QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.107.037101v1 72/3/553    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Foraker, A. B. Articles by Swaan, P. W. Search for Related Content PubMed PubMed Citation Articles by Foraker, A. B. Articles by Swaan, P. W.

Dynamin 2 Regulates Riboflavin Endocytosis in Human Placental Trophoblasts

Department of Pharmaceutical Sciences, School of Pharmacy, University of Maryland, Baltimore, Maryland (A.B.F., A.R., T.C.D.S., L.M.B., P.W.S.); Drug Disposition, Lilly Research Laboratories, Eli Lilly and Company, Indianapolis, Indiana (K.M.H.); and Division of Pharmaceutics, College of Pharmacy, the Ohio State University, Columbus, Ohio (T.D.S.)

Received April 13, 2007; accepted June 12, 2007

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Riboflavin is thoroughly established to be indispensable in a multitude of cellular oxidation-reduction reactions through its conversion to coenzyme forms flavin mononucleotide and flavin adenine dinucleotide. Despite its physiological importance, little is known about specific mechanisms or proteins involved in regulating its cellular entry in humans. Studies involving biochemical modulators and immunological inhibition assays have indirectly revealed that riboflavin internalization and trafficking occurs at least in part through a clathrin-dependent receptor-mediated endocytic process. Here, using a two-tiered strategy involving RNA interference and the overexpression of dominant-negative constructs, we directly show the involvement of this endocytic mechanism through the requirement of the pluripotent endocytic vesicle scission enzyme, dynamin 2 GTPase, in human placental trophoblasts. Similar to the endocytic control ligand, transferrin, riboflavin is shown to exhibit 50% dependence on the functional expression of dynamin 2 for its active cellular entry. Furthermore, this reduced vitamin uptake correlates with >2-fold higher riboflavin association at the cell surface. In addition, fluorescent ligand endocytosis assays showing colocalization between rhodamine-riboflavin and the immunostained caveolar coat protein, caveolin 1, suggest that the active absorption of this important nutrient involves multiple and distinct endocytosis pathways.

Vitamin B₂ is an essential nutrient required by oxidation-reduction pathways critical in normal cellular growth, function, and maintenance. States of physiological B₂ deficiency have been correlated with clinical manifestations including cardiovascular disease, stunted growth, anemia, and neurodegeneration (Powers, 2003, 2005). Despite its clinical and physiological value, the cellular mechanism(s) regulating B₂ absorption is poorly defined. Recent reports suggest that one portal of entry for B₂ may involve CME (Huang and Swaan, 2000, 2001; Huang et al., 2003; D'Souza et al., 2006b). These studies demonstrated cytoskeletal dependence and clathrin- and/or Rab5-positive endosomal enrichment of B₂ in placental trophoblasts (BeWo). Because both clathrin and Rab5 are endocytic markers of CME, these data suggest that B₂ absorption involves CME. Furthermore, analogous to the well-characterized CME ligand, transferrin, a B₂ carrier protein has been proposed to sequester free extracellular B₂ and facilitate its transport to endosomal organelles (Mason et al., 2006). Although such a soluble carrier protein has been characterized in oviparous species (Abrams et al., 1988; Zheng et al., 1988), the identity of a human homolog remains elusive. A major limitation to the B₂-RME model is a lack of direct evidence identifying the functional dependence on critical proteins regulating this nutrient absorption mechanism in humans. Dynamin 2 is a candidate protein regulator of cellular B₂ levels.

Dynamin 2 (DNM2) is a ubiquitously expressed GTPase known to regulate the invagination and constriction of vesicles at the plasma membrane and trans-Golgi domain of mammalian endothelial and epithelial cells (Hill et al., 2001; Conner and Schmid, 2003). It is required by multiple endocytic mechanisms including those pathways dependent on clathrin or caveolin, and consequently it has been described as the "master regulator of membrane trafficking events at the cell surface" (Conner and Schmid, 2003). Based on prior studies suggesting a B₂-CME mechanism occurring in the placental trophoblast model (Huang et al., 2003; D'Souza et al., 2006a,b), we hypothesized that B₂ internalization requires dynamin 2 GTPase. Using the established BeWo system, a two-tiered approach involving RNAi and transient transfections of a GTPase-null (K44A) dynamin 2-expression construct was carried out to elucidate the role of this enzyme in B₂ RME.

To date, the proposed B₂-RME model has been defined exclusively by the involvement of the CME mechanism in humans. However, other distinct RME pathways have been shown to be critical in nutrient absorption. Folate uptake is documented to occur through caveolae-mediated endocytosis (CvME) in addition to CME (Birn, 2006). Furthermore, prior data that revealed B₂ enrichment to Rab5 GTPase-positive endosomes in human placental trophoblasts and enterocytes (D'Souza et al., 2006b) may reflect a CvME trafficking mechanism. Recent reports have implicated the potential for cross-talk in cargo transport between caveosomes (i.e., an organelle specific to CvME) and Rab5-positive vesicles of CME (Querbes et al., 2006). Thus, it becomes plausible that B₂-RME may involve multiple endocytic pathways. The involvement of CvME in B₂ trafficking in BeWo cells was determined using 3D fluorescence colocalization analyses between internalized rhodamine-labeled B₂ (Phelps et al., 2004) and immunostained caveolin 1 (i.e., a CvME membrane coat protein).

Overall, the objectives of this study were to define the general extent by which B₂ absorption depends on endocytosis as a function of the expression level of the major vesicle scission protein, dynamin 2 GTPase, and evaluate whether B₂-RME involves the CvME pathway in human placental trophoblast cells. Results from this study are the first to directly reveal the dependence of B₂ absorption on dynamin 2 expression in human epithelia and demonstrates, in addition to CME, the involvement of CvME in intracellular B₂ trafficking.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Cell Culture. Human placental trophoblasts (BeWo, passages 205–220) were routinely maintained in a controlled atmosphere at 37°C with 5% CO₂ in F-12K culture media (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, 1% nonessential amino acids, 100 U/ml penicillin, and 100 µg/ml streptomycin. All transfection treatments were tested for cytotoxic effects on cell viability using the lactate dehydrogenase-based CytoTox-ONE assay according to manufacturer's instructions (Promega, Madison, WI). All chemicals used in these studies were purchased from Sigma (St. Louis, MO) unless stated otherwise.

Duplex Short-Interfering RNA Transient Transfection. Cells grown to 50 to 60% confluence were seeded (4.2 x 10⁴ cells/cm²) onto 24-well plates using antibiotic-free F-12K media. Twenty-four hours later, cells were briefly washed and preincubated in serum-free and antibiotic-free Opti-MEM I media (Invitrogen) for 2 h at 37°C and under 5% CO₂. After this incubation, cells were transfected in Opti-MEM I media with 20 or 40 nM concentration of 21-mer duplex short-interfering RNA (siRNA) complexed with Lipofectamine 2000 [1.5% (v/v)] according to the manufacturer's instructions (Invitrogen). Cells were incubated with siRNA-lipid complexes at 37°C and 5% CO₂ for 6 h, at which time the transfection media was replaced with normal F-12K media lacking antibiotics. Cells were used in experiments 48 to 72 h after transfection. All RNA interference studies involved the use of experimentally validated duplex siRNA targeting the human dynamin 2 gene (DNM2 pool siGenome Smartpool, GenBank accession no. NM_004945; Dharmacon RNA Technologies, Chicago, IL) and human dynamin-like protein 1 (DLP1; GenBank accession no. NM_012063; QIAGEN, Cambridge, MA). Targeting siRNA effects were normalized to cells transfected with equivalent amounts of nontargeting duplex siRNA (siControl NonTargeting siRNA; Dharmacon RNA Technologies) and compared with mock (Lipofectamine 2000 alone) and untreated cell conditions.

Western Blotting and Chemiluminescence-Based Densitometry. Whole cells transfected 63 to 68 h earlier with siRNA were harvested and lysed on ice for 20 min in radioimmunoprecipitation assay buffer [10 mM Tris-HCl, pH 7.5, 140 mM NaCl, 1% (v/v) Triton X-100, 1% (w/v) sodium deoxycholate, and 0.1% (w/v) SDS] supplemented with a Complete Mini protease cocktail tablet according to manufacturer's instructions (Roche Diagnostics, Indianapolis, IN). Total protein (10–15 µg) was resolved on 12.5% Tris-HCl Criterion gels (Bio-Rad Laboratories, Hercules, CA), transferred to polyvinylidene difluoride membranes, and immunoblotted using monoclonal antibodies specific to dynamin 2 (final concentration, 0.15 µg/ml; Calbiochem, San Diego, CA) or dynamin-like protein 1 (final concentration, 0.25 µg/ml; BD Pharmingen, San Diego, CA). Primary antibodies were subsequently labeled with horseradish peroxidase-conjugated IgG [1:20,000 (v/v) working dilution; GE Healthcare, Chalfont St. Giles, Buckinghamshire, UK] and detected using the ECL plus system (GE Healthcare). Protein levels were quantitated using chemiluminescence-based densitometry on a chemi-doc universal hood II system (Bio-Rad Laboratories) and normalized to the corresponding housekeeping protein expression for either glyceraldehyde-3-phosphate dehydrogenase (final concentration, 2 µg/ml; Ambion Inc., Austin, TX) or -actin (final concentration, 2.4 µg/ml; Sigma-Aldrich).

GTPase-Null (K44A) and Wild-Type Dynamin 2 Plasmid Transient Transfections and Cotransfections with DNM2 Pool siRNA. Characterized wild-type (DNM2^WT) and dominant-negative (DNM2^K44A) dynamin 2 expression constructs (both in the mammalian expression vector, pCR3.1; Invitrogen) were kindly provided by Dr. Mark A. McNiven (Mayo Clinic and Foundation, Rochester, MN) (Cao et al., 1998). Before transfections, both plasmids were transformed into DH5- cells (Invitrogen), subcultured in Luria broth with ampicillin selection, and purified using the Plasmid Maxi Kit (QIAGEN). Expression vector identities were confirmed through restriction enzyme digests and sequencing using the following established dynamin 2 primers: forward primer, [5'-GAAGAGGGCCATACC-3']; and reverse primer, [5'-AGTTGCGGATGGTCTC-3'] (Cao et al., 1998). Cell seeding conditions were maintained as in RNAi studies. Twenty-four hours after seeding, cells were briefly washed and preincubated in serum-free and antibiotic-free Opti-MEM I for 2 h at 37°C and under 5% CO₂. Cells were transfected using Opti-MEM I with 400 ng of plasmid DNA alone or with 40 nM DNM2 pool siRNA and complexed with Lipofectamine 2000 [1.5% (v/v)]. Cells were exposed to these transfection complexes for 6 h at 37°C and under 5% CO₂. Thereafter, the transfection media were replaced with complete F-12K media devoid of antibiotics. Cells were used in experiments 48 to 72 h after transfection.

Radiolabeled Ligand Endocytosis Assays. Cells were dosed with either 5 nM [³H]riboflavin ([³H]B₂; 41 Ci/mmol; Moravek Biochemicals, Brea, CA) or 10 nM [¹²⁵I]transferrin ([¹²⁵I]TF; 400 cpm/pmol) iodinated using the IODOGEN method (Pierce Biotechnology Inc., Rockford, IL) according to established procedures (D'Souza et al., 2006b) in Hanks' balanced salt solution, pH 7.4, containing 25 mM D-glucose and 10 mM HEPES at 37°C for 4 min. Immediately, cells were placed on ice, and free ligands (i.e., ligands not bound to cell surface receptors) were removed by washing three times with ice-cold phosphate-buffered saline containing cations Ca²⁺ and Mg²⁺ (PBS), pH 7.4. Plasma membrane-bound ligands were then removed by washing cells two times on ice (5 min/wash) with ice-cold PBS with Ca²⁺ and Mg²⁺, pH 3.0. Cells were alkaline-lysed (1 N NaOH) at 4°C for at least 2 h before internalized ligand quantitation. The extent of plasma membrane-bound and internalized [³H]B₂ and [¹²⁵I]TF was determined using liquid scintillation- or counting, respectively. Both plasma membrane-bound and internalized radiolabeled ligands were normalized to total protein content using the Bradford assay (Bio-Rad). All [³H]B₂ uptake data generated at 37°C was corrected for passively absorbed B₂ that was approximated by performing parallel uptake assays exclusively at 4°C. Actively internalized B₂ (i.e., uptake at 37°C) was defined by subtracting internalized [³H]B₂ at 4°C.

Acid wash samples collected after uptake assays represented plasma membrane-bound ligands and were compared with internalized samples using the following equations:

(1)

(2)

The extent of bound ligand was expressed as a percentage of the sum of bound (b_e) and absorbed (i_e) ligand under experimental conditions and was normalized to the percentage of bound ligand under nontargeting conditions (b_nt) (eq. 1). Likewise, the extent of absorbed ligand (i_e) was defined as a percentage of bound plus internalized ligand under experimental conditions and then was normalized to nontargeting effects (i_nt) (eq. 2).

Fluorescent Ligand Endocytosis Assay and Immunofluorescence Staining. BeWo cells were seeded 3 to 5 days before experiments (5 x 10³ cells/cm²) in collagen-coated BD Falcon culture slides (BD Biosciences, Bedford, MA). After serum starvation for 2 h, pulse-chase assays with 500 nM rhodamine-riboflavin (Rd-RF) (Phelps et al., 2004), 15 nM Alexa Fluor 555-labeled cholera toxin subunit B (CTX; Invitrogen), or 30 nM 5-carboxytetramethylrhodamine-labeled transferrin (TF; Invitrogen) were carried out according to established methods (Huang et al., 2003; D'Souza et al., 2006b). Cells were dosed with the fluorescent ligands for 2 or 10 min at 37°C and then immediately fixed with 4% paraformaldehyde for 20 min at room temperature. The fixed cells were then permeabilized for 20 min with 0.1% Triton X-100 in PBS with Ca²⁺ and Mg²⁺, pH 7.4. Permeabilized cells were blocked with 3% (w/v) bovine serum albumin in PBS, pH 7.4, for 30 min before 1-h immunolabeling for either caveolin 1 with rabbit-anti-caveolin 1 (CAV1) [1:500 (v/v) final dilution; Sigma] or clathrin (final concentration, 1 µg/ml; BD Biosciences). Cells were thoroughly washed with bovine serum albumin/PBS and probed with Alexa Fluor 405-labeled goat-anti-rabbit- or sheep-anti-mouse IgG [1:400 (v/v) final dilution; Invitrogen]. Fluorescence treatments were preserved using GelMount (Biomeda Corporation, Foster City, CA) and kept at –20°C until used in fluorescence imaging.

3D Confocal Laser Scanning Microscopy and Colocalization Analysis. Internalized fluorescent ligands and immunostained endocytic marker proteins, caveolin 1 and clathrin, were imaged using a Nikon Eclipse TE2000 E inverted confocal laser scanning microscope (Nikon Instruments Inc., Melville, NY) outfitted with fixed lasers for 405 and 543 nm and corresponding emission filters of 450/35 and 605/75 nm, respectively. Three-dimensional images were acquired using the following settings on Nikon EZ-C1 software (Gold version 2.3; Image Systems Inc., Columbia, MD): Nikon plan apochromatic 60xA oil objective (1.4 numerical aperture), 3.6-µs scan dwell time, 512 x 512 pixel size resolution, 0.30 µm z-step, and a 150-µm detector pinhole. Raw images were iteratively deconvolved using a calculated point-spread function for individual channels and corrected for background noise using a combination of median filtration and setting threshold levels just greater than negative control treatments with either a nonreactive rhodamine derivative (carboxytetramethylrhodamine-4-amine, a byproduct of the rhodamine-riboflavin conjugation reaction; Phelps et al., 2004) or the secondary Alexa Fluor 405 antibody alone. Restored images were analyzed for 3D colocalized fluorescence between ligands and endosome markers using Volocity, version 3.6.1 (Improvision Inc., Lexington, MA). The extent of colocalization between ligands and endocytic markers was determined by calculating the percentage of total overlap volume (in cubic micrometers) over the corresponding total ligand volume.

Statistics. One-way ANOVA with Dunnett's or Neuman-Keuls multiple comparison tests were used to define statistical significance between the effects of targeting siRNA or wild-type and dominant-negative dynamin 2 construct treatments and control conditions (i.e., nontargeting siRNA, empty vector, mock, and untreated treatments) on ligand trafficking. Statistical significance between the effects of targeting siRNA and controls (i.e., nontargeting siRNA or untreated conditions) on dynamin protein levels was defined using the nonparametric Student's t test. True fluorescence colocalization between channels for fluorescent ligands and immunostained endocytic markers was defined using the Pearson's correlation (Manders et al., 1993).

	Results

Top Abstract Materials and Methods Results Discussion References

 
siRNA Targeting Dynamin 2 or Dynamin-Like Protein 1 Resulted in Specific Protein Knockdown While Maintaining Cell Viability. In addition to the use of nontargeting siRNA treatments and to clearly validate siRNA specificity, all RNAi transient transfection studies included external control conditions involving siRNA targeting a dynamin homolog, DLP1. Like DNM2, DLP1 is a member of the dynamin superfamily (Praefcke and McMahon, 2004) and shares 36% sequence homology to DNM2. In contrast to DNM2, DLP1 is not involved in endocytic vesicle scission events, and has been largely characterized to function in the morphological maintenance of peroxisomes and mitochondria (Praefcke and McMahon, 2004). Based on the current understanding of DLP1, we hypothesized that this protein is not involved in the endocytosis of B₂.

BeWo cells were transiently transfected with 20 or 40 nM targeting siRNA for dynamin 2 GTPase (DNM2 pool) or DLP1. Cells were harvested 63 to 68 h after initiating siRNA transfections and analyzed for target dynamin knockdown using immunoblotting and chemiluminescence-based densitometry. Upon normalization to housekeeping protein levels, the extent of dynamin expression was defined as a percentage of detected dynamin in cells treated exclusively with nontargeting siRNA (N). Parallel cell populations that were treated with the lipid transfection reagent (Mock) or untreated cells served as additional negative controls. Twenty and 40 nM DNM2 pool siRNA treatments substantially reduced dynamin 2 (100 kDa) expression levels to a similar extent (Fig. 1A). However, 20 and 40 nM DLP1 siRNA specifically knocked down DLP1 protein (83 kDa) expression in a dose-dependent manner (Fig. 1B). The lack of a dose-dependent effect on dynamin 2 expression with the DNM2 pool siRNA conditions may reflect a more robust siRNA formulation, which involves the coadministration of four different duplex siRNAs targeting different regions of the DNM2 mRNA. Quantitative densitometry revealed that 40 nM DNM2 pool siRNA treatments led to 78% (21.95 ± 7.03% N) and 82% lower dynamin 2 GTPase levels compared with nontargeting siRNA-treated and untreated cells, respectively (Fig. 1C). To a similar extent, 40 nM DLP1 siRNA conditions resulted in 62% (37.65 ± 11.13% N) and 63% lower DLP1 protein levels compared with nontargeting siRNA and untreated conditions, respectively (Fig. 1D). Neither of the dynamin siRNA treatments revealed significant off-target effects at the protein level. Furthermore, dynamin protein levels were shown to be similar between nontargeting siRNA treatments and untreated cell conditions. These data suggest that the negative control siRNA transfection effects are not significantly altering target protein levels and serves as a valid reference for targeting siRNA data normalization.

View larger version (42K): [in this window] [in a new window]   Fig. 1. Dynamin siRNA reveal target-specific knockdown at the protein level. Cells transfected with duplex siRNA targeting mRNA for DNM2 pool or DLP1 were examined for target protein knockdown using Western blotting and quantitative chemiluminescence-based densitometry and were normalized to housekeeping proteins, glyceraldehyde-3-phosphate dehydrogenase, or -actin. Twenty and 40 nM targeting siRNA (A) DNM2p, and DLP1 treatments (B) were compared with 40 N siRNA, mock, and untreated (U) cells. Cells treated with 40 nM targeting or nontargeting siRNA were assessed for the extent of DNM2 (C) or DLP1 (D) knockdown and were expressed as a percentage of nontargeting siRNA effects. The data represent the standard deviations of two to four separate experiments. Statistical significance between targeting and nontargeting effects on dynamin protein expression were defined by Student's t test (**, p < 0.01).

Silenced Dynamin 2 Resulted in a Significant Reduction in [³H]B₂ Internalization at Physiological Temperatures. It has been well-established that B₂ gains entry into human epithelial cells via multiple mechanisms, including a passive diffusion component that seems to dominate at oversupplemented riboflavin levels and an active component that has been reported to coincide with micromolar B₂ concentrations (Foraker et al., 2003). One of the salient features defining an active absorption mechanism is temperature-dependence. Saturable absorption kinetics have been consistently shown to correlate with physiological temperatures (37°C), whereas low temperatures (4°C) generate linear absorption profiles reflecting passive diffusion (Said and Ma, 1994). In fact, B₂ transport has been reported to be dependent on temperature in divergent cell models (Said and Ma, 1994; Huang and Swaan, 2001). However, temperature-dependence alone does not discriminate between a carrier/transporter and receptor-mediated endocytic process. Therefore, to differentiate between such active uptake mechanisms, the standardized RNAi methods involving siRNA-induced silencing of the conserved endocytic protein, dynamin 2 GTPase, in combination with the effects of temperature change on B₂ absorption were investigated. To date, the human placental trophoblast cell model (BeWo) has been shown to express high affinity for riboflavin (2 nM) (Huang and Swaan, 2001), and such nanomolar affinities further suggest the involvement of RME in this vitamin's cellular uptake. Combined with prior reports suggesting a B₂-specific RME absorption process (Huang and Swaan, 2001; Huang et al., 2003; D'Souza et al., 2006a,b), these data validate the utility of the BeWo cell model to characterize the B₂-specific RME pathway(s). BeWo cells transfected with 40 nM targeting or nontargeting siRNA 63 to 72 h prior were used in endocytosis assays. Transiently transfected cells were dosed with 5 nM [³H]B₂ for 4 min at 4 or 37°C. The 4-min internalization period was chosen for all endocytic assays because this is the characterized time interval coinciding with the logarithmic uptake phase typical of RME mechanisms (Schmid, 2004). Passively diffusing riboflavin was defined by vitamin uptake detected at 4°C. Actively internalized B₂ was determined by the amount of B₂ absorbed at 37°C minus that absorbed at 4°C. For the remainder of this report, all actively internalized B₂ data shown represents the active absorption component exclusively. In contrast to all other transfection conditions, DNM2 pool siRNA-induced silencing of dynamin 2 GTPase resulted in a significant reduction of 50% (i.e., 0.18 ± 0.11 S.D. and 0.37 ± 0.09 S.D. pmol/mg of protein/4 min for DNM2 pool siRNA-treated and untreated cells, respectively) in actively absorbed B₂ compared with internalized B₂ in untreated cell populations (Fig. 2A). As expected, passively diffusing B₂ (i.e., uptake at 4°C) was unaffected by all transfection conditions (Fig. 2B). Collectively, these results further substantiate the involvement of classic RME machinery (i.e., dynamin 2 GTPase) in regulating the active absorption of B₂ in placental trophoblasts.

View larger version (50K): [in this window] [in a new window]   Fig. 2. Silenced dynamin 2 states result in a significant reduction in actively internalized B2 in BeWo cells. Cells transfected 63 to 72 h earlier with 40 nM siRNA were dosed with 5 nM [3H]B2 for 4 min at 37°C (A) or 4°C (B). Actively internalized B2 (uptake at 37°C) was determined by subtracting passively absorbed riboflavin (uptake at 4°C). These data represent the standard deviations of three to five separate experiments. Statistical significance between targeting and nontargeting treatments was defined using one-way ANOVA with Newman-Keuls multiple comparison test (*, p < 0.05).

Iodinated transferrin ([¹²⁵I]TF or TF) was chosen to serve as a positive control ligand in all endocytosis assays. Transferrin is an iron carrier protein that has been thoroughly characterized to be internalized via the classic clathrin-dependent endocytic pathway, which is known to be regulated by dynamin 2 expression in A431 cells (Lamaze et al., 1993). Thus, transferrin uptake is expected to be significantly inhibited under DNM2 pool siRNA treatments. BeWo cells were transiently transfected with 40 nM targeting or nontargeting siRNA 63 to 72 h before performing uptake assays. Cells were then dosed with either 5 nM [³H]B₂ or 10 nM [¹²⁵I]TF for 4 min at 37°C. The effects of all treatments, including mock and untreated conditions, on ligand absorption were defined as the percentage of nontargeting siRNA effects. It is interesting that both actively internalized B₂ and TF were significantly reduced to a similar extent exclusively under silenced dynamin 2 states. Riboflavin internalization was reduced by 40% (59.45 ± 24.70% S.D. of N; Fig. 3A), and transferrin absorption was reduced by 30% (69.64 ± 4.27% S.D. of N; Fig. 3B). The attenuation in transferrin uptake seen exclusively with DNM2 pool siRNA-treated cells is in agreement with literature reports revealing this ligand's dependence on CME and dynamin 2 expression (Cao et al., 2003) and further validates the specificity of the RNAi methodology. A similar effect on B₂ absorption under these same conditions further corroborates prior evidence suggesting that this vitamin is internalized via CME (Huang and Swaan, 2000, 2001; Huang et al., 2003; Phelps et al., 2004; D'Souza et al., 2006a,b; Mason et al., 2006). Furthermore, the reduced absorption of B₂ (40%) directly correlated with 80% reduced protein expression for DNM2 (Fig. 1C). Thus, we can approximate that at least 50% of the active component regulating B₂ absorption in human placental trophoblasts requires dynamin 2-dependent RME events.

View larger version (53K): [in this window] [in a new window]   Fig. 3. To an extent similar to the control endocytic ligand transferrin, riboflavin uptake was reduced exclusively under silenced dynamin 2 conditions. BeWo cells transfected 63 to 72 h earlier with 40 nM siRNA were dosed with either 5 nM [3H]B2 or 10 nM [125I]TF for 4 min at 37 or 4°C. Both actively internalized B2 (A) and TF (B) are significantly reduced exclusively under DNM2 pool siRNA treatments. The results reflect the standard deviations of four to six separate experiments. Statistical significance between targeting siRNA and nontargeting treatments was defined using one-way ANOVA with Newman-Keuls multiple comparison test (*, p < 0.05).

B₂ Enrichment at the Plasma Membrane Doubled under Silenced Dynamin 2 States. Dynamin 2 GTPase has been extensively characterized to function as a critical gate keeper of intracellular trafficking in that it is required in endosomal vesicle formation and release from the plasma and Golgi membranes (McNiven et al., 2000). Without functional dynamin 2, RME mechanisms such as clathrin-dependent RME are prevented from forming endosomal vesicles and thus are unable to transport their cargo from the plasma membrane or Golgi domain to various other destinations in the cell. Instead, ligand-bound receptors are restricted to these membranes. In light of the evidence of attenuated uptake for both B₂ and TF shown exclusively under silenced dynamin 2 treatments, we would expect a concomitant increase in ligand concentrations bound at the cell surface. To test this hypothesis, we analyzed the extent of plasma membrane-bound ligand detected after the 4-min uptake period and compared these data with the extent of the internalized ligand.

Membrane-bound and internalized ligands detected under the varying 40 nM siRNA treatments were normalized to nontargeting siRNA effects and were expressed using eqs. 1 and 2 defined under Materials and Methods. As expected, DNM2 pool siRNA treatments led to a more pronounced increase in the extent of bound B₂ at the plasma membrane, which coincided with a concomitant decease in its internalization. Likewise, plasma membrane TF localization was enhanced under silenced dynamin 2 conditions. In particular, under silenced dynamin 2 states, plasma membrane-bound B₂ was shown to be 150% higher than bound ligand revealed for untreated cells (Fig. 4A). In addition, internalized B₂ reduced by 40% compared with untreated conditions. Although to a lesser extent than that shown for B₂, membrane-bound TF detected for cells treated with DNM2 pool siRNA increased 110% over bound ligand detected in untreated cell populations (Fig. 4B). Furthermore, TF uptake reduced 30% under attenuated dynamin 2 levels compared with untreated cells. Both B₂ and TF showed substantial localization at the plasma membrane with dynamin 2-silenced treatments compared with the corresponding internalized ligand data. Both ligands revealed nearly 200% higher enrichment at the plasma membrane than within the cell. In addition, when the extent of bound over the extent of internalized ligand was expressed as a ratio, both B₂ and TF were shown to be 3.1- and 2.0-fold higher than that defined for untreated conditions, respectively (Fig. 4, C and D).

View larger version (46K): [in this window] [in a new window]   Fig. 4. siRNA-induced silencing of dynamin 2 leads to 3-fold higher B2 enrichment to the plasma membrane. BeWo cells transfected with 40 nM duplex siRNA were used in uptake assays (63–72 h after transfection) to examine the extent of bound versus internalized ligand upon dosing cells for 4 min with either 5 nM [3H]B2 or 10 nM [125I]TF. Data analyses involved eqs. 1 and 2 defined under Materials and Methods for membrane-associated and internalized B2 (A) and TF (B). Ligand localization was further characterized for the overall extent of plasma membrane-bound over internalized B2 (C) and TF (D). These data represent the standard deviations of three to eight separate experiments. Statistical significance between targeting and nontargeting effects on ligand distribution was determined using one-way ANOVA with Newman-Keuls multiple comparison test (*, p < 0.05; ***, p < 0.001).

B₂ Was Substantially Enriched at the Membrane Surface in Cells Transiently Transfected with GTPase-Null Dynamin 2 (K44A). A common approach in corroborating RNAi data involves the use of wild-type and dominant-negative expression vectors. We obtained fully characterized DNM2^WT and DNM2^K44A expression constructs, the latter of which is unable to hydrolyze GTP and thus unable to pinch off nascently formed endocytic vesicles at the plasma membrane (Cao et al., 1998). BeWo cells were transiently transfected with 400 ng of plasmid DNA (DNM2^WT, DNM2^K44A, or the empty expression vector) in the presence or absence of 40 nM DNM2 pool siRNA. Cells were dosed with either 5 nM [³H]B₂ or 10 nM [¹²⁵I]TF for 4 min. When both B₂ and TF results were expressed as a ratio of plasma membrane-bound ligand over internalized ligand, a strikingly similar trend was revealed (Fig. 5). Both ligands were shown to be largely localized at the cell membrane, as opposed to the intracellular environment, under GTPase-null dynamin 2 alone and cotransfection conditions (i.e., DNM2^K44A or DNM2 siRNA + DNM2^K44A, respectively). Although not significant, plasma membrane bound B₂ under GTPase-null dynamin 2 conditions was shown to increase 1.5- and 1.7-fold over cells transfected with wild-type dynamin 2 or the empty vector alone, respectively (Fig. 5A). The enhanced enrichment of B₂ to the membrane surface was shown to be more pronounced under cotransfection conditions (i.e., DNM2 siRNA + DNM2^K44A), which led to 1.8-, 2-, and 1.4-fold higher membrane localization compared with the effects of wild type, the empty vector, and the cotransfection treatment involving nontargeting siRNA and the empty vector, respectively. A similar trend was noted for the control ligand, transferrin. Transient transfections involving GTPase-null dynamin 2 revealed nearly 2- and 2.4-fold higher TF localization at the plasma membrane versus the effects of wild-type dynamin 2 and empty vector treatments, respectively (Fig. 5B). Similar to B₂, the cotransfection condition led to a more pronounced enrichment of TF at the membrane surface that was 1.5- and nearly 3-fold higher than the effects of the negative control cotransfection treatment (i.e., N siRNA + empty vector) and the empty vector, respectively. These results are in agreement with the data from cells transfected with DNM2 pool siRNA alone and provide additional evidence that actively internalized B₂ requires, in part, the functional expression of the highly conserved mechanoenzyme, dynamin 2, for its endocytic translocation in human placental trophoblasts.

View larger version (50K): [in this window] [in a new window]   Fig. 5. Overexpression of GTPase-null dynamin 2 results in enhanced ligand localization at the plasma membrane. BeWo cells were transiently transfected with 400 ng each of either DNM2WT, DNM2K44A, or a combination of DNM2K44A and 40 nM DNM2 pool siRNA. Sixty-three to 72 h after transfection, cells were dosed with 5 nM [3H]B2 or 10 nM [125I]TF for 4 min at 37°C and examined for membrane-associated and internalized ligands. Ratios were defined for plasma membrane-bound over internalized B2 (A) and TF (B). The effects of dynamin 2 expression vectors or GTPase-null dynamin 2 cotransfections with DNM2 siRNA on ligand localization were defined through comparisons with single transfections with an empty vector or empty vector cotransfections with 40 nM N siRNA, respectively. These data represent the standard deviations of three to four separate experiments. Statistical significance between experimental treatments and control conditions were determined using one-way ANOVA with Newman-Keuls multiple comparison test (*, p < 0.05; **, p < 0.01).

Rhodamine-Labeled-B₂ Colocalized with the Caveolae Coat Protein Caveolin 1. CAV1 is characterized to function in caveolae-mediated endocytosis in mammalian cells (Pelkmans et al., 2004). Specifically, CAV1 is a cytoplasmically oriented integral membrane protein known to bind cholesterol (Uittenbogaard and Smart, 2000) and associates with cholesterol- and sphingolipid-rich plasma membrane domains termed caveolae. CAV1 is one of three known caveolin isoforms that has been characterized to be critical in signal transduction pathways, membrane organization, and ligand bound receptor-mediated trafficking specific to CvME in many divergent cell systems (Cheng et al., 2006). Some of the classic ligands that have been characterized to be absorbed and trafficked via this pathway include cholera toxin subunit B and folate (Shajahan et al., 2004; Birn, 2006). Furthermore, like CME, CvME is dependent on dynamin 2 for vesicle formation and release from the plasma membrane (Yao et al., 2005).

Internalized Rd-RF, Alexa Fluor 555-labeled CTX, or 5-carboxytetramethylrhodamine-labeled TF were examined for colocalization with immunofluorescence detected endosome markers, CAV1 or clathrin, in BeWo cells. All fluorescence colocalization assessments were further analyzed for intact cell morphology using differential interference contrast imaging (data not shown). After a 2-min uptake period, both Rd-RF and CTX resulted in punctate staining resembling endosomal organelle localization (Fig. 6A). Similar to CTX, Rd-RF resulted in substantial overlap with CAV1. However, signal overlap between Rd-RF and CAV1 resulted in a largely peripheral cytoplasmic staining pattern compared with the perinuclear staining noted for colocalized CTX and CAV1 channels. In addition, Rd-RF and the positive control ligand, transferrin, revealed punctate, perinuclear signal overlap with clathrin-positive endosomes after this same time period (Fig. 6B).

View larger version (87K): [in this window] [in a new window]   Fig. 6. Rd-RF colocalizes with the caveolae coat protein CAV1 and clathrin after 2-min uptake in BeWo cells. Rd-RF and positive control ligands, CTX (ligand of CvME) and TF (ligand of CME), were examined for colocalization with the immunostained endocytic protein markers CAV1 (A) or clathrin (B) after ligand internalization in BeWo cells for 2 min. Images represent 3D orthogonal profiles with the inset view defining the XY plane and the outer panels reveal the YZ (right narrow panel) and XZ (top narrow panel) focal planes. Fluorescence signals for each channel were merged to reveal regions of colocalization (indicated by arrows and yellow regions). Scale bars (10 µm) are shown in the lower left corner of the merged XY panel. All images were acquired using 120x magnification.

View larger version (40K): [in this window] [in a new window]   Fig. 7. Rd-RF colocalizes with caveolin 1-positive endosomes to a similar extent as with clathrin-positive vesicles. A, overlapping (i.e., colocalized) volumes (in cubic micrometers) for ligands and endosome marker channels were expressed as a percentage of the total volume for each ligand and represent the mean ± S.E.M. for three to four regions of interest. B, the PC was chosen to define the likeness in 3D shapes between overlapping ligand and vesicle marker channels. A positive value for PC reflects a positive correlation in fluorescence signal overlap. Data in this graph are expressed as the mean ± S.D. for three to four regions of interest.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
A dual approach involving RNAi and wild-type or dominant-negative plasmid transfections was carried out to delineate the involvement of the conserved endocytic mechanoenzyme dynamin 2 in B₂ absorption in the human placental trophoblast model BeWo. We showed that the endocytosis of B₂ is regulated, in part, by the functional expression of dynamin 2 GTPase. The extent of dynamin 2-dependence in regulating cellular B₂ levels was similar to that of the clathrin-dependent RME ligand, transferrin. Attenuated internalization of both B₂ and TF (60 and 70% of control conditions, respectively) correlated with 80% reduced dynamin 2 levels. Thus, at least 50% of actively absorbed B₂ is estimated to involve a dynamin 2-dependent RME process. In addition, reductions in ligand uptake under these same conditions revealed a concomitant increase in B₂ and TF binding along the cell surface. In further support of the siRNA treatments alone, cotransfection results involving the GTPase-null dynamin 2 expression construct and DNM2 pool siRNA generated an amplified effect on the predominant ligand accumulation at the plasma membrane for both ligands. Although a B₂-specific receptor and the proposed B₂-soluble carrier protein remain to be identified, these data provide the most substantial evidence to date supporting the existence of such molecular machinery in humans.

Transferrin uptake in BeWo cells under silenced dynamin 2 protein expression is less pronounced compared with other reports (Hinrichsen et al., 2003; Huang et al., 2004); however, this observation may be cell type- and transfection reagent-specific. For instance, Huang and coworkers (2004) revealed 70 to 80% inhibition in transferrin absorption in HeLa cells that were transfected twice with siRNA targeting the clathrin heavy chain (i.e., a major protein component required in clathrin polyhedral lattice formation) or dynamin 2. However, two consecutive transfections using the DNM2 pool siRNA treatments in BeWo cells drastically attenuated cell growth and compromised cell morphology when examined under light microscopy (data not shown). In another study by Soulet and colleagues (2005), a single siRNA transfection in HeLa cells using siRNA targeting sorting nexin 9, a protein that binds and regulates dynamin 2 activity and clathrin-mediated endocytic efficiency, resulted in 45% reduction in transferrin absorption. These data are comparable with the effects we observed for this control ligand in BeWo cells treated with DNM2 pool siRNA (i.e., 30% reduced transferrin uptake). This close agreement between our results in BeWo and that of Soulet and coworkers' studies in HeLa suggests that our RNAi strategies are specific and effective in eliciting functional effects on the trafficking of the control ligand, transferrin.

Our laboratory recently showed a clathrin-mediated endocytic component regulating riboflavin absorption and trafficking in divergent cell systems (Huang and Swaan, 2000, 2001; Huang et al., 2003; D'Souza et al., 2006b). Based on our results that showed the requirement for dynamin 2 expression in directing B₂ cellular entry and considering the pluripotent nature of this scission enzyme on regulating different RME mechanisms, the question remained as to whether multiple and distinct endocytic pathways were regulating intracellular B₂ trafficking. The caveolae-mediated endocytic mechanism is a candidate pathway that, like CME, requires dynamin 2 activity to allow for endocytic vesicle release of ligand-receptor cargo from the plasma membrane. Established fluorescent ligand endocytosis assays involving the characterized rhodamine-B₂ conjugate were carried out to define the extent of colocalized signal intensities with the immunofluorescence detected caveolar endosome marker protein, caveolin 1, and were compared with colocalization assessments between the B₂ conjugate and clathrin-positive vesicles. Three-dimensional confocal laser scanning microscopic analyses revealed similar extents in B₂ localization to both clathrin- and caveolin 1-positive vesicles after 2- and 10-min uptake periods. In addition to this vitamin's dependence on clathrin-mediated endocytosis, the intracellular distribution of absorbed B₂ in human epithelia is implicated to involve for the first time the caveolar-mediated pathway.

The concept of multiple endocytic processes regulating the cellular trafficking of a single ligand is not unique to B₂. Evidence of this phenomenon has been demonstrated for folate, TGF-, and cholera toxin subunit B (Di Guglielmo et al., 2003; Shajahan et al., 2004; Birn, 2006). Considering the importance of maintaining cellular B₂ levels required for normal growth and development, the existence of multiple cellular transport mechanisms would provide additional controls to meet such nutritional demands in states of physiological distress. In the case for TGF-, Di Guglielmo and colleagues (2003) demonstrated a biochemical feedback mechanism regulating its internalization via either the clathrin-mediated or caveolar-mediated pathways. Specifically, their data revealed the absorption of TGF- along the CME pathway correlated with a signal transduction response as defined by interactions with the Smad anchor for receptor activation protein. In contrast, caveolae-mediated endocytosis of TGF- was coupled with the Smad7-Smurf2-dependent receptor degradation response and led to ubiquitin-dependent degradation of the TGF- receptor. In this particular instance, the CvME mechanism seems to be involved in receptor degradation and ultimately receptor turnover, whereas the CME pathway functions in promoting signal transduction cascades. However, studies with epidermal growth factor (de Melker et al., 2001) have shown the CME pathway to facilitate receptor degradation through trafficking to lysosomal organelles. In the case for B₂, the involvement of two distinct RME pathways regulating its intracellular distribution may reflect a homeostatic mechanism of molecular sensors that either promote B₂-receptor activation coupled with increased vitamin endocytosis or initiate receptor degradation and reduced ligand uptake.

In summary, the RNAi and dynamin 2 plasmid DNA transfection data provide definitive and direct evidence that a B₂ receptor-mediated endocytic mechanism exists in human placental trophoblasts. To date, dynamin 2 GTPase is the first protein identified in humans to serve as a regulator of B₂ cellular entry through RME. In addition to the clathrin-dependent B₂-RME process, we report for the first time that B₂ trafficking involves the CvME pathway in the BeWo model. Understanding the cellular absorption and trafficking itineraries specific to B₂ will aid in future studies aimed at understanding various pathological states correlated with B₂ deficiency and opens up novel drug targeting strategies that can be designed to bypass efflux transporters to potentially improve drug bioavailability. Conceptually and empirically, such drug formulations targeting either folate or transferrin RME pathways have been successfully demonstrated (Stephenson et al., 2003; Singh et al., 2006) and further promote the feasibility of exploiting the B₂-RME portal as an alternative drug targeting route. Furthermore, recent reports (Rao et al., 1999, 2006) revealing substantially elevated serum riboflavin carrier protein levels in patients with breast and liver cancer compared with healthy subjects propose a potential chemotherapeutic niche for such B₂-drug targeting initiatives.

	Acknowledgements

 
We thank the laboratory of Dr. Mark A McNiven (Mayo Clinic and Foundation, Rochester, MN) for graciously donating validated dynamin expression vectors. We also thank Dr. James I. Lee and Dr. Vanessa M. D'Souza for their valuable suggestions throughout this study.

	Footnotes

 
This work was supported, in part, by the National Institutes of Health under grant DK56631 (to P.W.S.), a postdoctoral fellowship grant from the Susan G. Komen Foundation (to P.W.S.), and by a predoctoral fellowship sponsored by Eli Lilly and Company (to A.B.F.).

Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org.

doi:10.1124/mol.107.037101.

ABBREVIATIONS: RME, receptor-mediated endocytosis; B₂, riboflavin; CME, clathrin-mediated endocytosis; RNAi, RNA interference; TGF-, tumor growth factor ; DNM2, dynamin 2 GTPase; CvME, caveolae-mediated endocytosis; siRNA, short-interfering RNA; DLP1, dynamin-like protein 1; TF, transferrin; CAV1, caveolin 1; CTX, cholera toxin subunit B; PC, Pearson's correlation; N, nontargeting; ANOVA, analysis of variance; PBS, phosphate-buffered saline; 3D, three-dimensional;. Rd-RF, rhodamine-riboflavin; DNM2^WT, wild-type dynamin 2; DNM2^K44A, dominant-negative dynamin 2.

Address correspondence to: Dr. Peter W. Swaan, Department of Pharmaceutical Sciences, University of Maryland, 20 Penn Street, HSF2–621, Baltimore, MD 21201. E-mail: pswaan{at}rx.umaryland.edu

	References

Top Abstract Materials and Methods Results Discussion References

 
Abrams VA, McGahan TJ, Rohrer JS, Bero AS and White HB 3rd (1988) Riboflavin-binding protein from reptiles: a comparison with avian riboflavin-binding proteins. Comp Biochem Physiol B 90: 243–247.[CrossRef][Medline]

Barouch W, Prasad K, Greene L, and Eisenberg E (1997) Auxilin-induced interaction of the molecular chaperone Hsc70 with clathrin baskets. Biochemistry 36: 4303–4308.[CrossRef][Medline]

Birn H (2006) The kidney in vitamin B12 and folate homeostasis: characterization of receptors for tubular uptake of vitamins and carrier proteins. Am J Physiol Renal Physiol 291: F22–F36.[Abstract/Free Full Text]

Cao H, Garcia F, and McNiven MA (1998) Differential distribution of dynamin isoforms in mammalian cells. Mol Biol Cell 9: 2595–2609.[Abstract/Free Full Text]

Cao H, Orth JD, Chen J, Weller SG, Heuser JE, and McNiven MA (2003) Cortactin is a component of clathrin-coated pits and participates in receptor-mediated endocytosis. Mol Cell Biol 23: 2162–2170.[Abstract/Free Full Text]

Cheng ZJ, Singh RD, Marks DL, and Pagano RE (2006) Membrane microdomains, caveolae, and caveolar endocytosis of sphingolipids. Mol Membr Biol 23: 101–110.[CrossRef][Medline]

Conner SD and Schmid SL (2003) Regulated portals of entry into the cell. Nature 422: 37–44.[CrossRef][Medline]

D'Souza VM, Bareford LM, Ray A, and Swaan PW (2006a) Cytoskeletal scaffolds regulate riboflavin endocytosis and recycling in placental trophoblasts. J Nutr Biochem 17: 821–829.[CrossRef][Medline]

D'Souza VM, Foraker AB, Free RB, Ray A, Shapiro PS, and Swaan PW (2006b) cAMP-coupled riboflavin trafficking in placental trophoblasts: a dynamic and ordered process. Biochemistry 45: 6095–6104.[CrossRef][Medline]

Daniels TR, Delgado T, Helguera G, and Penichet ML (2006) The transferrin receptor part II: targeted delivery of therapeutic agents into cancer cells. Clin Immunol 121: 159–176.[CrossRef][Medline]

Dautry-Varsat A, Ciechanover A, and Lodish HF (1983) pH and the recycling of transferrin during receptor-mediated endocytosis. Proc Natl Acad Sci U S A 80: 2258–22562.[Abstract/Free Full Text]

de Melker AA, van der Horst G, Calafat J, Jansen H, and Borst J (2001) c-Cbl ubiquitinates the EGF receptor at the plasma membrane and remains receptor associated throughout the endocytic route. J Cell Sci 114: 2167–2178.[Abstract/Free Full Text]

Di Guglielmo GM, Le Roy C, Goodfellow AF, and Wrana JL (2003) Distinct endocytic pathways regulate TGF-beta receptor signalling and turnover. Nat Cell Biol 5: 410–421.[CrossRef][Medline]

Fielding PE and Fielding CJ (1996) Intracellular transport of low density lipoprotein derived free cholesterol begins at clathrin-coated pits and terminates at cell surface caveolae. Biochemistry 35: 14932–14938.[CrossRef][Medline]

Foraker AB, Khantwal CM, and Swaan PW (2003) Current perspectives on the cellular uptake and trafficking of riboflavin. Adv Drug Deliv Rev 55: 1467–1483.[CrossRef][Medline]

Hill E, van der Kaay J, Downes CP, and Smythe E (2001) The role of dynamin and its binding partners in coated pit invagination. J Cell Biol 152: 309–323.[Abstract/Free Full Text]

Hinrichsen L, Harborth J, Andrees L, Weber K, and Ungewickell EJ (2003) Effect of clathrin heavy chain- and -adaptin-specific small inhibitory RNAs on endocytic accessory proteins and receptor trafficking in HeLa cells. J Biol Chem 278: 45160–45170.[Abstract/Free Full Text]

Huang F, Khvorova A, Marshall W, and Sorkin A (2004) Analysis of clathrin-mediated endocytosis of epidermal growth factor receptor by RNA interference. J Biol Chem 279: 16657–16661.[Abstract/Free Full Text]

Huang SN, Phelps MA, and Swaan PW (2003) Involvement of endocytic organelles in the subcellular trafficking and localization of riboflavin. J Pharmacol Exp Ther 306: 681–687.[Abstract/Free Full Text]

Huang SN and Swaan PW (2000) Involvement of a receptor-mediated component in cellular translocation of riboflavin. J Pharmacol Exp Ther 294: 117–125.[Abstract/Free Full Text]

Huang SN and Swaan PW (2001) Riboflavin uptake in human trophoblast-derived BeWo cell monolayers: cellular translocation and regulatory mechanisms. J Pharmacol Exp Ther 298: 264–271.[Abstract/Free Full Text]

Lamaze C, Baba T, Redelmeier TE, and Schmid SL (1993) Recruitment of epidermal growth factor and transferrin receptors into coated pits in vitro: differing biochemical requirements. Mol Biol Cell 4: 715–727.[Abstract]

Manders MM, Verbeek PJ, and Aten JA (1993) Measurement of co-localization of objects in dual colour confocal images. J Microsc 169: 375–382.

Mason CW, D'Souza VM, Bareford LM, Phelps MA, Ray A, and Swaan PW (2006) Recognition, co-internalization, and recycling of an avian riboflavin carrier protein in human placental trophoblasts. J Pharmacol Exp Ther 317: 465–472.[Abstract/Free Full Text]

McNiven MA, Cao H, Pitts KR, and Yoon Y (2000) The dynamin family of mechanoenzymes: pinching in new places. Trends Biochem Sci 25: 115–120.[CrossRef][Medline]

Pelkmans L, Burli T, Zerial M, and Helenius A (2004) Caveolin-stabilized membrane domains as multifunctional transport and sorting devices in endocytic membrane traffic. Cell 118: 767–780.[CrossRef][Medline]

Phelps MA, Foraker AB, Gao W, Dalton JT, and Swaan PW (2004) A novel rhodamine-riboflavin conjugate probe exhibits distinct fluorescence resonance energy transfer that enables riboflavin trafficking and subcellular localization studies. Mol Pharm 1: 257–266.[CrossRef][Medline]

Powers HJ (2003) Riboflavin (vitamin B-2) and health. Am J Clin Nutr 77:1352–1360. Powers HJ (2005) Interaction among folate, riboflavin, genotype, and cancer, with reference to colorectal and cervical cancer. J Nutr 135: 2960S–2966S.

Praefcke GJ and McMahon HT (2004) The dynamin superfamily: universal membrane tubulation and fission molecules? Nat Rev Mol Cell Biol 5: 133–147.[CrossRef][Medline]

Querbes W, O'Hara BA, Williams G, and Atwood WJ (2006) Invasion of host cells by JC virus identifies a novel role for caveolae in endosomal sorting of noncaveolar ligands. J Virol 80: 9402–9413.[Abstract/Free Full Text]

Rao PN, Crippin J, Levine E, Hunt J, Baliga S, Balart L, Anthony L, Mulekar M, and Raj MH (2006) Elevation of serum riboflavin carrier protein in hepatocellular carcinoma. Hepatol Res 35: 83–87.[CrossRef][Medline]

Rao PN, Levine E, Myers MO, Prakash V, Watson J, Stolier A, Kopicko JJ, Kissinger P, Raj SG, and Raj MH (1999) Elevation of serum riboflavin carrier protein in breast cancer. Cancer Epidemiol Biomarkers Prev 8: 985–990.[Abstract/Free Full Text]

Said HM and Ma TY (1994) Mechanism of riboflavine uptake by Caco-2 human intestinal epithelial cells. Am J Physiol 266: G15–G21.[Medline]

Schmid SL (2004) Endocytosis, in Cell Biology (Pollard TD and Earnshaw WC eds) pp 355–367, W.B. Saunders, Philadelphia.

Shajahan AN, Timblin BK, Sandoval R, Tiruppathi C, Malik AB, and Minshall RD (2004) Role of Src-induced dynamin-2 phosphorylation in caveolae-mediated endocytosis in endothelial cells. J Biol Chem 279: 20392–20400.[Abstract/Free Full Text]

Sheff D (2004) Endosomes as a route for drug delivery in the real world. Adv Drug Deliv Rev 56: 927–930.[CrossRef][Medline]

Singh M, Hawtrey A, and Ariatti M (2006) Lipoplexes with biotinylated transferrin accessories: novel, targeted, serum-tolerant gene carriers. Int J Pharm 321: 124–137.[CrossRef][Medline]

Soulet F, Yarar D, Leonard M, and Schmid SL (2005) SNX9 regulates dynamin assembly and is required for efficient clathrin-mediated endocytosis. Mol Biol Cell 16: 2058–2067.[Abstract/Free Full Text]

Stephenson SM, Yang W, Stevens PJ, Tjarks W, Barth RF, and Lee RJ (2003) Folate receptor-targeted liposomes as possible delivery vehicles for boron neutron capture therapy. Anticancer Res 23: 3341–3345.[Medline]

Uittenbogaard A and Smart EJ (2000) Palmitoylation of caveolin-1 is required for cholesterol binding, chaperone complex formation, and rapid transport of cholesterol to caveolae. J Biol Chem 275: 25595–25599.[Abstract/Free Full Text]

van der Ende A, du Maine A, Simmons CF, Schwartz AL, and Strous GJ (1987) Iron metabolism in BeWo chorion carcinoma cells. Transferrin-mediated uptake and release of iron. J Biol Chem 262: 8910–8916.[Abstract/Free Full Text]

Yao Q, Chen J, Cao H, Orth JD, McCaffery JM, Stan RV, and McNiven MA (2005) Caveolin-1 interacts directly with dynamin-2. J Mol Biol 348: 491–501.[CrossRef][Medline]

Zempleni J, Link G, and Bitsch I (1995) Intrauterine vitamin B2 uptake of preterm and full-term infants. Pediatr Res 38: 585–591.[Medline]

Zheng DB, Lim HM, Pene JJ, and White HB 3rd (1988) Chicken riboflavin-binding protein. cDNA sequence and homology with milk folate-binding protein. J Biol Chem 263: 11126–11129.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) All Versions of this Article: mol.107.037101v1 72/3/553    most recent Submit a response Alert me when this article is cited