Intestinal absorption of biotin occurs via a Na+-dependent carrier-mediated process that involves the sodium-dependent multivitamin transporter (SMVT; product of the Slc5a6 gene). The SMVT system is exclusively expressed at the apical membrane domain of the polarized intestinal epithelial cells. Whereas previous studies from our laboratory and others have characterized different physiological and biological aspects of SMVT, little is currently known about its structure-function activity relationship. Using site-directed mutagenesis approach, we examined the role of the positively charged histidine (His) residues of the human SMVT (hSMVT) in transporting the negatively charged biotin. Of the seven conserved (across species) His residues in the hSMVT polypeptide, only His115 and His254 were found to be important for the function of hSMVT as their mutation led to a significant reduction in carrier-mediated biotin uptake. This inhibition was mediated via a significant reduction in the maximal velocity (Vmax), but not the apparent Michaelis constant (Km), of the biotin uptake process and was not related to the charge of the His residue. The inhibition was also not due to changes in transcriptional or translational efficiency of the mutated hSMVT compared with wild-type carrier. However, surface biotinylation assay showed a significant reduction in the level of expression of the mutated hSMVT at the cell surface, a finding that was further confirmed by confocal imaging. Our results show important role for His115 and His254 residues in hSMVT function, which is most probably mediated via an effect on level of hSMVT expression at the cell membrane.
- biotin transport
- histidine residues
biotin, a member of the B-family of water-soluble vitamins, is essential for normal cellular functions and growth. The vitamin acts as coenzyme for a number of carboxylases in different metabolic pathways involving fatty acid biosynthesis, gluconeogenesis, and amino acid catabolism (10, 11, 17). Biotin deficiency leads to serious clinical abnormalities that include neurological disorder, growth retardation, and dermal abnormities (30). Furthermore, animal studies have shown that biotin deficiency during pregnancy can lead to embryonic growth retardation, congenital malformation, and death (5, 12, 27, 31).
Humans and other animals cannot synthesize biotin, and thus, must rely on exogenous supply of the vitamin and on normal absorption in the intestinal tract. The intestine, therefore, plays a central role in determining and maintaining normal body biotin homeostasis. The intestine is exposed to two sources of biotin: a dietary source, which is absorbed in the small intestine, and a bacterial source, which is absorbed in the large intestine (17). Studies from our laboratory and others over the past two decades on the mechanisms of intestinal absorption of biotin have used a variety of intestinal preparations to show the involvement of an efficient Na+-dependent, carrier-mediated system in biotin uptake process (reviewed in 17, 18, 21). This system involved was subsequently identified as the sodium-dependent multivitamin transporter (SMVT; product of the Slc5a6 gene) (14, 26), a transporter that can also transport the unrelated water-soluble vitamin pantothenic acid and the metabolically important substrate lipoate (14, 17, 20, 26). The human SMVT (hSMVT) system is exclusively expressed at the apical membrane domain of the polarized intestinal epithelial cells (23) and is predicted to have 12 transmembrane domains (TMD) with both the NH2- and COOH-termini being inwardly directed (14, 26). Studies from our laboratory have also demonstrated the hSMVT to be the main biotin uptake system in human intestinal epithelial cells (2) and characterized different aspects of its transcriptional and posttranscriptional regulation (4, 6, 13, 16). To date, however, there is little known about the structure-function activity relationship of the hSMVT system. Previous studies from our laboratory have used a chemical approach of amino acid/group specific reagents to show possible involvement of histidine residues in the function of the intestinal brush-border membrane (BBM) biotin uptake system (19). This chemical approach, however, could not identify which of the seven conserved histidine residues across species (human, rat, rabbit, mouse) are involved. To address these issues and to confirm and complement the previous findings with the chemical approach, we utilized in this study a molecular (site-directed mutagenesis) approach. Our results showed the involvement of His115 and His254 in the functionality of hSMVT and suggest that their role is most probably mediated via maintaining proper expression of the transporter at the cell membrane.
MATERIALS AND METHODS
[3H]Biotin (specific activity > 30 Ci/mmol, radiochemical purity > 98%) was obtained from American Radiolabeled Chemicals (ARC, St. Louis, MO). All chemicals used in the study were either of analytical or molecular biology grades and were obtained from commercial sources.
Cell culture and transient transfection.
As described previously (14, 23) human retinal pigmented epithelial cells (ARPE19) were used for uptake studies, while Madin-Darby canine kidney (MDCK) cells were used for confocal imaging. Both the cells were obtained from ATCC (Manassas, VA) and maintained in DMEM and MEM medium (Sigma, St. Louis, MO), respectively. For optimized growth, the media were supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin, and cells were grown at 37°C in presence of 5% CO2 at a humidified atmosphere. Lipofectamine 2000 (Invitrogen, Carlsbad, CA) was used for the transient transfection as mentioned in the manufacturer's protocol. For transfection, 3 μg of purified plasmid DNA was used for each well of a 12-well tissue culture plate (Corning, NY), and the uptake was carried out after 48 h of transfection.
Generation of hSMVT constructs and site-directed mutagenesis.
Full-length hSMVT was amplified from the human small intestinal SMVT cDNA previously cloned in GFP-N3 vector (23) using specific primers containing XbaI and EcoRI restriction site sequences respectively on their 5′ ends (Table 1). For PCR, following initial denaturation of 4 min at 94°C, the annealing was done at 57°C. The amplified hSMVT was cloned into pGMT-Easy vector (Promega, Madison, WI), and the construct was digested with XbaI and high-fidelity EcoRI-HF enzymes (NEB, Ipswich, MA). The products were then gel separated, and the released insert was purified by GENECLEAN II kit (MP Biomedicals, Solon, OH). The insert was then ligated into mammalian expression vector pcDNA 3.1(−) (Invitrogen), which was also gel purified after digesting with same set of restriction enzymes. The nucleotide sequence of the cloned construct was verified by DNA sequencing (Laragen, Los Angeles, CA).
Site-directed mutagenesis was performed using Quick Change II kit from Stratagen (La Jolla, CA) according to the manufacturer's protocol. Briefly, the template was amplified using specific mutant primers (Table 1) for 16 cycles, and the PCR product was incubated with DpnI (10 U/μl) to digest the parental DNA strand. The digest was then transformed into competent XL1-blue cells, and the clone harboring mutant plasmid was identified by sequencing of the isolated plasmids (Laragen). To generate the green fluorescent protein (GFP)-tagged hSMVT mutants, we used hSMVT-GFP construct generated by us previously (23) as template, and selective mutations were introduced into the hSMVT-GFP construct using specific primers (Table 1) as described above.
Real-time PCR analysis.
Total RNA was isolated from cells 48 h after transfection using TRIzol reagent (Invitrogen) following manufacturer's protocol. RNA samples were then treated with DNAse I (Invitrogen), and the DNAse-I-treated RNA sample was reverse transcribed using iScript cDNA synthesis kit (Bio-Rad). The expression level was quantified in CFX 96 real-time PCR system (Bio-Rad) using gene-specific primers for hSMVT (Table 1) using the conditions as described previously (16). β-Actin was used as an internal control for normalization, and data were calculated using relative relationship method provided by vendor (Bio-Rad) (9).
Uptake measurements were done 48 h posttransfection using Krebs-Ringer (KR) buffer (in mM: 133 NaCl, 4.93 KCl, 1.23 MgSO4, 0.85 CaCl2, 5 glucose, 5 glutamine, 10 HEPES, and 10 MES, pH 7.4) as described by us previously (16). Briefly, cells were incubated with radiolabeled biotin (9 nM) in KR buffer for 10 min (initial rate of uptake; data not shown) at 37°C. The reaction was stopped by adding 2 ml of ice-cold KR buffer followed by immediate aspiration. Cells were then rinsed twice with ice-cold buffer and lysed with 1 ml of 1 N NaOH, neutralized using 10 N HCl, and radioactivity was measured in a scintillation counter (Beckman Coulter LS6500). The endogenous biotin uptake was measured by using ARPE19 cell transiently transfected with empty pcDNA 3.1(−) vector. Carrier-mediated biotin uptake was determined by subtracting passive diffusion component form the total uptake. Passive diffusion was quantified by performing uptake in the presence of excess unlabeled biotin (1 mM). Protein concentrations were measured by Dc protein assay kit (Bio-Rad) (uptake was expressed by fmol·mg protein−1·unit time−1).
Western blot analysis.
To determine the level of expression of hSMVT, cell lysate was prepared by incubating the cells with 20 mM Tris-base, 150 mM NaCl, 1% Triton X-100, 0.1% SDS, and 1 mM EDTA buffer. The protein was dissolved in sample buffer and loaded into NuPAGE 4–12% bis-Tris gradient minigels (Invitrogen). After electrophoresis, the gel was electro blotted onto nitrocellulose membrane (Amersham Biosciences, Piscataway, NJ) and blocked overnight with 5% nonfat dried milk (Bio-Rad) in Tris buffer saline (pH 7.5) supplemented with 0.1% Tween 20 (TBST). The blot was then probed with specific anti-human polyclonal hSMVT antibodies (Santa Cruz Biotechnology) for 90 min in TBST buffer. The blot was washed twice in the same buffer and then incubated with goat anti-rabbit IgG conjugated with horseradish peroxidase (1:2,500 dilutions in TBST) for 1 h at room temperature. After being washed for three times for 10 min each, the bands were developed by an enhanced chemiluminescence kit (Amersham Biosciences). Level of human β-actin (used for normalization) was determined using anti-human β-actin polyclonal antibodies (Santa Cruz biotechnology). Density of the individual band was determined by UN-SCAN-IT gel software (version 6.1) from the scanned image.
Cell surface biotinylation assay.
Cell surface proteins were biotinylated as described by us and others before (1, 32). Briefly, after 48 h of transfection ARPE19 cells were incubated with sulfo-NHS-SS-biotin (Pierce Biotechnology, Rockford, IL) in biotinylation buffer (PBS, pH 7.5, supplemented with 10 mM triethanolamine, 2 mM CaCl2, and 150 mM NaCl). Followed by 90 min of incubation on an orbital shaker at 4°C, the reaction was quenched by adding quenching buffer (PBS, pH 7.5, supplemented with 2 mM CaCl2, 1 mM MgCl2, and 100 mM glycine), and the cells were then lysed with lysis buffer (in mM; 150 NaCl, 50 Tris·HCl, 5 EDTA, and 1% Triton X-100, pH 7.4).
Proteins bound to the beads were extracted in NuPAGE LDS sample loading dye (Invitrogen) by heating and resolved in NuPAGE 4–12% bis-Tris gradient minigels (Invitrogen) for Western blot analysis. hSMVT was identified using specific anti-hSMVT antibodies followed by horseradish peroxidase-tagged secondary antibody as described above. The amount of biotinylated hSMVT was quantified from the Western blot and normalized with expression of total amount of cellular hSMVT.
For confocal imaging, MDCK cells were grown on glass-bottomed petri dishes (MatTek, Ashland, MA) and imaged by a Nikon C1 confocal scanner head attached to a Nikon inverted phase-contrast microscope. Fluorophores were excited using the 488-nm wavelength from an argon ion laser, and emitted fluorescence was measured at 530 nm with 20-nm band-pass filter.
Comparative protein structure modeling.
We subjected the hSMVT polypeptide sequence to PSIPRED fold recognition program (3) and obtained possible templates for the most likely three-dimensional model for hSMVT. To identify the most suitable template, we generated several comparative models by Multiple Mapping Method (15). The quality of each model was checked by PROSA energy function (29), which compares the score of energy value of the input structure to the experimentally determined native proteins of similar size. The analysis suggests the nucleo base-cation-symport-1 family transporter (PDB ID: 2JLN) as the most suitable template (28), although it shares 11.4% sequence identity with hSMVT. We also determined the electrostatic potential of the model within the membrane environment by PBEQ solver program (8) (which uses the Poisson-Boltzmann equation), and the charge distribution was visualized in pyMol software (version 1.3) (http://www.pymol.org/). Results of the latter indirectly validated the model (see results).
Transport data are the means ± SE of multiple uptake determinations expressed in terms of femtamoles per milligram protein per 10 min. Statistical analysis was determined by Student's t-test, and significance level was set at P < 0.05. Carrier-mediated uptake was determined by subtracting the diffusion component from total uptake. For kinetic studies the diffusion component at each concentration was determined by multiplying the slope of the uptake line drawn between a high pharmacological concentration of biotin (1 mM) and the point of origin by the individual concentration. The apparent Michaelis constant (Km) and maximal velocity (Vmax) were determined by fitting the data to Michaelis-Menten equation using nonlinear regression in Graph Pad Prism software (version 5.03). All uptake studies, Western blot analysis, confocal imaging, and biotinylation assays were performed on three independent occasions.
The amino acid sequence of SMVT shows that it has nine His residues, seven of which are conserved across species (human, mouse, rat, and rabbit). The conserved His residues of the hSMVT polypeptide are located at positions 46, 54, 115, 125, 238, 254, and 533 (Fig. 1). Hydropathy plot (Kyte Dolittle algorithm) suggests that the hSMVT protein has 12 TMD and that both the NH2- and COOH-termini are oriented toward the cytoplasm (26). According to this prediction, the conserved His residues are located as follow: His46 and His533 residues are located at the membrane-aqueous interface, His54 and His125 residues are located in extracellular domains, His238 and His254 residues are in the large intracellular domain between sixth and seventh TMD, and His115 is located on third TMD.
Effect of mutating the conserved His residues on hSMVT function.
In this study, we mutated each of the conserved His residues individually (to the neutral alanine) and examined the effect of this conserved mutation on the functionality of the hSMVT. ARPE19 cells were transfected with mutant constructs, and biotin uptake was measured as described in materials and methods. The results showed that while mutating His residues located at position 46, 54, 125, 238, and 533 of the hSMVT had no effect on initial rate of carrier-mediated biotin (9 nM) uptake, mutating His115 and His254 lead to a significant (P < 0.01 for H115A and P < 0.05 for H254A) inhibition in the vitamin uptake (Fig. 2). To determine whether the inhibition in biotin uptake is due to the replacement of the His residue itself or due to the positive charge of the His residue, we mutated His115 and His254 to both negatively and a positively charged amino acids (aspartic acid and arginine, respectively) then examined the effect of these mutations on the initial rate of carrier-mediated biotin uptake in transiently transfected ARPE19 cells. The results showed comparable inhibition in biotin uptake in cells expressing hSMVT in which the His115 and His254 were mutated to alanine, aspartic acid, or arginine, indicating that the inhibition is charge independent but rather is due to the replacement of the His residue itself (Fig. 3).
Effect of mutating His115 and His254 on kinetic parameters of biotin transport via the hSMVT system.
Effect of mutating His115 and His254 residues of hSMVT on kinetic parameters of biotin uptake, i.e., the apparent Km and Vmax of the induced uptake process, was determined by examining the initial rate of biotin uptake as a function of concentration in ARPE19 cells expressing the His115 and His254 mutants (Fig. 4). Data were compared with those obtained with wild-type hSMVT. The results showed that mutating His115 and His254 had no effect on the apparent Km of the induced carrier-mediated biotin uptake process (17.30 ± 5.2, 19.54 ± 2.4, and 18.09 ± 2.9 μM for wild-type hSMVT and for His115- and His254-mutated hSMVT, respectively). On the other hand, a marked inhibition in the Vmax of the induced carrier-mediated biotin uptake was observed in hSMVT with mutations in His115 and His254 (1017.00 ± 106.6, 106.80 ± 4.7, and 95.01 ± 5.4 fmol·mg protein−1·10 min−1 for wild-type hSMVT, and that with His115 and His254 mutations, respectively). These findings suggest that the inhibition in hSMVT-mediated biotin uptake observed upon mutating His115 and His254 is due to an effect on the number (and/or activity) of the hSMVT carriers at the cell membrane with no effect on their affinity.
Effect of mutating His115 and His254 on the level of expression of hSMVT in mRNA and protein level.
To determine whether mutating His115 and His254 of the hSMVT polypeptide affect the steady-state mRNA level of the transporter, we performed quantitative PCR using total RNA isolated from ARPE19 cells transfected with the same amount (3 μg) of His115, His254, and wild-type hSMVT cDNA. The results showed that the steady-state mRNA level of the mutants to be similar to that of the wild-type hSMVT (Fig. 5A). In another study, we examined the effect of mutating His115 and His254 of the hSMVT polypeptide on the level of expression of the transporter protein. Expression was monitored by mean of Western blot analysis using specific polyclonal anti-hSMVT antibodies. On normalization with β-actin as described in materials and methods we found similar level of expression of total hSMVT protein in cells transfected with wild-type hSMVT construct and H115A and H254A mutants (Fig. 5B). Inset shows a representative Western blot image. The above-described findings suggest that mutating His115 and His254 of the hSMVT does not affect the transcriptional or translational efficiency of the carrier protein.
Effect of His115 and His254 mutations on cell surface expression of hSMVT.
To determine whether mutating His115 and His254 affect the level of expression of the hSMVT protein at the cell membrane (as suggested by the above data), we performed biotinylation assay followed by Western blot analysis on ARPE19 cells transfected with equal amounts (15 μg) of these mutants and compared the level of membrane expression with cells transfected with equal amount of wild-type hSMVT. The results showed a significant (P < 0.01) reduction in the level of cell surface expression of the His115 and His254 mutants compared with the wild-type hSMVT (Fig. 6A).
In other studies we performed live cell confocal imaging using constructs of wild-type and mutated hSMVT fused to GFP to determine cell surface expression of the proteins. Equal amounts of these constructs (3 μg) were transfected into MDCK cells, and cells were imaged 48 h posttransfection as described by us previously (23). The results showed that both of the His115 and His254 mutants as well as the wild-type hSMVT to be expressed at the apical membrane domain of MDCK cells (Fig. 6B); however, the level of expression in both the mutants was markedly lower than that of wild-type hSMVT.
Homology model for hSMVT and possible interaction between His254 and Thr537 in mediating the inhibition in biotin uptake.
We generated a homology model for hSMVT using the multiple mapping method as described in materials and methods. The charge distribution along the surface of the model indirectly validated the model as it predicted the TMD to be mostly hydrophobic, whereas the predicted extracellular and intracellular loops were mostly hydrophilic (charged) (Fig. 7A). The model shows that the His254 residue that was predicted to lie in the cytoplasmic sequence between the sixth and seventh TMD (26) is located in close proximity (hydrogen bond distance) to the conserved cytoplasmic Thr537 residue of the hSMVT polypeptide (Fig. 7B). Thus we examined whether Thr537 is important for the function of His254 and thus mutated this site (to alanine) and examined the effect of this mutation on functionality of the hSMVT system (Fig. 7C). The results showed that mutating Thr537 to have no effect on carrier-mediated biotin uptake.
The aim of this study was to investigate the structure-function relationship of the hSMVT system focusing on the role of the conserved (and positively charged) His residues in transporting the negatively charged biotin molecule (pKa of biotin = 4.5). Previous studies from our laboratory have shown that pretreatment of intestinal BBM vesicle (BBMV) with the histidine-modifying reagent diethyl pyrocarbonate (DEPC) to lead to a significant inhibition in carrier-mediated biotin uptake (19). That finding suggests a possible role for the His residues in the function of the involved system, which was subsequently identified as SMVT (14, 18, 21, 26). Primary sequence of the hSMVT protein shows that it has nine His residues of which seven are conserved among species, but it is not known as to which of these His residues is/are affected and how this chemical modification(s) affects the functionality of the hSMVT system. We have addressed these issues in the present study using refined molecular approach complemented with appropriate physiological and biological assays.
Results of our study showed that of the seven conserved His residues of the hSMVT polypeptide, only His115 and His254 appear to be important for the functionality of hSMVT. Mutating these two sites led to a significant inhibition in carrier-mediated biotin uptake by ARPE19 cells expressing these mutants compared with wild-type hSMVT. The His115 residue is predicted (by hydropathy analysis) to be located in a hydrophobic environment (cell membrane), while the His254 residue is predicted to be located in the cytosolic environment (it is located in the large intracellular loop between the sixth and seventh TMD). The inhibition in transport function of hSMVT as a result of mutating His115 and His254 to alanine was not due to change in charge, polarity, and size but rather due to loss of His residue itself. This conclusion is based on the observations that replacing His115 and His254 residues with the negatively charged aspartic acid (Asp) or with the positively charged arginine (Arg) failed to alter the level of inhibition in biotin uptake caused by mutating these His residues to the neutral alanine.
The inhibition in biotin uptake by hSMVT brought about by mutating His115 and His254 was found to be mediated via a significant decrease in the Vmax, but not the apparent Km, of the uptake process. These findings suggest that mutating His115 and His254 affects the number (and/or activity) of the hSMVT systems with no change in their affinity. This observation is in line with what we have observed previously using the chemical approach of histidine-specific reagent (19). Involvement of a residue predicted to be located in an extra membrane (cytoplasmic) sequence of a membrane transporter in the function of that transporter is not unusual phenomenon as similar finding was recently described for the His residue located at position 247 of the proton-coupled folate carrier (25).
The decrease in the activity of hSMVT His115 and His254 mutants compared with wild-type hSMVT was not due to a decrease in the transcription or translational efficiency of the two mutants compared with wild-type hSMVT. This is because quantative RT-PCR and Western blot analysis (on total cell lysate) showed similar level of mRNA and total cellular hSMVT protein level between the mutants and the wild-type hSMVT. However, when surface biotinylation assay was performed, a significant reduction in cell surface expression of hSMVT was observed in the case of the two His mutants compared with the wild-type protein. In the confocal imaging studies, we found that although both the mutants were being expressed on the cell surface, their level of expression is markedly lower than that of wild-type hSMVT confirming our finding with the biotinylation assay. These findings provide direct support for the conclusion drawn from the kinetic analysis data that the inhibition in biotin uptake as a result of mutating His115 and His254 is due to a decrease in the number of hSMVT at the cell surface. How mutating His115 and His254 ends up affecting membrane expression of hSMVT is not clear but may be due to changes in protein folding that then affect its interaction with an accessory and/or motor protein(s). Further studies are needed to clarify this issue.
The three-dimensional structure of hSMVT is not known. However, by using fold recognition program and PROSA energy function calculation as described previously (25), we found the three-dimensional structure of a nucleobase-cation symport-1 family transporter (Mhp1; PDB ID: 2JLN) can be a suitable template for hSMVT. Mhp1 is a cation symporter (22, 28), which has structural similarity with members of the solute-sodium-symporter family (7, 24, 28). Experimentally resolved crystal structure of the Mhp1 also shows that it has 12 TMD with both the NH2- and COOH-termini oriented in the same side relative to the plan of the membrane as is the case with hSMVT (26). According to this comparative model, the hSMVT polypeptide makes several hydrophobic stretches, and electrostatic calculations showed that these regions are largely solvent inaccessible. On the other hand, the extra-membrane hydrophilic regions of hSMVT polypeptide was solvent accessible according to this model. Since the His254 of the hSMVT protein is predicted to be exposed to the cytoplasm (26), we searched for a polar residue(s) in the hSMVT polypeptide that is located in close proximity to (and may interact with) this His residue. We have identified Thr536 and Thr537 as possible interacting partners. We focused on Thr537 because, unlike Thr536, it is oriented toward the His254 residue and within a hydrogen bond distance. Mutating Thr537 residue to the neutral Ala, however, did not affect hSMVT function, suggesting that even if an interaction exists between His254 and Thr537, this interaction is not functionally significant.
In summary, our study demonstrates the involvement of His115 and His254 residues in the function of the hSMVT system and that this is mostly due to an effect on membrane expression of the transporter.
This study is supported by grants from the Department of Veterans Affairs and the National Institutes of Health (DK-58057 and DK-56061).
No conflicts of interest, financial or otherwise, are declared by the author(s).
Authors are thankful to V. S. Subramanian for assisting in the confocal imaging and critically reviewing the manuscript.