (72°C for 50 s) were performed, followed by a final elongation period of 5 min at 72°C. The PCR products were then separated on a 1.5% agarose gel and photographed under UV illumination.
Unless otherwise specified, unlabeled reagents were obtained from Sigma (Madrid, Spain). Reagents used to determine sucrase activity, and O-glycosidase and neuraminidase were from Roche (Madrid, Spain); those used in Western blot analysis were from Bio-Rad (Barcelona, Spain), and PNGase-F and λ-PPase were purchased from New England Biolabs. d-[U-14C]glucose and the ECL reagent were from Amersham Biosciences, and the membrane filters were obtained from Millipore (Barcelona, Spain).
Total d-glucose fluxes from four independent experiments were analyzed by nonlinear regression, using the Enzfitter program (Biosoft, Cambridge, UK). Because the errors associated with experimental fluxes were roughly proportional to their values, it was considered appropriate to apply a proportional weighting to the data.
Results are expressed as means ± SE. SHR and WKY rat groups were compared by the unpaired, two-tailed Student's t-test, and differences were considered significant at P < 0.05.
Body weight, systolic blood pressure, and serum determinations.
SHR systolic blood pressure at the age of 14 wk was significantly higher than that of WKY, but there were no significant differences in body weight values between both strains. Serum glucose and insulin levels were not modified between SHR and WKY rats. On the contrary, aldosterone levels were significantly enhanced en SHR when compared with WKY rats (Table 1).
Characterization of BBMVs.
The activity of the BBMV marker sucrase was decreased in the jejunum and ileum of hypertensive rats compared with normotensive rats (Table 2). Enrichments and overall recoveries in the sucrase activity were high enough in jejunal and ileal segments from both rat groups. As for the marker of basolateral membranes, Na+-K+-ATPase, the specific activity was not affected by hypertension; no significant differences in either jejunum and ileum were found in enrichment and recovery values between SHR and WKY rats, which were low enough to ensure little contamination of BBMV preparations.
d-glucose transport across BBMVs.
Before kinetic studies, the transport capability of jejunal and ileal BBMVs was assessed by measuring the time course of d-glucose uptake (0.1 mM) in the presence and absence of a Na+ gradient across the vesicles. In the first case (Na+o > Na+i), there was a transient accumulation of d-glucose, with an overshoot at 5 s in both groups of animals, which disappeared in the absence of Na+ (i.e., when the Na+ gradient was replaced by a K+ gradient). In normotensive rats, the magnitude of the accumulation ratio for Na+-dependent d-glucose transport was 10- to 12-fold in both intestinal segments, this parameter being significantly lower in hypertensive rats (7-fold and 5-fold in jejunum and ileum, respectively). In addition, the intravesicular volume, estimated from d-glucose distribution at the equilibrium (incubation time = 30 min) was similar in SHR and WKY rats (in μl/mg protein: 0.42 ± 0.06 vs. 0.44 ± 0.05 in the jejunum and 0.37 ± 0.04 vs. 0.38 ± 0.05 in the ileum for SHR and WKY rats, respectively; n = 4).
The total fluxes for d-glucose initial uptakes in the range of 0.01–100 mM are shown in Fig. 1. Curve fitting was best achieved when considering a kinetic model in which the total uptakes were broken down into one saturable, Michaelian component plus a linear, nonsaturable one. The calculated kinetic constants are shown in Table 3. A significant decrease in the Vmax was observed in the jejunum (52 ± 2%) and ileum (60 ± 5%) of hypertensive rats compared with normotensive ones. However, no differences were noted in the Km and Kd values between the two groups of animals in any intestinal segment.
A single band of ∼72 kDa was recognized by the anti-SGLT1 antibody in the jejunum and ileum of WKY rats. However, in SHR we observed a decrease in the molecular mass of the band, which appeared as a 67 kDa protein. These bands were blocked by preabsorption with the antigenic peptide (Fig. 2A). Blots were normalized with an anti-actin antibody to assure equivalence of loading and absence of artifacts due to manipulation. The anti-actin antibody recognized a single band of 45 kDa with no significant abundance differences between samples (Fig. 2B). The relative abundance of SGLT1 was reduced in the jejunum (48%) and ileum (62%) of SHR with respect to WKY rats (Fig. 2C).
To know whether the changes observed in the molecular mass of SGLT1 were due to a loss of glycosylation or phosphorylation of the protein in the intestine of hypertensive rats, BBMVs were incubated with either N-glycosidase F (PNGase-F) or lambda protein phosphatase (λ-PPase) enzymes. The treatment with PNGase-F reduced the apparent molecular size of the specific immunoreactive band of SGLT1 by 5 and 7 kDa in the jejunum of WKY and SHR, respectively, thus leading to molecular mass values of 67 and 60 kDa (Fig. 3A). After the treatment with λ-PPase, the size of the SGLT1 decreased by 2 and 3 kDa, respectively, in the jejunum of WKY and SHR, leading to 70- and 64-kDa bands (Fig. 3B). Finally, sequential deglycosylation (including potential O-linked glycosylation) and dephosphorylation of the transporter resulted in the detection of 64- and 57-kDa bands in BBMVs from the jejunum of WKY and SHR, respectively. Similar results were obtained in the ileum (not shown).
Northern blot analysis.
Purity of the total RNA solution was assessed by measuring absorbance at 260 and 280 nm. In all cases, the ratio of 260:280 was higher than 1.8, indicating a high purity and low contamination by protein fractions. Hybridization with a specific SGLT1 probe showed a single band of 3.9 kb in both intestinal segments and WKY and SHR rats (Fig. 4A). Each well contained equivalent amounts of total RNA, assessed by rehybridizing the same blot with ribosomal 18S RNA (Fig. 4B) and with a GAPDH probe (Fig. 4C). There was a decrease of 51 and 61% in the jejunum and ileum, respectively, in the abundance of the SGLT1 mRNA levels from SHR group compared with WKY group (Fig. 4D).
As expected from the rat SGLT1 cDNA sequence, the specific pairs of primers used for RT-PCR gave final products of 624, 781, and 1,348 bp using both WKY and SHR total RNA templates isolated from jejunal enterocytes (Fig. 5). \. Therefore, no size differences appeared when comparing parallel fragments of the mRNA encoding for SGLT1 transporter from both groups of rats. On the other hand, the density of PCR products seemed to indicate once again a lower expression of the transporter in SHR jejunum compared with WKY rats. Similar results were obtained when using RNA templates from ileal enterocytes (not shown).
Intestinal glucose absorption is mainly mediated by SGLT1 protein that is a sodium-dependent cotransporter located in the apical membrane of enterocytes. Before the present study, we had reported a decrease in the sodium-dependent glucose transport in the jejunum and ileum of SHR when compared with normotensive rats (43, 51, 52). In this work, we have extended this finding by analyzing both the kinetic properties and the regulation of the transporter SGLT1. We have demonstrated that the maximum capacity (Vmax) for SGLT1 is diminished in BBMVs prepared from jejunum and ileum of SHR and that the reduced activity of SGLT1 in hypertensive rats was associated with a lower expression of SGLT1 protein and mRNA. However, these findings were not accompanied by changes in the serum glucose and insulin levels, suggesting that the homeostasis of glucose and insulin are maintained in these hypertensive rats when compared with WKY rats.
The Vmax for SGLT1-mediated glucose transport was 52 and 60% lower, respectively, in jejunal and ileal BBMVs prepared from SHR compared with those from WKY rats. Moreover, Vmax values were higher in the jejunum than in the ileum in both WKY and SHR, according to previous work in different animal species (19, 20, 49). On the other hand, the affinity of the transporter (given by Km values) was similar in both groups and within the range of previous reports (8, 11, 38).
To check whether the reduction of Vmax observed in SHR was due to alterations in the transporter density in the apical membranes, we measured the relative amount of SGLT1 in jejunal and ileal BBMVs by Western blot analysis. A single band of 72 kDa was detected in BBMVs from WKY rats, which matches with the molecular mass described in previous reports for rat intestinal SGLT1 (27, 29). Surprisingly, SGLT1 was expressed as a 67-kDa protein in both jejunal and ileal SHR samples. Regardless of this observation, the abundance of the transporter in hypertensive rats was 48 and 62% lower in the jejunum and ileum, respectively, i.e., a reduction equivalent to that observed in Vmax values. A parallel immunoblotting of actin as a control protein revealed no differences between WKY and SHR; therefore, the impaired d-glucose absorption observed in the latter would be due, in part, to a lower number of SGLT1 molecules at their apical membrane.
The abundance of SGLT1 was further studied at mRNA level, to clarify the regulation of the expression of this transporter in the absorptive intestinal cells of hypertensive rats. A single transcript of 3.9 kb was obtained in jejunal and ileal enterocytes from both WKY and SHR, a similar transcript size to that obtained by several authors in the rat intestine (7, 36) but slightly lower than the 4.5-kb transcript reported by others (34, 37, 44). After normalization of blots with 18S rRNA and GAPDH probes, SGLT1 mRNA levels were once again reduced by 51 and 61% in the jejunum and ileum, respectively, of SHR. Therefore, our results show similar variations of mRNA and protein for SGLT1, suggesting that transcriptional modulations are later reflected in the alterations in protein tissue content of the transporter in hypertensive rats.
Our data are in agreement with previous reports, but they also differ from some others. We have previously observed a decrease in sugar transport in the kidney of hypertensive rats, which was also related to changes in the protein content (41, 42). A decrease of sugar transport has been also reported in renal BBMVs from SHR compared with control WKY rats by Morduchowicz et al. (46). In contrast, other hypertensive rats models, such as renovascular hypertensive rats (3) and high-salt loaded Dahl salt-sensitive rats (32) showed an increase in the renal sodium-dependent d-glucose transport. However, all of these studies have been done in the kidney, due to its role in blood pressure regulation, and, to our knowledge this is the first report concerning the regulation of SGLT1 in small intestine from SHR.
It still remains unclear whether the observed alterations in the intestinal sugar transport are of relevance in the pathophysiology of hypertension or constitute rather another consequence of the disease. In this regard, the renin-ANG-aldosterone system, which is considered crucial in the pathogenesis of hypertension (4, 10), has been reported to modulate sugar transport through ANG II actions (3, 33, 55). The increased activity and expression of renal SGLT2 found in renovascular, hypertensive rats by Bautista el al. (3) disappeared after either inhibition of ANG II synthesis or ANG II type 1 receptor (AT1) blockade. In addition, aldosterone regulates the expression of SGLT1 in the chicken intestine under low-sodium intake posttranscriptionally (2, 22). In the present study, although we have not measured ANG levels, we have found an increase in the aldosterone levels in SHR when compared with WKY rats, which could be related with the observed changes in SGLT1 function and expression in the intestine of these rats. Therefore, the study of the effect of AT1 blockade and/or ANG-converting enzyme (ACE) inhibitors and aldosterone antagonist on intestinal d-glucose uptake in SHR could be particularly interesting for this purpose. In addition, since the present study has been carried out in rats with an established hypertension status, it would be interesting to study whether changes in SGLT1 follow a pattern related to the development of the disease.
On the other hand, as both oxidative stress (23) and cytokines (50) gain relevance in the development of hypertension in SHR, and since the activity and expression of SGLTs can be influenced by both factors (21, 24), further investigation is also needed to check whether intestinal sugar transport is modulated by a local and/or systemic high production of reactive oxygen species and/or cytokines in hypertensive animals.
Besides the alterations in SGLT1 function/expression in SHR small intestine, Western blot analysis experiments revealed an unexpected 5-kDa difference in the molecular mass of this transporter. We have previously observed a decrease in the activity of SGLT1 in the kidney of hypertensive rats; however, in that study there were no changes in the molecular mass of the SGLT1 protein between WKY and SHR (42), nor were there in the study of the intestinal fructose transporter, GLUT5, in the same pathology (40). Then, where could the changes in the structure of the intestinal SGLT1 in SHR come from? Since glycosylation and phosphorylation constitute two of the most common posttranscriptional modifications in membrane proteins including SGLT1 (59), a possible explanation for the 5-kDa difference among the two rat strains would be a loss of glycosylation and/or phosphorylation in the SGLT1 protein from SHR intestine compared with that present in WKY rats. However, the treatment with both glycosidase and phosphatase enzymes reduced the apparent molecular size of SGLT1 in a similar manner in WKY and SHR. Since the consensus sequence for N-glycosylation (NXS/T) is also a potential site for phosphorylation, which may result in incomplete deglycosylation or dephosphorylation of the protein when these treatments are carried out separately, we also performed a sequential treatment with both glycosidase (including enzymes that eliminate O-linked glycosylations) and phosphatase enzymes. Nevertheless, the difference in the apparent molecular size of SGLT1 between WKY and SHR remained also after this treatment. Moreover, the treatment with O-glycosidase alone produced no effect on the apparent size of the transporter, which is in agreement with previous experiments in rabbits showing that O-linked glycosylation is unlikely to be present in the intestinal SGLT1 (28).
Since posttranscriptional modifications, in terms of different glycosylation or phosphorylation levels, are not responsible for the changes in SGLT1 structure in hypertensive animals, these changes might, in part, be attributable to the presence of a slightly different SGLT1 isoform in this hypertensive strain. Although this hypothesis could probably be confirmed only by cloning and sequencing of the transporter, we made a last attempt to study a possible alternative splicing of the protein by performing RT-PCR with specific pairs of primers from the rat intestinal SGLT1 encoding fragments of ∼625–1,350 bp. However, none of the bands revealed size differences in the PCR products obtained from either SHR or WKY intestinal RNA templates, indicating that there was no alternative splicing in SHR that could account for the observed differences in the apparent molecular size of the transporter.
In conclusion, we have found a decrease in the activity of sodium-dependent d-glucose transporter in SHR, which was due to a reduction in the levels of SGLT1 protein and mRNA. Decreased sodium absorption through this transporter might participate in the observed dysregulation of sodium transport in hypertensive rats.
Grants from Ministerio de Sanidad, Fondo de Investigación Sanitaria (FIS 97/1143, 99/1142) and Generalitat de Catalunya (2001SGR0149) supported this work. A. Barfull was supported by a training grant (FPU) from Ministerio de Educación, Cultura y Deporte. L. Gómez-Amores was supported by a research grant (Contrato de Apoyo a la Investigación) from Junta de Andalucía.
The authors thank Dr. M. Kasahara and Prof. E. M. Wright for providing the SGLT1 antibody and cDNA, respectively, and Dr. J. Vitorica for expert advice on RT-PCR.
↵* A. Mate and A. Barfull contributed equally to this work.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2006 the American Physiological Society