Renal Tubular Reabsorption of Folate Mediated by Folate Binding Protein 1
Henrik Birn*,
Ofer Spiegelstein,
Erik I. Christensen* and
Richard H. Finnell
* Department of Cell Biology, University of Aarhus, Aarhus, Denmark; and Institute of Biosciences and Technology, Texas A&M University System Health Science Center, Houston, Texas
Address correspondence to: Dr. Henrik Birn, Department of Cell Biology, Institute of Anatomy, University of Aarhus, University Park, Building 234, Aarhus C, Denmark, DK-8000. Phone: +45-8942-3051; Fax: +45-8619-8664; E-mail: hb{at}ana.au.dk
Renal tubular reabsorption of filtered folate is essential forthe conservation and normal homeostasis of this important vitamin.Different molecular mechanisms have been implicated in epithelialfolate transport, including folate receptors. Defective expressionor antibody inactivation of these is associated with embryonicdefects also correlated with low folate intake; however, theircontribution to renal tubular folate reabsorption has not beenestablished. With the use of targeted inactivation of the folatebinding protein 1 (folbp1) and folate binding protein 2 (folbp2)genes in mice, the role of folate receptors in renal epithelialfolate reabsorption was evaluated during low and normal folateintake. Inactivation of folbp1 was associated with (1) lossof 3H-folic acid binding to crude kidney membranes, (2) increasein renal folate clearance, and (3) increase in urinary excretionand decrease in renal uptake of injected 3H-methyltetrahydrofolate.No changes in renal folate handling were observed as a resultof folbp2 inactivation. Thus, folbp1 is essential for normalrenal tubular folate reabsorption, preventing excessive urinaryfolate loss. Folbp1 is heavily expressed in choroid plexus,yolk sac, and placenta, supporting a role of folbp1 in folatetransport in other tissues. The greatest significance of folbp1for renal folate uptake was observed at conditions of low folateintake, providing a possible explanation for the ability offolate supplementation to prevent developmental defects associatedwith folbp1 inactivation.
Epithelial folate transport is essential for normal folate homeostasis,regulating intestinal uptake, renal tubular reabsorption, andtissue distribution of folate. Defects in folate homeostasishave been implicated in a number of conditions, including embryonicabnormalities, cancer, and cardiovascular disease. Several differentmechanisms have been suggested to regulate epithelial folateuptake, including folate carriers and glycosylphosphatidylinositol–linkedfolate receptors (FR) (1–5). Thus far, four FR isoformshave been characterized in humans. FR- is present predominantlyin epithelial cells (6–8), FR- is expressed at low tomoderate levels in several different tissues (8,9) and differswith respect to affinities for different forms of folate (10).FR-/' are specific for hematopoietic cells (11,12). The fourthhuman FR gene (FR-) predicts a 27.7-kD protein with a uniqueexpression pattern in both adult and embryonic tissues (13).Both FR- and FR- and the reduced folate carrier (RFC) are expressedin the kidney (8,14–17), and in vitro studies have suggesteda role for each of these transport systems in cellular folateuptake; however, their individual contribution varies betweencell lines (1,2,5). FR have been located in the proximal tubuleluminal membranes and endocytic apparatus (18,19), and kineticstudies indicate a role in renal tubular folate transport (20).Knockout of the folate binding protein 1 (folbp1) and folatebinding protein 2 (folbp2) genes, the mouse equivalents of FR-and FR-, has shown that deletion of folbp1 is embryolethal butcan be rescued by supplementing the dams with folate (21,22).Surviving mice have low plasma folate levels and may presentwith congenital defects involving the craniofacies, heart, eyes,and abdominal wall (22–24). Furthermore, the injectionof anti-FR into pregnant rats was shown to cause embryonic damage(25), and autoantibodies against FR were recently identifiedin women who previously gave birth to an infant with neuraltube defects (26). These data point to an important role ofFR in folate homeostasis, although whether this is linked toepithelial folate transport or some other function of FR iscurrently unknown.
To establish whether FR are involved in renal folate transport,we examined tubular reabsorption of folate in mice with targetedgene knockout of folbp1 and folbp2. The data suggest an importantrole of folbp1 in renal tubular reabsorption of filtered folate,establishing for the first time a significant role for folbp1in epithelia folate transport in vivo.
Animals
All animal experiments were carried out in accordance with theprovisions for animal care license provided by the Danish NationalAnimal Experiments Inspectorate. Male mice that were defectivein folbp1 or folbp2 were produced by targeted gene knockout(21) and previously characterized (22,23,27). Beginning at 7wk of age, folbp1–/– (n = 6), folbp1+/– (n= 6), folbp2–/– (n = 7), and wild-type (n = 6) micewere fed a low-folate diet (0.3 mg folate/kg with succinyl-sulfathiazole;Dyets #518841, Bethlehem, PA) for 38 d followed by a shift toa folate-replete diet (3 mg folate/kg; Dyets) for 25 d. At theend of both periods, mice were placed in metabolic cages withfree access to water and received an intraperitoneal injectionof 430,000 to 580,000 CPM (approximately 40 pmol) of 3H-(6S)-5-methyltetrahydrofolate( 3H-MTHF; 15 Ci/mmol; Moravek Biochemicals, Brea, CA) in 0.9%NaCl. Urine was collected into 2 ml of 10 mM sodium-ascorbate(Sigma, St. Louis, MO) for 24 h both before 3H-MTHF injectionand 24 h after this. Blood was sampled from the retro-orbitalsinus after urine collection on low-folate diet. After the micehad been fed the normal diet for 25 d and urine collection hadbeen completed, the mice were anesthetized, blood was collected,and the kidneys were removed. All blood samples were heparinizedand centrifuged for the preparation of plasma. Urine, plasma,and kidneys were stored at –80°C. Kidneys from additionalfolbp1–/–, folbp2–/–, and wild-typemice were fixed by retrograde perfusion through the abdominalaorta with 2% paraformaldehyde or 1% glutaraldehyde in 0.1 Msodium cacodylate buffer (pH 7.4).
Crude Membrane Preparation
Mouse kidneys were homogenized in 0.3 M sucrose, 25 mM imidazole,1 mM EDTA, 8.5 µM leupeptin, and 1 mM PMSF (pH 7.2), usingan Ultraturrax T8 homogenizer (IKA Labortechnik, Staufen, Germany)at maximum speed for 30 s. After centrifugation at 4000 x gfor 15 min, crude membranes were isolated by centrifugationof the supernatant at 200,000 x g for 1 h. Membrane proteinconcentration was determined using a Coomassie blue commercialassay (Bio-Rad Laboratories, Hercules, CA).
Membrane Folate Binding 3H-folic acid binding was estimated using a method modifiedfrom Mason and Selhub (28). Approximately 13 to 28 µgof crude membrane preparations in 100 µl of 10 mM HEPES,140 mM NaCl, 2 mM CaCl2, and 1 mM MgCl2 (pH 7.4) was incubatedwith 200 µl of 1 M acetate buffer (pH 3.5) to dissociateendogenous folate, followed by the addition of 7.2 kBq 3H-folicacid (approximately 6 pmol; Moravek Biochemicals) and normalizationof pH with 300 µl of 1 M K2HPO4 (pH 8). The mixture waspassed through equilibrated 0.45-µm cellulose filters(Millipore, Billerica, MA) and washed with 20 ml of 0.1 M K2HPO4(pH 7.4). Filters were dried, and retained radioactivity wascounted in a LKB-Wallac 1211 Rackbeta scintillation counterwith Ecoscint H scintillation fluid (National Diagnostics, Atlanta,GA). Additional kidney membrane binding experiments were performedin the presence of excess 0.5 nmol unlabeled folic acid. Countsfrom blanks representing buffer without kidney membranes wereapproximately 10% of total counts retrieved with crude wild-typemembrane samples. Blanks were subtracted from total counts,and the amount of bound 3H-folic acid was determined by theparallel counting of appropriate 3H-folic acid standards.
Folate Analyses
Folate in plasma and urine was determined using a Lactobacilluscasei assay essentially as described by Molloy and Scott (29).In short, 0.25 M sodium ascorbate was added to 25 µl ofsample, making a total volume of 1 ml. Duplicate volumes of50 and 100 µl of diluted sample or 0 to 100 µl ofa folate standard, 500 pg/ml, prepared from folic acid (Sigma,St. Louis, MO), were added to 96-well microtiter plates. Additional0.25 M sodium ascorbate was added to some wells to make a totalvolume of 100 µl in all. To each well was added 200 µlof folic acid medium prepared from a 57-mg/ml folic acid medium(Becton Dickinson, Sparks, MD) solution that contained 30 µg/mlchloramphenicol (Sigma), 0.75 mg/ml ascorbic acid (Sigma), 30µl of Tween 80 (Sigma), and 200 µl of cryopreservedL. casei (donated by Dr. J.M. Scott, Trinity College, Dublin,Ireland). The plates were incubated for 48 h at 37°C, gentlyshaken, and read at 590 nm using an InerMed ImmunoReader NJ-2000.The amount of folate in each sample was determined from a standardcurve based on the folate standards.
3H activity in urine and kidney tissue was determined by liquidscintillation counting. Because the mice received an injectionof 3H-MTHF on low-folate diet 25 d before repeated injectionon normal diet, urine was collected 1 d before the second injectionto examine whether labeled folate retained from the first injectionwould interfere with the results of the second injection. Theactivity in the urine before the second injection was <2%of the activity collected after the second 3H-MTHF injectionin all animals and, thus, considered insignificant. For countingthe kidney, approximately 100 mg of whole tissue was homogenizedin 1 ml of 0.9% NaCl and solubilized in 2 ml of Solvable (PackardInstrument Company, Meriden, CT). The solution was heated to55°C for 2 h, and a total of 0.2 ml of H2O2 was added followedby heating to 55°C for 30 min. Ten milliliters of scintillationfluid was added before counting.
Other Serum and Urine Analyses
Creatinine and urea in mouse serum and urine were measured usingan automatic equipment (Cobas-Integra, Hoffmann-La Roche, Switzerland).
Calculations
Renal folate, creatinine, and urea clearances were estimatedby dividing 24-h folate, creatinine, or urea excretion by plasmafolate, creatinine, or urea concentrations, respectively, Thelatter were determined from blood collected at the end of the24-h urine sampling period. Thus, because calculations are basedon single, end point serum concentrations, the calculated clearancesshould be considered estimates.
Antibodies
A polyclonal antibody raised against purified rat placentalFR was donated by Dr. S.P. Rothenberg (State University of NewYork). Polyclonal sheep anti–rat-megalin antibodies havebeen described previously (30).
SDS-PAGE and Immunoblotting
Membrane samples were subjected to nonreducing SDS-PAGE usingpolyacrylamide minigels (Bio-Rad Mini Protean II) and transferredto nitrocellulose membranes. Blots were blocked with 5% skimmilk in PBS with 0.1% Tween 20 (PBS-T; pH 7.5) for 1 h. Afterovernight incubation with primary antibody diluted in PBS-Twith 1% BSA, the blots were incubated with horseradish peroxidase–conjugatedgoat anti-rabbit IgG (1:3000; Dako, Glostrup, Denmark) and visualizedusing ECL enhanced chemiluminescence system (Amersham Biosciences,Little Chalfont, UK). Controls involving incubation withoutprimary antibody or incubation with nonspecific serum revealedno significant labeling. The expression of immunoreactive FRwas evaluated by densitometry of immunoblots. Equal amountsof membrane protein samples from different genotypes were blottedfor the FR as described. Identical samples were run on similargels and Coomassie blue stained. Both the ECL films and theCoomassie blue–stained gels were scanned using an AGFADuoscan f40 scanner and Corel Photo Paint software. The labelingdensity of the specific bands on the immunoblots was quantifiedusing custom-made software (31). Values were adjusted for theloaded amount of protein by dividing the density of the specificbands obtained by immunoblotting with the density of all bandsin the corresponding Coomassie-stained gel. Values thus representrelative expression levels in arbitrary units given as means± SEM and adjusting wild-type to 100%.
Ultrastructure and Immunocytochemistry
Small blocks of perfusion-fixed kidney cortex were postfixedin the same fixative. For ultrastructural analysis, glutaraldehyde-fixedkidney cortex was further postfixed in 1% OsO4, en bloc stainedwith uranyl acetate, dehydrated in ethanol, embedded in Epon812 (TAAB; Aldermaston, Berks, UK), sectioned on a ReichertUltracut S microtome (Reichert, Vienna, Austria), stained withlead and uranyl acetate, and observed in a Philips CM 100 electronmicroscope. For immunocytochemistry, fixed kidney cortex wasinfiltrated with 2.3 M sucrose and frozen. Semithin cryosectionswere incubated with the primary antibody, followed by incubationwith horseradish peroxidase–conjugated secondary antibodyand visualization by incubation with diaminobenzidine and 0.03%H2O2. Sections were counterstained with Mayors hematoxylin beforeexamination under a Leica DMR microscope equipped with a SonyS CCD color video camera. Controls involving sections that wereincubated without the primary antibody or incubated with nonspecificrabbit or sheep serum revealed no significant labeling.
Statistical Analyses
Initial Bartless testing of the data suggested that the datado not represent samples of similar variance. Thus, the equivalentnonparametric test (the Kruskall-Wallis test) was applied followedby Dunn’s multiple comparisons test as post hoc testingcomparing folbp1–/–, folbp1+/–, and wild-type,or folbp2–/– and wild-type. Wilcoxon matched pairssigned rank test was used for paired comparisons. P < 0.05was considered significant. Data represent mean ± SEM.
Kidney Function and Structure
One folbp1–/– mouse had to be excluded during theinitial experimental phase, as it seemed ill. All other micethrived and survived throughout the study. Except for a highermean weight of folbp2–/– mice compared with wild-typemice that were maintained on the low-folate diet, no differenceswere identified between genotypes when comparing animal andkidney weight, urine output, urinary creatinine and urea excretion,or creatinine and urea clearance (Table 1). A small yet significantincrease in urinary creatinine and urea excretion was observedin wild-type mice when going from the low-folate to normal diet.No other significant changes were observed in the basic parametersas a result of the different diets, when comparing mice of thesame genotype (Table 1). The proximal tubule ultrastructureof folbp1–/– mice appeared normal (Figure 1), andimmunocytochemistry revealed no apparent difference in the immunolocalizationof other proximal tubule membrane proteins, including megalin(data not shown).
Figure 1. Electron micrograph showing the apical part of a proximal tubule from perfusion-fixed folate binding protein 1–deficient (folbp1–/–) mouse kidney cortex. General ultrastructure appears normal with extensive microvilli (MV) and a well-preserved endocytic apparatus including invaginations, numerous endocytic vesicles (EV), dense apical tubules representing the membrane recycling compartment (arrows), and structurally normal mitochondria (MIT). Bar = 0.5 µm.
Expression of Folate-Binding Proteins
Immunoblotting on crude kidney membranes revealed an approximately40-kD protein band in wild-type, folbp1+/–, and folbp2–/–representing immunoreactive FR (Figure 2a). No band was observedin folbp1–/–, and no other bands were identifiedin kidney membranes from any of the genotypes. When similaramounts of total membrane protein were blotted, the intensityof the band seemed greater in wild-type compared with both folbp1+/–and folbp2–/– specimens, although this was not evaluatedstatistically.
Figure 2. Expression of immunoreactive folate receptor (FR; a) and 3H-folic acid binding (b) in crude kidney membranes from folbp1–/– (lanes 1 to 3), folbp+/– (lanes 4 to 6), wild-type (lanes 7 to 9), and folbp2–/– (lanes 10 to 12) mice. Similar amounts (approximately 8 µg) of total membrane protein as reflected by the similar stained lanes in the Coomassie blue–stained gel (a, insert, top left) were subjected to SDS-PAGE and immunoblotted using an antibody raised against rat placental FR. An approximately 40-kD protein band was identified in folbp1+/–, wild-type, and folbp2–/– mice, showing the expression of immunoreactive FR (a). No bands were observed in folbp1–/–, and no other bands were identified in any of the genotypes. The intensity of the bands was semiquantified as described (a, bottom bars). The positions of molecular weight markers are indicated to the right. The binding of 3H-folic acid to crude kidney membranes was determined as described previously (b, filled bars). No binding of 3H-folic acid to kidney membranes from folbp1–/– could be identified, whereas binding to kidney membranes from all genotypes in the presence of excess unlabeled folic acid was very low (b, open bars).
Immunocytochemistry confirmed the total absence of immunoreactiveprotein in the brush border of folbp1–/– in contrastto the normal labeling of proximal tubule apical membranes inwild-type and folbp2 –/– mice (Figure 3). Additionallabeling was observed in the parietal cells of Bowman’scapsule in these mice, demonstrating for the first time expressionof FR within the renal corpuscle.
Figure 3. Immunolocalization of FR in cryosections of kidney cortex of wild-type (a), folbp1–/– (b), and folbp2–/– (c) mice using a polyclonal antibody raised against rat placental FR. FR is located to the brush border and apical cytoplasm of proximal tubules. Additional labeling is observed in the parietal cells of Bowman’s capsule (arrows in a). No labeling is observed in folbp1–/– (b), whereas folbp2–/– (c) reveals an apparently normal labeling. Bars = 10 µm.
Folate Binding to Kidney Membranes
To quantify membrane-associated folate binding, we measuredhigh-affinity 3H-folic acid binding activity to crude kidneymembranes including both plasma membrane and intracellular membranes(Figure 2b). No binding of 3H-folic acid to kidney membranesfrom folbp1–/– was identified while binding to kidneymembranes in the presence of excess unlabeled folic acid wasvery low (Figure 2b).
Plasma and Urinary Folate
In accordance with previous observations (22), total plasmalevels of folate were significantly lower in the folbp1–/–compared with wild-type mice on both low-folate and normal diets(Table 2). Mean plasma folate levels in folbp1+/– andfolbp2–/– were in between levels in folbp1–/–and wild-type mice, although not statistically significant fromeither. On normal diet, no significant differences in the urinaryfolate excretion rate were observed between genotypes despitedifferences in plasma folate levels. In contrast, on the low-folatediet, folbp1–/– mice revealed an approximately 10-foldincrease in urinary folate excretion rate compared with wild-typeand folbp1+/– mice. No significant differences in urinaryfolate excretion were observed between folbp1+/– or folbp2–/–and wild-type mice.
Table 2. Plasma folate, urinary folate excretion, and calculated renal folate clearance in wild-type, folbp1–/–, folbp+/–, and folbp2–/– mice on low-folate and normal dieta
On normal diet, an approximately five-fold higher estimatedrenal folate clearance was observed in folbp1–/–compared with wild-type mice (Table 2). This was the resultof unaltered urinary folate excretion despite lower plasma folate.The estimated renal folate clearance in folbp1–/–mice on normal diet is comparable to creatinine clearance (Table 2).On the low-folate diet, an approximately 34-fold higherestimated renal folate clearance was observed in the folbp1–/–mice when compared with wild-type mice. This elevated folateclearance is the combined result of an increase in urinary folateexcretion and decreased levels of plasma folate (Table 2). Nosignificant differences in estimated folate clearance were observedbetween folbp1+/– or folbp2–/– and wild-typemice; however, a tendency toward higher urinary folate clearancewas observed in folbp1+/– compared with wild-type miceon both low-folate and normal diet.
Urinary Excretion and Renal Accumulation of Injected 3H-MTHF
For examining handling of an acutely administered physiologicdose of MTHF and to study tissue uptake of folate, approximately40 pmol of 3H-MTHF was injected intraperitoneally into miceon both the low-folate and the normal diet. Twenty-four-hoururinary 3H excretion confirmed the findings of total folateexcretion, showing higher urinary excretion of label in folbp1–/–mice when compared with wild-type (Figure 4). Thus, the handlingof an acutely injected physiologic MTHF dose is consistent withsteady-state observations. After the normal diet period, thekidneys were removed and tissue accumulation of 3H was counted.Twenty-four-hour urinary 3H excretion was increased approximately1.7-fold in folbp1–/– mice, which was associatedwith a similar, significant decrease in the accumulation oflabel within the kidney (Figure 4).This demonstrates that theincreased excretion of folate associated with the folbp1–/–genotype is not due to changes in filtration as increased glomerularfiltration along with unchanged tubular reabsorption would haveresulted in unaltered or increased renal accumulation. Similarto the results of total folate handling, no significant changesin the urinary excretion and renal accumulation of 3H-MTHF wereobserved as a result of the folbp2 deficiency.
Figure 4. 24-hour urinary excretion of 3H-label after intraperitoneal injection of approximately 40 pmol of 3H-MTHF during low-folate and normal diet conditions (left axis scale), as well as renal accumulation of label after normal diet (right axis scale). Values are expressed in percentage of total injected label. Significantly less label is excreted under low-folate conditions as compared with normal diet in folbp1+/–, folbp2–/–, and wild-type mice. A significant increase in urinary excretion of label is observed in folbp1–/– compared with wild-type mice on both low-folate and normal diet. On low-folate diet, a 5.6-fold increase in urinary folate excretion is observed compared with a 1.7-fold increase on normal diet. The latter is associated with a significant, approximately 45% decrease in renal 3H-accumulation. Data represent mean ± SEM. *Significantly different compared with wild-type mice on similar diet (Kruskal-Wallis test followed by Dunn’s multiple comparisons test); #significantly different compared with mice of the same genotype under normal diet conditions (Wilcoxon matched pairs signed rank test).
The present study establishes an important physiologic rolefor the folbp1 protein in mouse renal, tubular reabsorptionof folate. This conclusion is based on the finding of increasedrenal folate clearance in folbp1–/– mice both atlow and normal folate intake showing impaired tubular uptakeof filtered folate. The importance of the folbp1 protein isgreater during folate deficiency than during normal folate intake.On the low-folate diet, renal folate clearance in wild-typemice is reduced to <1% of creatinine clearance, whereas folbp1–/–mice lose folate in the urine at a rate corresponding to approximately23% of creatinine clearance despite an approximately 2.5-folddecrease in plasma folate. Plasma folate levels in folbp1–/–mice on normal diet are comparable to the levels in wild-typemice on the low-folate diet (26 ± 5 versus 27 ±2 nM). Therefore, the amount of filtered folate is assumed tobe similar in these situations. Nevertheless, the amount offolate excreted in the urine of folbp1–/– mice isapproximately 100 times higher than that detected in the wild-typemice (4.3 ± 0.4 versus 0.04 ± 0.01 nmol/24 h),indicating that the tubular reabsorptive capacity related specificallyto folbp1 in the low-folate situation is approximately 4 nmol/24h in mice. The essential role of folbp1 in tubular folate transportis further supported by the total lack of kidney membrane 3H-folicacid binding associated with folbp1 deficiency. Injection of3H-MTHF also confirms the important role of folbp1 for renaltubular reabsorption, showing increased urinary excretion anddecreased renal accumulation in folbp1–/–. Thesefindings, establishing for the first time in an in vivo modela direct role for folbp1 in epithelia folate uptake, confirmsthe hypotheses raised by the identification of FR in the kidneyproximal tubule and by observations on folate transport in thekidney and in cultured kidney cells (14,20,32,33). In addition,folbp1 is heavily expressed in other absorptive epithelia, includingthe rodent yolk sac (34) and choroid plexus (15). These findingsindicate a similarly important role for folbp1 in transepithelialfolate transport in these tissues and provide a potential mechanismfor the defective embryonic development observed in unsupplementedfolbp1–/– mouse embryos (21). The estimated urinaryloss of 4 nmol/24 h ascribed to folbp1 deficiency correspondsto the folate content in approximately 6 g of low-folate diet,which is more than the average daily intake. Thus, as a resultof urinary loss, it is impossible to maintain adequate folatelevels in folbp1–/– mice on the low-folate diet,contributing to the observed reduction in plasma folate levels.
The reduced folate clearance observed on low-folate diet inboth folbp1–/– and wild-type animals is consistentwith previous observations that urinary folate clearance decreaseswith decreasing plasma folate levels (35). Given a low-folateintake resulting in low plasma folate levels, reduced amountsof folate are filtered within the renal glomeruli. Under theseconditions, folbp1 permits almost complete tubular reabsorptionof folate, as shown by a wild-type folate clearance <1% ofcreatinine clearance. In contrast, folate clearance in folbp1-deficientmice on the normal diet is approximately 100% of creatinineclearance, suggesting little or no net tubular reabsorptionof folate. It is interesting that on the low-folate diet, thefolate clearance in folbp1–/–mice is only approximately20% rather than the 100% of creatinine clearance that may beexpected when folate is freely filtered and tubular reabsorptionis defective. However, in addition to free folate, plasma containsboth specific and nonspecific folate binders, the latter probablybeing albumin (36–38). Whereas total protein binding offolate in rat serum has been estimated to be approximately 20%over a wide range of concentrations (20), the concentrationof the approximately 35 kD, high-affinity plasma folate bindingprotein secreted from hemopoietic cells (11,12), is assumedto be more constant and thus carrying a larger fraction of totalplasma folate at low folate concentration. Filtration of thisfolate-protein complex is followed by tubular reabsorption,most likely by a megalin-mediated process (39), providing apotential mechanism of tubular folate uptake that is independentof membrane-associated folbp1 and may play a role at low plasmafolate concentrations. Finally, it cannot be excluded that other,very low capacity reabsorptive mechanisms may operate underlow-folate conditions.
The absence of 3H-folic acid binding to kidney membranes offolbp1–/– mice suggests that the folbp1 gene productis the major membrane folate binder in the kidney. This is furthersupported by the total lack of immunoreactive FR in folbp1–/–kidney cortex. Thus, although FR- mRNA has been identified inhuman kidney (8), the role of folbp2 in mouse kidney remainsunclear. The present data do not support a role for folbp2 inrenal folate reabsorption. Immunoblotting revealed lower levelsof FR in folbp1+/– and folbp2–/– kidneys,both showing a tendency of lower plasma folate concentrations,compared with the wild-type mice. Because it has been shownthat low folate intake and low plasma folate levels are associatedwith decreased expression of FR in rat kidney (40), it is possiblethat the decreased expression of immunoreactive FR in folbp1+/–and folbp2–/– kidneys reflects this rather thana direct effect of gene knockout.
Although this study clearly shows an important role of folbp1in renal epithelial uptake of folate, it does not exclude asignificant function of other folate transport proteins. Thekidney expresses both FR and RFC (17,41,42). Proximal tubulecell culture studies have implicated RFC in apical transportof MTHF (43,44), and RFC has been located by immunocytochemistryto basolateral membranes in kidney tubules (17). Thus, RFC maybe involved in apical folate transport at high folate concentrationsor in the cellular exit of reabsorbed folate.
In conclusion, this study has established the important roleof folbp1 in renal tubular reabsorption of folate in vivo. Thegreatest relative importance of folbp1 is observed at low-folateintake. Although folbp2 has been identified in kidney, the presentstudy does not support a role in folate reabsorption.
Acknowledgments
This work was supported in part by the Danish Medical ResearchCouncil, Aarhus University, the NOVO-Nordisk Foundation, LjgevidenskabensFremme, the Biomembrane Research Center, and the Birn-Foundation.R.H.F. and O.S. were supported by the National Institutes ofHealth, grant HL66398.
This study was presented in part at the American Society ofNephrology Annual Meeting, November 1–4, 2003, Philadelphia,PA.
We thank Dr. S.P. Rothenberg, State University of New York,for donating the anti-FR antibody, and Dr. J.M. Scott, TrinityCollege (Dublin, Ireland), for invaluable assistance in settingup the L. casei assay. The technical assistance by A.M. Hass,P. Nielsen, H. Sidelmann, and I. Kristoffersen is greatly appreciated.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
O.S. is currently affiliated with Teva Pharmaceutical Industries,Innovative R&D, Netanya, Israel.
Said HM: Recent advances in carrier-mediated intestinal absorption of water-soluble vitamins.
Annu Rev Physiol 66
: 419
–446, 2004[CrossRef][Medline]
Kamen BA, Smith AK: A review of folate receptor alpha cycling and 5-methyltetrahydrofolate accumulation with an emphasis on cell models in vitro.
Adv Drug Deliv Rev 56
: 1085
–1097, 2004[CrossRef][Medline]
Matherly LH, Goldman DI: Membrane transport of folates.
Vitam Horm 66
: 403
–456, 2003[Medline]
Elwood PC: Molecular cloning and characterization of the human folate-binding protein cDNA from placenta and malignant tissue culture (KB) cells.
J Biol Chem 264
: 14893
–14901, 1989[Abstract/Free Full Text]
Lacey SW, Sanders JM, Rothberg KG, Anderson RG, Kamen BA: Complementary DNA for the folate binding protein correctly predicts anchoring to the membrane by glycosyl-phosphatidylinositol.
J Clin Invest 84
: 715
–720, 1989
Ross JF, Chaudhuri PK, Ratnam M: Differential regulation of folate receptor isoforms in normal and malignant tissues in vivo and in established cell lines. Physiologic and clinical implications.
Cancer 73
: 2432
–2443, 1994[CrossRef][Medline]
Ratnam M, Marquardt H, Duhring JL, Freisheim JH: Homologous membrane folate binding proteins in human placenta: Cloning and sequence of a cDNA.
Biochemistry 28
: 8249
–8254, 1989[CrossRef][Medline]
Wang X, Shen F, Freisheim JH, Gentry LE, Ratnam M: Differential stereospecificities and affinities of folate receptor isoforms for folate compounds and antifolates.
Biochem Pharmacol 44
: 1898
–1901, 1992[CrossRef][Medline]
Shen F, Wu M, Ross JF, Miller D, Ratnam M: Folate receptor type gamma is primarily a secretory protein due to lack of an efficient signal for glycosylphosphatidylinositol modification: Protein characterization and cell type specificity.
Biochemistry 34
: 5660
–5665, 1995[CrossRef][Medline]
Shen F, Ross JF, Wang X, Ratnam M: Identification of a novel folate receptor, a truncated receptor, and receptor type beta in hematopoietic cells: cDNA cloning, expression, immunoreactivity, and tissue specificity.
Biochemistry 33
: 1209
–1215, 1994[CrossRef][Medline]
Spiegelstein O, Eudy JD, Finnell RH: Identification of two putative novel folate receptor genes in humans and mouse.
Gene 258
: 117
–125, 2000[CrossRef][Medline]
Kamen BA, Caston JD: Identification of a folate binder in hog kidney.
J Biol Chem 250
: 2203
–2205, 1975[Abstract/Free Full Text]
Selhub J, Franklin WA: The folate-binding protein of rat kidney. Purification, properties, and cellular distribution.
J Biol Chem 259
: 6601
–6606, 1984[Abstract/Free Full Text]
Holm J, Hansen SI, Høier Madsen M, Bostad L: A high-affinity folate binding protein in proximal tubule cells of human kidney.
Kidney Int 41
: 50
–55, 1992[Medline]
Wang Y, Zhao R, Russell RG, Goldman ID: Localization of the murine reduced folate carrier as assessed by immunohistochemical analysis.
Biochim Biophys Acta 1513
: 49
–54, 2001[Medline]
Birn H, Selhub J, Christensen EI: Internalization and intracellular transport of folate-binding protein in rat kidney proximal tubule.
Am J Physiol 264
: C302
–C310, 1993
Hjelle JT, Christensen EI, Carone FA, Selhub J: Cell fractionation and electron microscope studies of kidney folate-binding protein.
Am J Physiol 260
: C338
–C346, 1991
Selhub J, Emmanouel D, Stavropoulos T, Arnold R: Renal folate absorption and the kidney folate binding protein. I. Urinary clearance studies.
Am J Physiol 252
: F750
–F756, 1987
Piedrahita JA, Oetama B, Bennett GD, van Waes J, Kamen BA, Richardson J, Lacey SW, Anderson RG, Finnell RH: Mice lacking the folic acid-binding protein Folbp1 are defective in early embryonic development.
Nat Genet 23
: 228
–232, 1999[CrossRef][Medline]
Spiegelstein O, Lu X, Le XC, Troen A, Selhub J, Melnyk S, James SJ, Finnell RH: Effects of dietary folate intake and folate binding protein-1 (Folbp1) on urinary speciation of sodium arsenate in mice.
Toxicol Lett 145
: 167
–174, 2003[CrossRef][Medline]
Spiegelstein O, Mitchell LE, Merriweather MY, Wicker NJ, Zhang Q, Lammer EJ, Finnell RH: Embryonic development of folate binding protein-1 (Folbp1) knockout mice: Effects of the chemical form, dose, and timing of maternal folate supplementation.
Dev Dyn 231
: 221
–231, 2004[Medline]
Tang LS, Finnell RH: Neural and orofacial defects in Folp1 knockout mice.
Birth Defects Res Part A Clin Mol Teratol 67
: 209
–218, 2003[CrossRef][Medline]
da Costa M, Sequeira JM, Rothenberg SP, Weedon J: Antibodies to folate receptors impair embryogenesis and fetal development in the rat.
Birth Defects Res Part A Clin Mol Teratol 67
: 837
–847, 2003[CrossRef][Medline]
Rothenberg SP, da Costa MP, Sequeira JM, Cracco J, Roberts JL, Weedon J, Quadros EV: Autoantibodies against folate receptors in women with a pregnancy complicated by a neural-tube defect.
N Engl J Med 350
: 134
–142, 2004[Abstract/Free Full Text]
Spiegelstein O, Merriweather MY, Wicker NJ, Finnell RH: Valproate-induced neural tube defects in folate-binding protein-2 (Folbp2) knockout mice.
Birth Defects Res Part A Clin Mol Teratol 67
: 974
–978, 2003
Mason JB, Selhub J: Folate-binding protein and the absorption of folic acid in the small intestine of the suckling rat.
Am J Clin Nutr 48
: 620
–625, 1988[Abstract/Free Full Text]
Molloy AM, Scott JM: Microbiological assay for serum, plasma, and red cell folate using cryopreserved, microtiter plate method.
Methods Enzymol 281
: 43
–53, 1997[CrossRef][Medline]
Sahali D, Mulliez N, Chatelet F, Laurent Winter C, Citadelle D, Sabourin JC, Roux C, Ronco P, Verroust P: Comparative immunochemistry and ontogeny of two closely related coated pit proteins. The 280-kd target of teratogenic antibodies and the 330-kd target of nephritogenic antibodies.
Am J Pathol 142
: 1654
–1667, 1993[Abstract]
Marples D, Knepper MA, Christensen EI, Nielsen S: Redistribution of aquaporin-2 water channels induced by vasopressin in rat kidney inner medullary collecting duct.
Am J Physiol 269
: C655
–C664, 1995
Selhub J, Rosenberg IH: Demonstration of high-affinity folate binding activity associated with the brush border membranes of rat kidney.
Proc Natl Acad Sci U S A 75
: 3090
–3093, 1978[Abstract/Free Full Text]
McMartin KE, Morshed KM, Hazen Martin DJ, Sens DA: Folate transport and binding by cultured human proximal tubule cells.
Am J Physiol 263
: F841
–F848, 1992
Birn H, Crüger D, Hansen SI, Holm J, Høier Madsen M: Distribution and characterization of high-affinity folate binding proteins in rat yolk sac and human coelomic fluid. In:
Chemistry and Biology of Pteridines and Folates, edited by Pfleiderer W, Rokos H, Berlin, Blackwell Science, 1997
, pp 349
–352
Goresky CA, Watanabe H, Johns DG: The renal excretion of folic acid.
J Clin Invest 42
: 1841
–1849, 1963
Sasaki K, Natsuhori M, Shimoda M, Saima Y, Kokue E: Role of high-affinity folate-binding protein in the plasma distribution of tetrahydrofolate in pigs.
Am J Physiol 270
: R105
–R110, 1996[Medline]
Holm J, Hansen SI, Lyngbye J: High-affinity binding of folate to a protein in serum of male subjects.
Clin Chim Acta 100
: 113
–119, 1980[CrossRef][Medline]
Soliman HA, Olesen H: Folic acid binding by human plasma albumin.
Scand J Clin Lab Invest 36
: 299
–304, 1976[Medline]
Birn H, Zhai XY, Holm J, Hansen SI, Jacobsen C, Moestrup SK, Christensen EI: Megalin binds and mediates cellular uptake of the folate binding protein [Abstract].
J Am Soc Nephrol 13
: 486A
, 2002
da Costa M, Rothenberg SP, Sadasivan E, Regec A, Qian L: Folate deficiency reduces the GPI-anchored folate-binding protein in rat renal tubules.
Am J Physiol Cell Physiol 278
: C812
–C821, 2000[Abstract/Free Full Text]
Nguyen TT, Dyer DL, Dunning DD, Rubin SA, Grant KE, Said HM: Human intestinal folate transport: Cloning, expression, and distribution of complementary RNA.
Gastroenterology 112
: 783
–791, 1997[CrossRef][Medline]
Whetstine JR, Flatley RM, Matherly LH: The human reduced folate carrier gene is ubiquitously and differentially expressed in normal human tissues: Identification of seven non-coding exons and characterization of a novel promoter.
Biochem J 367
: 629
–640, 2002[CrossRef][Medline]
Sikka PK, McMartin KE: Determination of folate transport pathways in cultured rat proximal tubule cells.
Chem Biol Interact 114
: 15
–31, 1998[CrossRef][Medline]
Morshed KM, Ross DM, McMartin KE: Folate transport proteins mediate the bidirectional transport of 5-methyltetrahydrofolate in cultured human proximal tubule cells.
J Nutr 127
: 1137
–1147, 1997[Abstract/Free Full Text]
Received for publication August 29, 2004.
Accepted for publication December 13, 2004.
This article has been cited by other articles:
C. Muller, R. Schibli, E. P. Krenning, and M. de Jong Pemetrexed Improves Tumor Selectivity of 111In-DTPA-Folate in Mice with Folate Receptor-Positive Ovarian Cancer
J. Nucl. Med.,
April 1, 2008;
49(4):
623 - 629.
[Abstract][Full Text][PDF]
A. Qiu, S. H. Min, M. Jansen, U. Malhotra, E. Tsai, D. C. Cabelof, L. H. Matherly, R. Zhao, M. H. Akabas, and I. D. Goldman Rodent intestinal folate transporters (SLC46A1): secondary structure, functional properties, and response to dietary folate restriction
Am J Physiol Cell Physiol,
November 1, 2007;
293(5):
C1669 - C1678.
[Abstract][Full Text][PDF]
G. Luo, C. E. Garner, H. Xiong, H. Hu, L. E. Richards, K. L. R. Brouwer, J. Duan, C. P. Decicco, T. Maduskuie, H. Shen, et al. Effect of DPC 333 [(2R)-2-{(3R)-3-Amino-3-[4-(2-methylquinolin-4-ylmethoxy)phenyl]-2-oxopyrrolidin-1-yl}-N-hydroxy-4-methylpentanamide], a Human Tumor Necrosis Factor {alpha}-Converting Enzyme Inhibitor, on the Disposition of Methotrexate: A Transporter-Based Drug-Drug Interaction Case Study
Drug Metab. Dispos.,
June 1, 2007;
35(6):
835 - 840.
[Abstract][Full Text][PDF]
P.-X. Wang and P. W. Sanders Immunoglobulin Light Chains Generate Hydrogen Peroxide
J. Am. Soc. Nephrol.,
April 1, 2007;
18(4):
1239 - 1245.
[Abstract][Full Text][PDF]
H. Birn The kidney in vitamin B12 and folate homeostasis: characterization of receptors for tubular uptake of vitamins and carrier proteins
Am J Physiol Renal Physiol,
July 1, 2006;
291(1):
F22 - F36.
[Abstract][Full Text][PDF]
Top
Abstract
Introduction
is present predominantly in epithelial cells (6–8), FR-
is expressed at low to moderate levels in several different tissues (8,9) and differs with respect to affinities for different forms of folate (10). FR-
/
) predicts a 27.7-kD protein with a unique expression pattern in both adult and embryonic tissues (13). Both FR-
